Level Design

Level Design

Concept, Theory, and Practice

Rudolf Kremers

A K Peters, Ltd.
Natick, Massachusetts
Cover illustrations courtesy of Alex May

CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742

© 2009 by Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government works

Version Date: 20150515

International Standard Book Number-13: 978-1-4398-7695-4 (eBook - PDF)

This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to publish reliable data and information, but the author and publisher
cannot assume responsibility for the validity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material repro-
duced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged please write and
let us know so we may rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now
known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc.
(CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have
been granted a photocopy license by the CCC, a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com

and the CRC Press Web site at

http://www.crcpress.com
For Kate
Contents

Preface ix
Introduction xiii

I. Introduction to Level Design 1

1. Game Design vs. Level Design 3
2. Teaching Mechanisms 19
3. Level Design Goals and Hierarchies 41
4. Level Design Structure and Methodology 55
5. Single-Player vs. Multiplayer Considerations 81
6. Towards a Shared Grammer for Level
Design 87

II. Emotional Feedback Systems 93

7. Reward Systems 95
8. Immersion 137
9. Negative Emotions 167

III. Game Environments 189

10. Visual Experiences 191
11. Audio Design 219

IV. Game Stories 241

12. Story and Narrative 243

vii
viii Contents

V. Designing Gameplay 273

13. World Building 275
14. Puzzles 307
15. Artificial Intelligence 323
16. Challenge 351

VI. Final Thoughts 367

17. Considering the Future 369
Recommended Reading & Materials 373
Glossary 377
Preface

I hope that in finishing this tutorial you will be well on your way to
creating worlds of your wildest dreams . . . or nightmares.1
As in many endeavors of this kind, much of this book comes from personal moti-
vation and circumstances. I have tried to keep reference to those out of the main
text but I would like to acknowledge them somewhere in the book. I guess that
is why the preface was invented. Here, then, are some of the personal aspects of
why and how this book came to be.

On Level Design
One of the many beautiful things about level design is that it is an almost all-
encompassing field. There aren’t many creative interests and expressions that
cannot find a home in level design, yet it remains a unique profession with its
own rules and quirks, and there are few jobs in games as satisfying or important
as that of a level designer.
It is, however, a very young profession, and because of its youth, it is often
misunderstood or misrepresented, which is regrettable. A good level designer
can have as much impact as a good screenwriter or director, a great artist or mu-
sician. Those are vocations that people aspire to and recognize as being benefi-
cial to society as a whole.
Video games have not yet come this far. Many people still look surprised to
even hear video games mentioned in conjunction with other, more established
art forms, although this is slowly changing. Perhaps one day this mistrust of vid-
eo games as a medium that can compare to other art forms will be quaint and
slightly amusing. Hopefully on that day this book will still be in print and the
words you have just read will seem odd and out of place.
1
Worldcraft 1.6 Tutorial, available at http://hosted.planetquake.gamespy.com/world-
craft/tutorial/index.shtm, 1997.
ix
x Preface

Either way, regardless of being greeted with a cheer or a puzzled face, being
able to work or have a career as a level designer should be something to be im-
mensely proud of. I certainly am.

Why This Book?

Working as a game designer or level designer for a substantial amount of time
has shown me there is a lack of understanding of how level design works at a
fundamental level. There are many very basic questions that go unasked:
• What fundamental concepts matter to this field?
• How do they translate into level design theory?
• How do we translate theory into practice?
Level design is an incredibly exciting and important field. Good or bad level de-
sign can make or break any game, so it is surprising how little reference material
there is available for level designers to give them a deeper understanding of what
exactly their field is about. Level designers have a limited understanding of what
tools and techniques they can use to achieve their goals, or even define them.
This is perhaps, considering the youth of the profession, (15–20 years or so2)
not entirely surprising.
This is, however, a completely different situation from comparable profes-
sions in other industries; there are countless sage-like tomes available on top-
ics as diverse as direction, production, set design, camera work, acting, lighting,
writing and so on. In those disciplines, in stark contrast to level design, an enor-
mous amount of knowledge has been collected and made available to others.
Even within the game industry, there is a large amount of high-quality literature
and other material available to programmers and artists, but not so much for
designers.
Therefore, most people who work in the field of level design either learn
through apprenticeship on the job, or are self-taught. This is less than ideal be-
cause, due to the high-pressure work environment of the game industry, it is very
difficult for people to take a step back and incorporate a more formalized ap-
proach to their profession. Nor, as mentioned already, is relevant knowledge and
reference readily available. Instead, a lot of level design isn’t actually designed.
Often it is a derivative of other successful work, implemented without a full un-
derstanding of why the original work was successful. Other times it is just reac-
tionary, quickly improvised to vague parameters, sometimes based on nothing
more than some nebulous sense of “fun.” This is not meant as a harsh criticism
2
At the time of writing.
Preface xi

of other people; I certainly have done it many times myself. It is just a logical
consequence of working in a creative field where there is not enough support at a
core level for people who already have limited time and are under much pressure
to perform. This is one of many reasons why I believe there is a great need for a
book that provides strong conceptual foundations, formulates working theories
and, crucially, shows how to apply these in practice.
After learning about my craft the hard way, like other level designers, through
self-training on the job, I realized that most of the hard-earned lessons in our field
stay isolated from other level designers. The people most likely to hold the right
answers to difficult level design questions posed in practice are often too busy
working on their next game, while people outside of games don’t have enough
exposure to the hard reality of commercial or practical game development. This
is why I have set out to write a book that bridges this gap between theory and
practice, useful to people on either side of the spectrum—a book that will still
be relevant as a work of reference or as a practical guide many years after be-
ing published. I wanted to write a book that uses a conceptual and theoretical
foundation to build a set of practical tools and techniques that can be universally
applied within the field of level design. Or, to put it more simply, I have tried to
write the book that wasn’t available to me when I first started as a level designer.
Hopefully it will fill that role for some of the readers.

Thanks and Dedications

While writing this book, I have gone through a number of events that have af-
fected me in one way or another. I bought a house with my partner, sat through
two earthquakes (yes, two), added two rescue dogs to our menagerie, started my
own business making independent games and became a full-fledged vegetarian.
All of the above provided major distractions, as you can imagine; yet, in all
that time, I was lucky enough to receive the love and unwavering support of my
wonderful partner, Kate, without whom I would have been lost several times
over. This book is therefore firmly dedicated to her.
Additional thanks go to my publisher, A K Peters, and the people there that
have helped me get the book into a readable state: Alice and Charlotte, espe-
cially. And boy did I need the help!
I have also been lucky to receive the support of family and friends and all
the people who have helped me with proofreading and advice all throughout the
project. Thank you all very much! Special thanks go to Alex May for his work on
Dyson. Extra special thanks go to Colin and Diana for their unfaltering support
and belief in me.
xii Preface

I must also mention Neemo the courageous dog, Tom the wonder dog, and
Billy the naughtypuss.
Finally, thanks to all the people who create the levels that inspire me every
day.
Introduction

M ake things as simple as possible, but not simpler.

—Albert Einstein
This book is a hefty affair; there are many ideas, concepts theories, practical ex-
amples, tangents, and footnotes to read. I have done my best to present it in a
way that makes sense to most readers. But, to really get the most of the book, I
give you this introduction that explains how it all hangs together.

How to Use This Book

Ultimately it is up to you, the reader, to use this book in any way you see fit.
However, the book has been written within a certain logical structure and flow
that made sense to me while writing it. Wherever possible I have tried to intro-
duce concepts and theories before I start showing their practical applications.
This not only applies to the later chapters that are actually divided into concept,
theory, and practice sections but also applies to much of the preceding content.
A good example of this is the chapter on “Teaching Mechanisms” (Chapter 2),
which explores ideas that are so fundamental to what level design is about that
it made sense to cover it early on. I do advise people to read this chapter before
the following ones.
This kind of hierarchy is applied throughout the bulk of the book—where
understanding of certain ideas paves the way to understanding of others. Since
later sections often refer to earlier ones, the book may be harder to use if read out
of sequence, though I have tried to facilitate this by referring to related material
in later chapter when I thought it appropriate. Nonetheless, there is plenty of
material in the book that can be read in its own right. Indeed, it is almost impos-
sible within the framework of such a wide field to adhere to strict linearity.
It is my hope that I have made a book that is useful to people with a general
interest in the subject of games as well as to people with a more specific interest
in level design.
xiii
xiv Introduction

Who Is This Book For?

Although this book can be read (and hopefully understood) by anybody who
picks up a copy out of curiosity, I would like to highlight some potential readers
who may be more interested than others.

Level Designers
Naturally, it is important that this book is useful to level designers. The book is
written in such a way that it should prove useful to both inexperienced and vet-
eran designers alike. I made a conscious choice to mainly focus on level design
as opposed to level construction.1 It is important to make that distinction, as level
design is a universal field, whereas level construction is technology dependent.
Too often have I come across texts that claim to be about level design but read
more like a technical instruction manual for some 3D software application. This
is only a useful approach for those people who will be using the same technology
as the author of that text.
Instead, I have done my best to make the book as much about level design
as possible, ignoring platform- or technology- or level-construction-specific is-
sues. The book is based on the assumption that design knowledge has to come
first. The way in which vertices are manipulated in a 3D mesh or how entities are
placed in a level editor is so dependent on unpredictable technical factors that
they are best tackled outside of the context of this book.
What this book will do, however, is cover material that teaches and informs
level designers of all levels of expertise what to aim for in their designs, why they
should do so, and how to achieve those goals in as many diverse situations as
possible.

Students and Teachers

Similarly, the book is aimed at future level designers and their teachers. There
should be much useful content for students and teachers to work with. Due to
the fact that there are no insurmountable technical barriers barring readers from
engaging with the text, it is ideal for general academic use.
There is a large amount of theory and high-level conceptual content to be
found in this book. Much of it is supported by information derived from other
fields of interest.

1
Or production.
Introduction xv

Nonetheless, this still only scratches the surface of the vast amount of
knowledge that is relevant to level design. The book, therefore, could be used as
a stand-alone text as well as a base for even more detailed course material.
The practical2 aspect of the book provides an invaluable tool for assessing
the progress made within a course. Since the material covered always has a final
practical application, it is always possible to test the knowledge acquired against
the harsh realities of real-world level design scenarios. This provides scope for
benchmarking as well as the added advantage of providing real training and
preparation for actual industry work.

Other Interested Parties

There are, of course, many other reasons why somebody could be interested in
the field of level design. Just as I have throughout the book found inspiration
and knowledge applicable to level design through the study of other creative dis-
ciplines, the same can be imagined of people in other disciplines in regards to
level design. A certain amount of cross-pollination between art forms is to be
expected and encouraged.
Furthermore, the book can be enjoyed on a non-professional level. Anybody
who has an interest in level design or video games should find more than enough
to keep them occupied for quite a while. Due to the fact that level design is such
a wide-ranging discipline, it can easily spark off or feed other existing interests
as well. Indeed, many of the chapters can function as short introductions and
springboards to entire fields of related knowledge. This is one of the beautiful
things about level design in general, and hopefully this multi-connected aspect
will speak to people reading about it. It still does to me.

Organization
Whichever way you choose to read the book, it may be useful to understand
the way the book has been structured. For this purpose a short summary of the
chapters and parts of the book follows.
The main body of the book is organized into parts, then chapters, and then
sections. Parts are thematically related: they handle areas of interest that can be
seen through the filter of a general theme. The chapters handle more distinct ar-
eas of interest and cover specific subjects. The sections break the subjects down
into three important aspects: concept, theory, and practice.

2
The sections dealing with level design practice especially.
xvi Introduction

Part I: Introduction to Level Design

Before we can even begin to look at either high-level concepts or more practical
aspects, we have to understand at a basic level exactly what it is that we are do-
ing.
These chapters are about level design itself. It covers the nature of the beast
all the way from examining its function within the field of game design to ex-
plaining what level design is—and, more importantly, what good level design
is.
Furthermore, the text goes into a fair amount of detail regarding methodol-
ogy and structure. This section of the book is fundamental to understanding the
point of level design and the ability to define sensible goals and tasks springing
from that knowledge. The section provides, among other things, much interest-
ing information about the psychology of play, without which it is difficult to
really grasp the impact of the techniques at our disposal. It also gives a first ex-
ample of the concept-theory-practice breakdown used more extensively in later
parts of the book.

Part II: Emotional Feedback Systems

This part of the book examines how game levels work as feedback systems.
That is to say that game levels can provide players with emotional feedback—
fun, tension, immersion and so on—based on their actions within the game
world.

Part III: Game Environments

Part III looks at some of the senses through which we experience levels as game
environments. Specifically, the visual and audio aspects of level design are stud-
ied in a fair amount of detail.

Part IV: Game Stories

Narrative and story are very controversial subjects within the framework of
game studies. Yet, there is no escaping the fact that level designers have to deal
with the subject on a regular basis and will be better able to do their work if they
have an informed opinion or understanding of what factors are at play. I decided
that due to the somewhat specific focus of the subject, it should be covered in a
separate part of the book.
Introduction xvii

Part V: Designing Gameplay

This is a central part of the text, where many of the diverse parts about actually
designing levels come together. Or, to paraphrase one famous level designer, Jay
Wilbur; this is “where the rubber hits the road.”
Many of the bread and butter issues of level design like puzzles, challenge
levels, or item placement are examined in detail.

Part VI: Final Thoughts

In this final part of the book, matters are wrapped up by a taking a look at what may
lie in the future of level design. There is also a small list of recommended material,
reading or otherwise, that may be of interest to those readers who want to study
more on the subject in the future, as well as a glossary and the book’s index.

Notes on Concept,Theory, and Practice Sections

A large portion of the text is presented in chapters that each, in a fairly linear
manner, cover a topic by addressing a basic concept, incorporating it in level de-
sign theory, and finally apply it in practice through examples and case studies.
This structure—where concept leads to theory leads to practice—is the only way I
could conceive to collate this much diverse information in one coherent text.

Concept
In these sections I will identify a large number of concepts, by which I mean any
topics and areas of interest related to fundamental elements of level design. These
are topics that anybody involved in level design will encounter at one point or
another and are worth exploring. The concept section of a chapter presents the
reader with a short essay or exploration that examines a specific topic that in
some way matters to level design. These essays will mainly deal with high-level
principles and ideas that take inspiration from sources as diverse as general psy-
chology, language studies, other art forms like film or literature, or anything else
that is relevant to the topic.
Through examination and analysis that goes beyond the limited framework
of video games, the concept section will lead to a number of general findings.

Theory
The theory sections will take the general findings from the previous section and
show what relevance they have when they are applied directly to the field of level
xviii Introduction

design. It will show that, through this process and through general extrapolation,
it is possible to arrive at a number of logical conclusions that can be useful and
relevant to specific level design theory.
These sections will demonstrate that these theories can be applied to all
kinds of different level design situations, largely irrespective of genre. In other
words, they are fundamental to level design.

Practice
Finally, in the practice sections of the book, the level design theory of the previous
section will get exposed to real-world practical examples. At this stage the subject
matter becomes one of practical application and will be much more specific than
before.
These sections will document numerous examples of typical level design
situations or scenarios and show how they benefit from the conclusions of the
previous sections. By using a wide and diverse range of examples, they will show
how useful it is to be able to apply universal techniques and theories to real-
world level design situations.
I will give practical examples on many level design situations, some of
which will be drawn from existing levels, sometimes my own work, or will be
completely new scenarios. We will look at examples from as many diverse genres
as possible, including puzzle games, 3D action adventures, first-person shooters,
2D platform games, and many more.
The key point is that by using the lessons learned in the concept and theory
parts of the books, it is possible to apply solutions to level design questions in
almost any game genre or situation.

Goals
Finally, I would like to explain something about my general philosophy on books
like this one. I do not expect the reader to agree with all my assertions and opinion
in this book. I think it is futile to even try to do this, especially in a field like level
design where there is still so little consensus in some of the key areas of interest.
What I am trying to offer instead is a book that gives the reader a number
of useful tools to work with. This includes tools for examining the subject itself,
tools to form or study theories, and tools for applying knowledge in practical
and tangible situations. I firmly believe that an informative approach, rather than
a dogmatic one, works best.
Within that spirit I hope that the reader will find it a useful and empowering
work, and that it will open doors or avenues that may have been left unexplored
otherwise.Giamcon henim do ex ecte facin volum vulputatum iriurer aesendi
Introduction to Level I
Design

“Gleicharmige Waage”, Wikipedia, http://commons.wikimedia.org/wiki/File:Gleichar

1

mige_Waage.png, 2009.
1
Game Design vs. Level
Design 1

I t is often asserted that level design is a subset of game design. This is mirrored
in the game industry, where, at the time of writing this book, level design
does not have the same status as game design, and often, level designers have
to aspire to game design roles to progress in their career. The pay grades within
the industry enforce this bias, as well; a game designer will typically earn more
money than a level designer. This situation is very unfortunate and based on a
number of misunderstandin gs or misconceptions. Among these are that game
design “trumps” level design, or that level design work is somehow subordinate
to, or simpler than, game design. This assertion misses an important aspect of
level design: it should not be underestimated how much of an impact level de-
sign has on a game. It is a well-known truism that:
Bad level design can ruin a good game.
If this is true (it is), the impact of a game’s level design is as strong as that of the
game design. Conversely, we can therefore say:
A bad game cannot be saved by good level design.1
The two statements seem to put level design and game design on somewhat
equal footing, at least in terms of impact on the development of the game itself.
It is clear that game design and level design are not the same discipline. How-
ever, they clearly cannot exist without each other. There is an interrelationship
at work that operates on a deep and fundamental level. It is useful to try to find a
way to interpret both fields and show how they are interconnected and what sort
of relationship they form. To make this possible, we must have a look at the basic
function that both fields perform.
1
Although it can be argued that good level design can make a mediocre game enjoy-
able.
3
4 Introduction to Level Design

The Function of Game Design

There are many differing interpretations of what game design actually is; some
are derived from detailed academic studies, while others come from experienced
game designers who have gained their knowledge through practical application.
The sheer amount of differing definitions almost inevitably leads to as many dis-
agreements and arguments, although many disagreements seem to arise as much
out of semantic differences as out of interpretive ones. The various game design
definitions can arise from origins including ideology, commercial function, or
practical analysis.
Because of this lack of consensus, and because of the sheer number of pro-
posed definitions, there is currently no easily identifiable unifying concept of
what constitutes gameplay, let alone game design. There are many sizeable books
and academic studies devoted to games and gameplay but they often contradict
each other and in some cases even attack each other. This presents a problem,
since game design and level design are interrelated, and we need to agree upon
a certain amount of theoretical common ground to formulate workable univer-
sal (or at least wide ranging) theories and techniques. So despite the lack of a
universally accepted view on games and gameplay, we should nonetheless exam-
ine the diversity of opinion more closely. Even if no definitive consensus can be
agreed upon by everybody in the field, we should still aim to find commonalities
and individual defining features that can be used for the purpose of this book.
Perhaps by looking at all the differing ideas on what constitutes a game or game-
play, we can derive a useful set of descriptions that can be used to characterize
the function of game design.

Definitions
One of the most famous definitions of game design is the one from Sid Meier
that states that:
A game is a series of interesting choices.
Further explanation of “interesting” often includes the following qualifiers:
• No choice should be consistently better than the others. (Or it would make
the other choices uninteresting or redundant.)
• The choices shouldn’t be the same. (It becomes meaningless to differentiate
between choices.)
• Choices must be informed. (Lest they become arbitrary or random.)
On the surface, this rings true. People are easily engaged in gameplay when they
are mentally challenged in an interesting way, especially if this is done in such a
Game Design vs. Level Design 5

way that the player stays engaged. Chess, for example, keeps the player engaged
by the multitude of options available, made interesting by the far-reaching stra-
tegic and tactical consequences within the game. Unfortunately, this description
of a game or gameplay is not all-encompassing; it simply doesn’t always apply,
something that I am sure Sid Meier himself would agree with.
Jesper Juul offers this:
But some games do away with interesting choices altogether. The
object of the music/rhythm games Dance Dance Revolution and Vib-
Ribbon is simply to hit the right buttons on the PlayStation controller
or dance mat at the correct time. These games do in fact not contain
any interesting choices whatsoever - but performing the non-interesting
choices is marked by some other form of enjoyment, namely that of
being in time with the music. They are still enjoyable games, which goes
to prove that interesting choices is not all there is to it.2
Unfortunately, within the field of game design, both in a practical professional
sense, as well as within the academic realm, there is no clear consensus on the
nature of game design, partially because it is hard enough to find agreement on
what it is that constitutes a game! To illustrate this point further, when I started
to research game definitions for this book in the hope to shed some light on the
topic, I found, much to my irritation, that I couldn’t find much overlap between
differing viewpoints. Instead, many views were contradictory, even when they
tried to incorporate as many “accepted” elements as possible. Several people
who have spent more time than me trying to define games have commented on
this.
Katie Salen and Eric Zimmerman, in their book “Rules of Play,” formally
compared eight notable definitions or descriptions of games and put the defin-
ing characteristics in a comparative grid. This is one of their conclusions:
All of the authors except Costikyan include rules as a key component.
Beyond this there is no clear consensus.3 Although 10 of the 15 elements
are shared by more than one author, apart from rules and goals, there is
no majority agreement on any one of them. (Emphasis mine.)4

2
Jesper Juul, “Just What Is It That Makes Computer Games So Different, So Appealing?”
IGDA The Ivory Tower, April 2003, available at http://www.igda.org/columns/ivory-
tower/ivory_Apr03.php.
3
Nonetheless, the authors then proceed to create a definition of games that is easy to
disagree with.
4
Katie Salen and Eric Zimmerman, Rules of Play: Game Design Fundamentals, MIT Press,
Cambridge, MA, 2004, p. 93.
6 Introduction to Level Design

Perhaps a definition of games is too much to ask for, as it is clear that it can-
not be unambiguously captured within a single concept. The same is true for the
concept of play, which is very much related, of course. It is, however, a topic that
is unsurprisingly, discussed as fiercely as that of game definition. Unsurprisingly,
because gameplay is the logical consequence of a game, and therefore they seem
to be aspects of the same thing, leading to the same disagreements. Isn’t it fair to
say that games cannot exist without gameplay? It is clear that play is central to the
experience. But is it the defining element? It is, according to some game research-
ers who argue that exact point. But what of other elements often represented
within games? For example, what of the narrative elements present in many game
types? Once again, disagreements or lack of consensus come to the fore.

Ludologist vs. narrativist perspectives

As much as is the case of practical game development or commercial game fund-
ing, some disagreements within academic circles are quite profound. A good ex-
ample of this can be found in the differing viewpoints often attributed to the
ludologists and the narrativists.5 Theirs is a disagreement that stems from a differ-
ent interpretation of what games are, and within which context to place the play
experience, a disagreement that has led to many articles and books, impassioned
speeches, and even heated arguments. Although their defining features are often
contested, they seem to be most clearly understood as follows.
Ludology is: A branch of game studies that approaches the subject through
the prism of play.
While narrativism is: A branch of game studies that approach the subject
through the prism of narrative.
This sounds straightforward enough, yet it has led to many, sometimes ill-
tempered disputes. Let’s look at the opinion of Michael Mateas,6 who offers the
following on narrativism:
The narrativists generally come out of literary theory, take hypertext as
the paradigmatic interactive form, and use narrative and literary theory
as the foundation upon which to build a theory of interactive media.
And on ludologists:
Ludologists generally come out of game studies [e.g., Avedon and
Sutton-Smith 1971], take the computer game as the paradigmatic

5
Narrativism is also often referred to as narratologism.
6
A scholar active in the field of artificial intelligence, among other things, and one of the
authors of Façade, an experimental interactive drama (Façade can be downloaded here:
http://www.interactivestory.net/).
Game Design vs. Level Design 7

interactive form, and seek to build an autonomous theory of inter-

activity. 7

In some instances, ludologists have placed themselves in direct opposition to

narrativist thinking, and vice versa. Differences between ludologist and narrativ-
ist thinking have produced so much friction that some even started attacking
each other’s viewpoints in public. (I am not going to go into specific examples,
but things got quite heated.)
Ultimately, in the eyes of many, neither view is correct, or rather, they are
both right and wrong at the same time, insofar as they both focus on legitimate
aspects of the equation but try to invalidate other equally appropriate ones. (I
am old and wise enough, however, to leave this debate to those willing to spend
their time on it.)
Far from being the only positions available, the narrativist vs. ludologist
standpoints illustrate how deeply entrenched people can become in their very
particular beliefs about games and gameplay. It is notable that in many ways,
both parties to some degree perpetuate a false dichotomy. Most narrativists, for
example, don’t deny the relevance of play, just the weight and importance given
to it by some ludologists. We can still agree that play is fundamental to games,
while also acknowledging the importance of narrative within a gaming experi-
ence. Narrative, however, will be discussed in detail in its own section of the
book: Part IV, Chapter 12.
To be fair, much of this debate is suspect insofar as that often, when pressed,
it seems there is only limited disagreement, which often stems from method-
ological issues more than from anything else.
Gonzalo Frasca states in the abstract of one of his articles:
During the last few years, a debate took place within the game scholars’
community. A debate that, it seems, opposed two groups: ludologists and
narratologists. Ludologists are supposed to focus on game mechanics
and reject any room in the field for analyzing games as narrative, while
narratologists argue that games are closely connected to stories. This
article aims at showing that this description of the participants is
erroneous. What is more, this debate as presented never really took
place because it was cluttered with a series of misunderstandings and
misconceptions that need to be clarified if we want to seriously discuss
the role of narrative in video games.8
7
Michael Mateas, “Michael Mateas Responds in Turn.” Electronic Book Review, http://
www.electronicbookreview.com/thread/firstperson/bestyled, 2004.
8
Gonzalo Frasca, “Ludologists Love Stories, Too: Notes from a Debate That Never Took
Place,” Ludology.org, http://ludology.org/articles/Frasca_LevelUp2003.pdf, 2003.
8 Introduction to Level Design

Even though the dust has now settled a bit, it is worth reflecting on how difficult
matters of definition can be, and game-related definitions are notoriously hard,
by any standard.

External Goals
Since we are looking at matters of function and purpose, we should look at game
design’s external goals. If we know what game design is supposed to achieve, we
have a much better idea of game design function. External goals are as funda-
mental as describing what something is for, and how this is achieved. Let’s take a
real world example: a chair’s design, and use it as a simple test case.
A chair’s design is subject to many requirements, but the main identifiable
goal is to allow a person to sit on it. An observation that may be banal in its sim-
plicity, but one that needs to be noted, nonetheless.
This basic goal leads to other related requirements that describe what the
chair has to be:
• strong enough to take the weight of most people,
• stable,
• affordable,
• moveable,
• aesthetically pleasing.
At this point, a designer comes in and starts to formulate these external goals
and requirements in a number of functional designs. Regardless of the content
of those designs, the design’s function or purpose is partly described by these
external rules.
The same principle occurs in game design. The basic function of a game de-
termines the game design’s function. So a first step in game design is to correctly
identify the game’s external goals and to interpret those in such a way that they
get represented well in the game’s rules.
This means, for example, that a game whose defining external goals are of a
commercial nature will end up radically different from a game whose external
goals are centered on delivering, for example, a disturbing artistic narrative.
If we decide that the main defining external goals are simply aimed at game-
play and profit, and if we apply the chair analogy to a game, we could state that
the game exists in order to provide a fun and profitable gameplay experience.
This then leads to other related requirements that describe what the game
has to be:
• pretty,
• easy to learn,
Game Design vs. Level Design 9

• hard to master,
• of sufficient quality,
• showcase high production values,
and so on.
It is important, however, not to confuse these with the game’s internal or
intrinsic goals.

Gameplay Goals
It has been noted earlier that a defining feature of most games is that they have
goals. These are the level design or general gameplay objectives that the game
itself presents to the user. We encounter these all the time, and it is easy to name
typical examples. Take the following list of player objectives:
• Attain the high score.
• Unlock the dungeon.
• Defeat the boss character.
• Win the race.
• Score more goals than your opponent.
• Explore the environment.
• Shoot the enemy soldiers.
Most gameplay is driven by these kinds of explicit objectives and motivations for
the player, a fact that is hardly controversial. To the level designer, however these
goals must be designed. And in order for them to be designed, they need to have
a logical source or reason to be included in the levels. This reason is generally
found in the game’s internal or intrinsic goals.

Internal goals
These are similar to a game’s external goals, insofar as they describe high level
goals from which we can derive gameplay requirements. The difference with ex-
ternal goals lies in the fact that internal goals govern the high level goals that are
directly related to aspects of the game and gameplay itself, as opposed to external
factors. They tend to cover things like the following:
• Empower the player.
• Teach the player how to have fun with the game.
• Don’t break the player’s suspension of disbelief.
• Give the player a sense of achievement.
• Reward the player for exploration.
• Provide addictive, fun gameplay.
10 Introduction to Level Design

We will look in more detail at these kinds of goals and how they fit in a develop-
ment and level design hierarchy in Chapter 3.

Defining Goals and Designing Rules

What all this tells us is that both internal and external goals are part of game de-
sign function. It describes functional necessity related to gameplay AND related
to the game’s more existential9 goals, like turning a profit or making a particular
artistic statement.10 Goals like these are useless by themselves unless they get
translated into actions. How do we achieve these goals? In the case of game de-
sign, this means that once the external or meta goals have been defined, the de-
signer needs to design the actual game, which means designing the game’s rules
in such a way that they best support the external goals.

Rules
A fundamental aspect shared by most, although not all, video games, is that they
adhere to a formal set of rules. Games without rules can exist, but they are ei-
ther very abstract in form, or function more on the level of toys. Nonetheless, it
cannot be denied that by far the majority of all games, not just video games, are
based on or reliant on a formal set of rules, often predetermined, that the player
has to follow in order to successfully play the game.
Some people go even further and argue that games cannot exist without
rules:
Rules are what differentiate games from other kinds of play. Probably
the most basic definition of a game is that it is organized play, that is to
say rule-based. If you don’t have rules you have free play, not a game.
Why are rules so important to games? Rules impose limits—they force
us to take specific paths to reach goals and ensure that all players take
the same paths. They put us inside the game world by letting us know
what is in and out of bounds.11
However, this construction of formal game rules is completely abstract until ex-
ecuted in play. This is similar to theater, where the actual play does not exist until
the performance takes place. The text of the play can be read in its own right, of
9
Suggestions for a less pompous term are welcome.
10
Chapter 3, “Level Design Goals and Hierarchies,” will cover these subjects in much
more detail, focusing both on external and internal goals and how they relate to level
design.
11
Marc Prensky, Digital Game-Based Learning, McGraw-Hill, New York, 2001, p. 14.
Game Design vs. Level Design 11

course, but the actual theatrical play only takes place during the performance.
The same is true for games. Although a game’s design may be able to formalize
the rules of the game, until actual play occurs, this design is unfulfilled, and in
many ways the game itself is incomplete. Game design facilitates play by design-
ing rules under which play can occur.12

A quick word on toys

Toys are often excluded from definitions of games because they don’t have a set
of formalized rules associated with them. This is true to a degree but is somewhat
misleading. It is more accurate to say that toys don’t have a set of predetermined
rules associated with them. This does not stop those who play with toys from
formulating their own rules spontaneously at the time of play. The end result is
the same as in other games: the player is actively engaged in gameplay. The newly
formalized rules may be simple, for example a game of catch between two, such
as parent and child, but they are gameplay rules nonetheless.
The conclusion that follows from this is that toys are facilitators of games that
ask players to define their own rule set. Crucially, the player(s) temporarily take on
the guise of game designer and level designer.

Game Design Function Summary

It appears to be difficult to agree on a definitive view on what games are. There
are countless definitions of games and their associated viewpoints, and they of-
ten are in disagreement with each other. But study of games, gameplay, and game
rules shows that there are a number of commonalities that can be highlighted:
• Play is central to games.
• Diverse and unrelated goals can motivate the production of games.
• Most games rely on rules, or facilitate the definition of them.
If seen in this light, we can describe a game as featuring: an often predetermined,
agreed-upon set of rules, which are designed to facilitate gameplay. The motivation
behind the creation of a game itself can be diverse, for example including commercial,
educational, artistic, or other elements.
All of these elements are individually fairly obvious; yet taken as a whole,
they spell out something fairly useful with regards to finding a workable concept
of the function of game design. Within all of these observations lies an answer to
the question about game design’s function or purpose, because they describe what
a designer needs to achieve.

12
It is also good to note that just facilitating play is not enough to guarantee a good game.
12 Introduction to Level Design

Ultimately, a game designer is the person who determines the rules by which
a game is formulated, in order to achieve the goals for which it is created. And in
some ways, a good game designer is one who is good at determining what rules
are appropriate for the desired gameplay.
A game design is a coherent set of rules that formalizes a game’s content in such
a way that it facilitates appropriate gameplay, in order to achieve the game’s funda-
mental goals.

The Function of Level Design

Now that we have spent considerable time looking at game design function, we
need to compare this with level design function. We have seen that in game design
it is very important to define what the game’s external goals are and design rules
of play that correctly support them. Questions about level design in many ways
seem to start from a completely opposite position. The rules of play are a known.
How else can we construct a level if we don’t already know what rules it has to
facilitate?13
On closer inspection, however, we are left with a similar definitional prob-
lem as we had at the beginning of this chapter with regard to game design. Where
does the rubber hit the road? How do we define this? Instead of trying to find an
ultimate definition of level design, I would like to focus, just as we did with game
design, on finding a useful description of the function of level design. The reason
for trying to formulate an overall function or purpose for level design is that it
should give us a way to determine what is within the level designer’s responsi-
bilities. This will give us more than a job description; it gives us a conceptual
framework within which we can do our work. This is something that may sound
unimportant on paper, but is nonetheless of vital importance in practice when
we need to defend or explain our professional or artistic choices. (Even in those
cases when we have to justify them to ourselves.) Or to put it in less dramatic
terms; it gives us a practical framework through which we can approach level
design.
A useful start to this endeavor is to look at level design in a historic context.

Level Design in a Historic Context

It is beyond the scope of this book to provide a complete history of level design,
although it would be a fascinating project to attempt to do so. Instead, a short
13
Scarily, in commercial level design, one is often asked to design levels without a clear
understanding of the final gameplay parameters.
Game Design vs. Level Design 13

look at a number of historically interesting examples of level design or related

fields will have to do.14 Even this limited focus should produce some insights, as
there is much to be learned. At the least, it should provide us with some historic
context in which we can place level design.

Sports
Almost all sports take place within defined spaces. And, more importantly, most
sports take place in designed spaces. At some point in time, somebody actually
decided on the dimensions of a soccer field, the size of a hockey goal, or the
placement of hurdles in an equestrian15 course. How these original decisions by
proto-level designers were made we don’t always know, but it is clear that they
allow for an important function of sports: competition. In order to compete un-
der fair terms, their design allows a level playing field. (A term that sounds much
like a video game level to me.) A sporting field or environment usually cannot
be altered or bypassed by the participants of the sport. It is literally against the
rules, and the offender typically gets punished heavily or even disqualified from
participating further.

Board-game layouts
Board-game design is even closer to level design for video games, partly because
it allows the creation of an abstract representation of an environment. There is not
always a need to create a field of even grassland with play zones demarcated by
chalk lines, or to run divisions of soldiers through complex tactics and strategy
drills out in some field, if a similar effect can be created by an approximation or
an abstraction in the form of a board game. Chess, for example, is a good case;
the game portrays warfare and enhances strategic thinking, despite using a play-
ing area that is rather abstract.
Furthermore, board games provide scope to introduce elements of the
fantastic into play. In the context of a board game, it is fine to teleport players
through the world, or to introduce mythical monsters as adversaries. A board
game can introduce elements of chance (pick a card) and encourage the use of
avatars.16 Many of these choices are affected by the board’s layout, which had to
be designed at one point in time. Literally thousands of board games have been
designed through the years, and the inherent level design choices that were
made provide a rich source of information. They are especially interesting from
a historic point of view, because they go back many hundreds of years.
14
I do encourage people to do some of their own research in this area.
15
Am I the only person who thinks of a platform game when these horses jump?
16
A game piece that represents the player.
14 Introduction to Level Design

A good exercise for budding level designers would be to choose any board
game, try to find out why the board was designed the way it was, and try to im-
prove on its design. This is a guaranteed way to improve as a level designer17 and
as a byproduct is likely to teach some appreciation of board game designers, as
well.

Pinball machines
Another beautiful example of proto-level design can be found in pinball games.
The basic rules of pinball games can be summed up on the back of a napkin. As
a set of rules describing a game, there really isn’t much to it. Yet many hundreds
of iterations of such games have successfully persuaded players all over the world
to feed them coins. There are a huge number of pinball tables whose layout and
content design, or in other words, their level design, showcases new and success-
ful interpretations of those old and basic rules. The player still controls flippers,
the table is still slanted so the ball rolls down, and the game offers three “lives” to
earn a maximum amount of points.
Yet there is no shortage of unique and wildly differing pinball tables. To-
gether, they provide an interesting and enduring example of an interactive game
type that predates video games.

Dungeons and Dragons

In 1974, Gary Gygax and Dave Arneson designed a new type of game still en-
joyed proudly by the geek tribes of the world. They created one of the first suc-
cessful pen-and-paper roleplaying games and called it Dungeons and Dragons.
The basic setup of the game consists of a group of players sitting around a table
and enacting the roles of diverse player characters within a virtual fantasy setting,
designed and described by the dungeon master. The dungeon master literally de-
scribes this virtual world to the other players in such a way that the players can
imagine themselves to be there in their own imagination. The dungeon master
tells the players what they encounter within this world, and the players describe
their actions and reactions to the dungeon master, role-playing (play acting)
their player characters. The physics and mechanics of this world are documented
in complex and extensive rule books sold by the publisher, the adventures (mod-
ules) that the players experience within this virtual setting are designed before-
hand, either by the dungeon master or by an independent designer. A skilled
dungeon master can take the somewhat impersonal, systematic rules on how
the world and its inhabitants or processes behave, and through the use of a well
17
Game designers should try creating better rules or even design the board games them-
selves.
Game Design vs. Level Design 15

designed adventure, really bring it to life, providing the other players with an
extremely compelling play experience.
Key to this, though, is the earlier prepared adventure, which functions as a
perfect example of a level design outside of video games. The dungeon master
or independent designer takes on the role of level designer because he or she
ends designing the in-game encounters and dramatic occurrences that define
the player reactions in the game. Although often maligned as an activity, Dun-
geons and Dragons pioneered a fascinating new way of playing deeply immersive
games. This was borne out by the huge number of players who since have bought
the games and associated products, the countless other similar games that have
since become successful, and the countless tie-in products sold. But ultimately
for the purpose of this book, this provides an interesting example of alternative
level design.

Lessons from history

What these historic examples show us is that level design is not exclusive to video
games, but instead can be found throughout very diverse other types of games.
Level design never exists purely on its own terms. But what the above examples
have shown is that most game designs also don’t exist within a vacuum; instead,
they can only work together, in unison. This so far shows us that there is indeed
a very close link between game design and level design, but it does not explicitly
tell us what level design’s main function is.
Unfortunately, due to the very young age of level design as a profession,
there is not much recourse to be found in professional literature. Hardly any
serious texts exist on the subject, and many of those are relatively old or cover
the mechanics of level production more than anything else. Even so, some
opinions and views have been expressed in the past that provide unique in-
sights.

Various Views on Level Design

Throughout the limited history of level design as a unique discipline, various
people have tried to describe or define the field, often based on nothing but
their own hard-won professional experience. These definitions are fascinating in
many ways. It is always wise to heed the words of pioneers in any creative genre,
as typically these early works were the result of completely original thought pro-
cesses led by novel problems, and not led by existing conventions or styles. And
especially since many of these early designers have demonstrated their skills and
abilities by providing high-quality examples of their craft. Let’s take a look at a
number of notable examples.
16 Introduction to Level Design

Example 1: Jay Wilbur

Level design is where the rubber hits the road.18
Much of level design is about making sure the player is taught19 the rules of play.
An important part of the act of level design is taking all the diverse game ele-
ments, teaching the player the associated rules, and using the means available to
put them together into one coherent whole. This is what Jay Wilbur meant by his
quotation. The game design at one point has to be put into practice, and to do
so, the level designer needs to be able to put all the diverse elements of a game
together in such a way that it doesn’t fall apart when tested in the real world.

Example 2: Sam Sharami

Level designers, or map designers, are the individuals responsible for
constructing the game spaces in which the player competes. As such,
the level designer is largely responsible for the implementation of the
game play in a title.20
This is an interesting view of level design insofar as it talks about its goals and
the level designer’s responsibility. It touches upon the important fact that level
design is responsible for gameplay implementation.

Example: John Romero

A level designer has a very responsible position, because maps are where
the game takes place.21
Again, this is an interesting observation because it makes a comment on what
maps (levels) are. If they are “where the game takes place,” it follows that levels
allow the game to exist, or at least to be played. An obvious point perhaps, but it
tells us something about the strong link between game and game levels.
18
Jay Wilbur in conversation with Cliff Bleszinski, as reported in Game Design: Secrets
of the Sages, MacMillan, New York, 1999, Ch. 6. Wilbur is a very famous designer who
worked on titles like Doom and Quake, or more recently, Gears of War.
19
See also Chapter 2, which discusses level design as a “teaching mechanic.”
20
Sam Shahrani, “Educational Feature: A History and Analysis of Level Design in 3D
Computer Games—Pt. 1,” Gamasutra - Features, http://www.gamasutra.com/view/fea-
ture/2674/educational_feature_a_history_and_.php, 1999.
21
John Romero, notorious level designer, programmer, and game designer, who has
been involved in many famous and even controversial games. In John Romero, “ION
Storm,” Chapter 6 of Game Design: Secrets of the Sages, Macmillan, New York, 1999; also
available at Gamasutra - Features, http://www.gamasutra.com/features/19990723/
levdesign_chapter_05.htm.
Game Design vs. Level Design 17

Summary of Views
We have heard that level design has to take all the disparate elements in a game
and make them gel, that level designers are responsible for the implementation
of the game, and that levels are where the game takes place. These observations,
combined with what we have learned from historic examples of level design,
provide us with something we can draw some initial conclusions from, and give
us some guidance into what to examine next.

Level Design Function Summary

There seems to be no shortage of opinion and ideas on the topic of level design.
It is disturbing, therefore, to note that, just as with game design, there is no clear
underlying theme to these observations or definitions. This chapter has touched
on a large number of diverse subjects, and throughout, we have reached a num-
ber of general conclusions. Many of these conclusions aren’t individually that
revelatory, but if taken together and placed in a shared context, they do provide
useful results. Let’s summarize and see where that takes us.

Codependency
It should be clear by now that level design and game design are not the same
thing. It has also become clear that they are codependent and interrelated: one
is useless without the other. Most games are unfulfilled without some kind of
level design, while level design is an interpretation of a game’s rules. Because of
this codependency, it is very important to realize that we shouldn’t study one
without studying the other. To understand level design we have to understand
game design. How can we interpret a game design without knowledge of it? And
conversely, how can we define rules for an experience we do not understand?

Game and Level Design Function

This book is about level design, but if we take into account the findings of this
chapter, this means that, at least to some degree, it is also about game design.
If a game designer designs the gameplay rules, the level designer designs how
the player is confronted with those rules. Looking at it from that angle, a level
designer and a game designer have completely different jobs. A game designer
formulates the game’s rules, while a level designer interprets them for maximum
results. To some degree, one represents theory while the other represents prac-
tice. This is the basic function of level design.
18 Introduction to Level Design

Play and Application of Game Rules

Just as a theatrical play needs a performance to be complete, a video game’s rules
need gameplay to occur. This is a basic purpose of level design, to interpret the
game rules, and to translate them into a construct (a level) that best facilitates
play. Another way of expressing this is by stating that “level design is applied
game design.” Not the most impressive definition of level design out there, but
it is suitable for the needs of this book. It describes much of the function and
purpose of level design, and therefore, much of the work of a level designer.
So, for the purpose of this book, level design will be defined as applied game
design, not as a separate function subordinate in a game design hierarchy,22 but
as a description of its main function and purpose. This does not mean that I will
focus overly on game design issues in this book to the exclusion of level design
issues; far from it. Rather, it means that in order to understand certain level de-
sign issues, we have to understand certain game design issues. They are different
sides of the same coin.

Area of Responsibility
These observations on level design’s function or matters of definition are not
unimportant or abstract. If we ourselves don’t understand the nature of our work
it will be impossible to confidently defend it. This isn’t always achieved by the
content we create, although that is obviously of the utmost importance. We need
to be able to explain to others (as much as to explain to ourselves) why we made
those choices in the first place. As often as not, we need to be able to make clear
and defensible choices from the get-go; to do this; we need to know within what
area of responsibility we work. It is helpful to be able to work from within a clear
framework and to be able to say what the function of our work is and what ar-
eas it covers. This furthermore allows us to create clearly defined goals for our-
selves, a theme that will be further explored in the imaginatively named Chap-
ter 3, “Level Design Goals and Hierarchies.”

22
Is a written play superior to its theatrical performance?
Teaching Mechanisms
2
W e have established in the previous chapter that: “level design is applied
game design.” This tells us something about the function of level design,
but it does not give us enough information on the core content that a level de-
signer needs to provide. This leaves important questions unanswered. What are
the intrinsic internal workings of level design, as opposed to its external goals?
What is it that we are trying to do in the context of the game?
We already know how closely game design and level design are intertwined.
If we examine both at the point where they overlap the most, we start to take a
look at fundaments of gameplay. In order to define the nature of level design, we
will have to define the nature of gameplay, and how it relates to level design.
The following three sections provide a detailed examination of these mat-
ters. They also provide a preview of the methodology used later in this book,
where a given subject is examined through chapters covering the subject’s ba-
sic concept, how it applies to level design theory, and what applications it has in
practice.

Concept
The Nature of Gameplay
Anybody who has spent any time around animals, perhaps a pet dog or a cat,
knows that they are very playful creatures. It is very easy to start playing a game
with them on the basis of rules that are surprisingly easily understood.
• Catch the ball.
• Retrieve the stick.
• Let’s pretend my hand is prey!
• Obstacle course!

19
20 Introduction to Level Design

It is clear when we watch the behavior of animals of a certain level of in-

telligence, that play comes naturally to them. This becomes even clearer when
we watch them at play when they are young. When young animals at play are
observed, it is clear that almost anything in their environment can be an excuse
to initiate gameplay. To a puppy, for example, almost anything can be incorpo-
rated into gameplay, and that is true without anybody teaching the animal how
to engage in this behavior. It is easy to observe that gameplay comes naturally to
animals of sufficient intelligence, which hints at the possibility of play being fun-
damental to their well-being, due to evolutionary reasons. It may be linked to the
animal’s survival. This is as true for animals as it is for human beings.

Survival Skills and Make-Believe

Play is a relatively well-understood phenomenon. People from fields as diverse
as behaviorism, anthropology, and biology have studied it, and a number of gen-
eral findings can be agreed upon. First and foremost, it has to be understood that
it is clear that there is very serious reason for this innate ability to be playful; it
helps maximize the animal’s chances of survival. Through play, valuable lessons
are taught that clearly demonstrate this point. Through play, the young animal’s
skills are honed that are necessary in order to hunt, fight, mate, hide, or one of
many other activities that are key to survival as an adult. Games provide a safe
context in which these lessons can be learned through play. With this in mind, it
is not controversial to state that “Gameplay teaches skills that are important and
necessary in order to survive in real life.”
This seems a straightforward-enough statement, but upon examination, a
number of startling further conclusions can be reached. Not the least of which is
the realization that animals are capable of grasping abstract concepts like games
or make-believe It is irrefutable, however, because to engage in this kind of safe
play, human beings (or animals that are capable of gameplay) need to be able to
accept imposed boundaries and rules to their behavior. They need to understand
that the gameplay experience is an artificial one. This means that fairly complex
and abstract concepts are at play. We are, after all, talking about the understand-
ing of something that is by definition an abstract construct, governed by a set of
formalized rules.
Or to put it in other terms, when engaged in gameplay, we need to under-
stand the difference between the rules that govern our reality, and those that
govern the make-believe, or virtual, world of a game. This is quite an amazing
skill, and the fact that we as human beings are adept at constructing and ma-
nipulating our experiences within these parameters is nothing less than remark-
able.
Teaching Mechanisms 21

Our propensity towards gameplay has far-reaching and interesting conse-

quences. We see this ability to accept artificial, invented realities reach into areas
beyond gameplay. For example, it is easily identifiable as crucial to the enjoy-
ment of film, literature, art, music, and countless other forms of art and enter-
tainment.
I will go into further detail with regard to this ability in Chapter 8, on im-
mersion.

“Fun” as a Reward for Gameplay

It is safe to say that good games are rewarding. A good game is fun, or makes us feel
good. But what is it exactly that creates this reward for us, and how does it work
on a biological level? It needs no explanation that the answers to these questions
are valuable to any level designer.
From studying gameplay in animals, we have learned that engaging in game-
play makes the animals feel good. They want to play games from a very young age
on and need no prompting by external factors. In fact, they often do their best
to initiate gameplay tendencies in others. Some of this behavior may originate
from the fact that engaging in gameplay causes chemicals to be released in the
bloodstream that act as a reward for playful behavior. This in turn makes animals
feel encouraged to engage in this behavior. This is no accident, as the rewarding
aspect of gameplay is biologically necessary. (This necessity stems from the need
to learn survival skills within the safe context of a game, as we established previ-
ously.)
It only requires a small step to take this information and extrapolate to hu-
man behavior, which is basically the same. In fact, as already noted at the be-
ginning of this chapter, our ability to understand games crosses the species
boundary,1 which is a strong indicator that comparable processes are at work.
Much of human gameplay, when examined, bears striking similarities with
gameplay in some advanced animals. Hide and seek or tag come to mind. We
also are rewarded when we engage in gameplay, and in the case of humans, the
chemical award is the release of certain pheromones, which make us feel good,
or in other words, we experience “fun.2” Our large brains enable us to engage in
games that are much more complex than those enjoyed by animals, but all the
basic principles still apply.

1
If we take a moment to think about this, it should be apparent how truly extraordinary
this is.
2
See Raph Koster’s book, A Theory of Fun for Game Design (Paraglyph Press, Scottsdale,
AZ, 2004), for similar sentiments.
22 Introduction to Level Design

Aptitude for Gameplay

All around the world, people, adults as well as children, play games in which the
rules are easy and even intuitively understood. This suggests that we as a species
are very good at “speaking the language of games.” We know, for example, that
we can play games with people whose linguistics aren’t ours, whose culture we
don’t share, and whom we have never met before. It is even clear that we can play
games with creatures that don’t even belong to our species! However we choose
to look at the subject, it is clear that human beings possess a certain amount of
aptitude for gameplay.

Universal gameplay grammar

One could even argue that human beings use something akin to a universal game-
play grammar that allows us to understand and play games easily, often without
regard to their origins.
Although beyond the scope of this book, it would be fascinating to study
this topic further, as has been done in the field of linguistics, where much study
and discussion exist around the field of generative grammar, for example through
the work of Noam Chomsky, not least because of his work in the field of transfor-
mational grammar. Here are some key concepts:
[T]ransformational-generative grammar, [a] linguistic theory asso-
ciated with Noam Chomsky, particularly with his Syntactic Structures
(1957), and with Chomsky’s teacher Zellig Harris. Generative gram-
mar attempts to define rules that can generate the infinite number of
grammatical (well-formed) sentences possible in a language. It starts
not from a behaviorist analysis of minimal sounds but from a rationalist
assumption that a deep structure underlies a language, and that a similar
deep structure underlies all languages.3
Exciting as a “unified theory of gameplay” sounds, I think I will leave further
examination of this for another book for now.

Support from Within the Field

Many notable people within our own industry have reached the same or similar
conclusions. Raph Koster, the author of A Theory of Fun for Game Design, has the
following things to say:
3
“Transformational-Generative Grammar,” The Columbia Electronic Encyclopedia, Sixth
Edition, http://www.encyclopedia.com/topic/transformational-generative_grammar.
aspx, 2001–2007.
Teaching Mechanisms 23

One of the subtlest releases of chemicals is at that moment of triumph

when we learn something or master a task. This almost always causes
us to break into a smile. After all, it is important to the survival of
the species that we learn—therefore our bodies reward us for it with
moments of pleasure. There are many ways we find fun in games, and I
will talk about the others. But this is the most important.
Fun from games arises out of mastery. It arises out of comprehension. It
is the act of solving puzzles that makes games fun.
In other words, with games, learning is the drug.4
Taking inspiration from Flow Theory,5 which offers a scientific and well-re-
searched approach to “happiness,” Raph Koster describes processes that are very
similar to our findings. Although he was talking specifically about game design,
it is nonetheless as relevant to level design if one thinks of level design as applied
game design. (See Chapter 1, “Game Design vs. Level Design.”)
To find more support from the game industry community, let us also con-
sider the following from Carolyn Handler Miller:
The earliest games were developed not for idle amusement but for
serious purposes: to prepare young men for the hunt and for warfare.
By taking part in games, the youths would strengthen their bodies
and develop athletic skills like running and throwing. By playing with
teammates, they would also learn how to coordinate maneuvers and
how to strategize. Over time, these athletic games evolved into formal
competitions. Undoubtedly, the best known of the ancient sporting
events are the Greek Olympic games.6
In these ancient games, we find another clear indication of gameplay as a
teaching device for events in real life.
Many other examples illustrate that there is some support for the view that
gameplay has a strong basis in teaching mechanisms.

Support from Other Fields

This notion that human beings possess a native ability to understand and engage
in gameplay is not a new one. People in other disciplines who have been studying
4
Raph Koster, A Theory of Fun for Game Design, Paraglyph Press, Scottsdale, AZ, 2005,
p. 40.
5
Explained in Chapter 8 in the section “The Zone.”
6
Carolyn Handler Miller, Digital Storytelling: A Creator’s Guide to Interactive Entertain-
ment, Elsevier Science, Amsterdam, 2004, p. 27.
24 Introduction to Level Design

this concept and approached it from differing angles have come up with similar
conclusions. Currently, a reasonable amount of data has been collected to sup-
port these findings, and more is uncovered on a regular basis.
Researchers suggest that social play may safely teach young the skills
they will later use in aggressive social competition. However, such
play may also simultaneously strengthen social bonds between group
members, a process that serves to limit the amount of actual aggression
between group members.
If social play can mirror real aggression in appearance, what then lets
animals know that their partners are only playing? Animals, it turns
out, communicate playful intentions with certain stereotyped signals.
The most widespread play signal is the play face, a relaxed, open-mouth
expression seen in many mammal species, used virtually from birth.
The human smile almost certainly evolved from this ancestral trait. If
someone smiles or laughs while hitting you in the arm, you realize that
his or her intentions are very different than if he or she is frowning, with
a tightly closed mouth.7
More and more studies and articles are appearing that explore gameplay for edu-
cational purposes. Consider the following example:
Games[‘] greatest potential is that they’re worlds in a box. They allow
you to create a world that somebody can be in and take on an identity.
People learn most deeply when they take on a new identity that they
really want. Let’s say I really want to know what it’s like to be a biologist
of a certain sort. I really want to know what it’s like to feel that way, to
value that way, to talk that way. I can do that now. I can be in that world.
That’s going to be a deeper form of learning.8
A whole subgenre called serious games has appeared recently, which heavily relies
on the educational aspects of gameplay and uses them specifically with the con-
text of educational games.

Concept Summary
Our findings have taught us that from a very young age, human beings are pre-
disposed towards playing games, just like many animals, in order to learn impor-
7
Alex Hawes, “Jungle Gyms: The Evolution of Animal Play,” National Zoo| FONZ,
http://nationalzoo.si.edu/Publications/ZooGoer/1996/1/junglegyms.cfm, 1996.
8
Joel Foreman, James Paul Gee, J. C. Herz, Randy Hinrichs, Marc Prensky, and Ben Saw-
yer, “Game-Based Learning: How to Delight and Instruct in the 21st Century,” EDU-
CAUSE Review 39:5 (September/October 2004), pp. 50–66.
Teaching Mechanisms 25

tant survival skills, in a safe environment. It may even be hardwired in our brains,
and we are rewarded with pleasure if we engage in playful behavior. We call this
pleasurable feeling fun. Related to this predisposition towards gameplay is our
ability to easily and readily suspend our disbelief, when confronted with virtual
experiences.
Games can be considered a teaching mechanism. One of their most important
purposes is to teach vital life lessons through gameplay. This concept of educa-
tional gameplay, from a biological point of view, is far-reaching and fundamental
to us as human beings. It is literally a part of our behavioral makeup for reasons
of survival, and understanding the processes at work is vital to our understand-
ing of level design. Many of the same underlying mechanics at work with regard
to play and traditional games apply to level design. Gameplay and educational
processes can form a natural match, examples of which can be found in new
gameplay-based educational programs, serious games, and scientific literature.
Nonetheless, it is important to keep in mind that evolutionary play is a very
controversial subject, and to this day, arguments in favor and against are still de-
bated.
Explanations of play that involve either proximate or ultimate cause,
or both, are common in the literature. However, though evolutionary
explanations—and hence ultimate explanations—of play pepper the
literature, their success in answering the question, “What is play for?”
has been limited.9
Nonetheless, play as a teaching mechanism provides a worthy area of knowledge
for level designers to dip into. There is much useful data there to link to level
design theory.

Theory
If we are predisposed to gameplay because it teaches us survival skills, it stands
to reason that we examine the teaching aspects of this concept further. The idea
that games can be seen as a teaching mechanism, born out of biological or evolu-
tionary necessity, is certainly interesting. Based upon our finding so far, we can
formulate the following statement:

9
Garry Chick, “What is Play For? Sexual Selection and the Evolution of Play,” keynote
address presented at the annual meeting of The Association for the Study of Play, St.
Petersburg, FL, February 20, 1998.
26 Introduction to Level Design

In purely biological terms, a game is an artificial construct, designed to

safely teach survival skills, and in doing so, rewarding the player with
pleasure.
This statement seems fair and accurate, but how does it help us understand or
apply level design? Let’s see if there is conceptual overlap between this general
concept and specific level design theory by rephrasing the previous statement so
it fits within the parameters of video games. A translation into video game terms
would look like this:
A video game is an artificial construct that, when well made, rewards
the player with fun.
If we equate the concept of teaching survival skills from the first statement with
well made from the second statement, we find that they equate very well.
There is a strong correlation between the rewards we receive from play in
general, and those we receive from playing video games. This is not strange, since
video games are just a different form of games, and therefore are subject to many
of the same underlying principles. However, we are not used to looking at video
games in that way, even though it makes sense to do so.
Much traditional play is all about teaching skills, and testing the player’s profi-
ciency. It is rewarding to master a task, or to be good at something. Games teach
motor skills, mental skills, skills of reasoning, and so on. Gameplay allows us to
put those skills to the test in a controlled manner. A good game strikes a balance
between teaching these skills and providing the player with an enjoyable testbed
in which to try them out. There is no reason to think that video games are any
different, and in extension, the same is true for level design.

“Good” Level Design

As we have already established in Chapter 1, “Game Design vs. Level Design,”
we can state that level design is the application of game design. We now also know
that in practical terms, we can say that “good level design teaches the player how to
play and enjoy the game.”
This is one of the most important concepts in level design, and when better
understood, one that good level designers will keep coming back to, time after
time. It can be used in almost any aspect of level design and can relate the small-
est gameplay mechanic to the largest span of levels.
Let’s look at a small number of areas where this can be applied:
• the physical rules of the environment,
• the abilities of the player’s in-game character,
Teaching Mechanisms 27

• the behavior of enemies,

• the game’s reward systems.
These are all major gameplay areas, and the game as whole would suffer if their
rules were not properly taught to the player. Conversely, if the player is taught
well how to deal with these areas of gameplay, this will form a solid basis for a
good gameplay experience. This is part of the reason why the game cannot be
good if the level design is not good. The underlying game design can be the best
in the world, but if the level design does not support it by teaching the player the
rules of this brilliant game, it is all for nothing. Furthermore, if the game design
does not support education though level design, it will never be fully enjoyed by
the player.
If we accept that this teaching role is absolutely fundamental to level design,
it makes sense to try to define the best techniques available to us in achieving this
teaching goal. This is what I mean by teaching mechanisms.
Good level design is not just teaching the player the rules of the game, but
also allowing the player to use those rules in a way that is rewarding and fun.
Much of the fun comes to the surface when the player is tested.

Teaching Mechanisms vs. Testing Mechanisms

Teaching mechanisms are meaningless unless that which is taught is tested and
put into practice. There has to be a way to test the player’s knowledge or profi-
ciency within the game, or the game may lack purpose. This is another intrinsic
goal of level design. If the gameplay is taught well and the player gets tested in an
enjoyable manner, the level designer has done a good job. Countless lessons and
techniques can be derived from this fact alone. Examples of this balance can be
found in many successful video games, and throughout the book many of these
will be referenced or new ones will be explored.
Testing10 the player’s skills and knowledge is an integral part of the teaching
mechanisms in level design. From now on, when I refer to teaching mechanisms,
it can be assumed that I am also talking about testing mechanisms.

Teaching Gameplay and Reward Systems

At the basis of much level design success lies the ability to show the player how
the game’s play mechanics tie in with the game’s reward mechanisms. What ac-
tions and skills does the player have to master in order to get rewarded by the
10
It is important to note that testing does not necessarily equal challenging. A test can
constitute a challenge, but not every test has to be a challenge.
28 Introduction to Level Design

game? Where are the most enjoyable sections to be found, which weapons pro-
vide the best results, what creatures are most fun to play with, and which skills
should the player train first? All these questions present a teaching dialogue be-
tween level designer and player.

Inappropriate Gameplay
When we teach the player how to play the game at its most effective or its most
enjoyable, it is expected that those lessons are meaningful. Players implicitly
trust the game to teach them techniques that are consistent and trustworthy
throughout the game.
It is therefore important for us as level designers not to betray that trust, and
to create gameplay scenarios or puzzles not only for their own sake,11 but also in
context of the gameplay appropriate to the game as a whole. To put it simply, if
possible, we should not create situations where the player’s skills are useless.
For example, we should avoid situations that arbitrarily or in an unan-
nounced way deviate from the needed skill set and require something from the
players that they have never been taught.
This lesson is easily (and often) forgotten. A typical example is found in boss
fights12 that don’t use taught gameplay skills—and they are rife in game levels.
Please be aware of pitfalls in this regard.

Teaching Mechanisms in General Areas of

Level Design
It is impossible to try to identify all gameplay mechanics and try to find ways to
teach them best to the player. Not only is this impossible due to the scope of the
task; it also wouldn’t cover new genres of gameplay or unexpected gameplay oc-
currences. Furthermore, mechanics differ wildly between genres.
A more sensible approach lies in trying to find common themes that can be
applied to diverse situations. I have identified a number of areas that cover most
aspects of level design and deserve a further look:
• the game’s main goals and rules,
• the abilities and limitations of the game’s player character,

11
Although there is place for discreet isolated puzzles, or even games dependent on
them.
12
Boss fights are climactic fights between a player and an extra-powerful computer con-
trolled opponent.
Teaching Mechanisms 29

• the physics and scope of the gameplay world,

• the abilities and limitations of the non-player characters,
• success strategies available to the player.
Each of these areas covers a very wide range of gameplay issues and is worth
exploring further.

The Abilities and Limitations of the Game’s

Player Character
A player’s in-game character or avatar provides the player with much of his or
her interface with the game world. To a large degree, players experiences and
interacts with the world through their player character. This is of course less
important to games where the main gameplay does not derive from interacting
through an avatar, but the principle still stands that players need to know how
they can manipulate or interact with the virtual world they are exposed to. For
example, it is just as important in Tetris to know how to rotate, direct, or drop
down shapes as it is in Tomb Raider to know what jump distances are viable or
what surfaces can be climbed.
This need to know defines the player’s abilities and limitations in context
of the game’s (virtual) reality. It is very important to teach this early on in the
gameplay experience, because the earlier this is accepted by players, the ear-
lier they are able to suspend their disbelief in what is after all an artificial con-
struct.
This need is actually fairly typical in all kinds of games. In chess, for example,
half of the rules are linked to knowing what the abilities and limitations of the
individual chess pieces are. There is in fact no point in playing the game until you
know. (This does not mean that in video games these rules can’t change later on,
but in most cases, these changes have to come after the player has been taught
the fundamental and basic rules.)

The Physics and Scope of the Gameplay World

Just as it is important to define the abilities of the player character, it is important
to teach players how they interact with the game’s environment. This is impor-
tant because it gives further context in which to perform actions. In some ways,
the game world can be seen as a character in itself.
Players’ actions by themselves don’t have much meaning unless they are
performed within a defined physical context. The game world is a large part of
this.
30 Introduction to Level Design

The Abilities and Limitations of the

Non-Player Characters
If a level designer works on a game that features non-player characters, which is
extremely likely, the player needs to be taught how to interact with them. This
does not only cover enemy AI (artificial intelligence) but any AI characters pres-
ent in the game.

The Game’s Main Goals and Rules

In most cases (there are exceptions) it is wise to assume that players need to
learn what constitutes the core gameplay experience of the game they are play-
ing. In other words, they need to know early on what the game’s main objectives
are and how to achieve them. This is true on a level-by-level basis as much as it is
true for the game as a whole.

Success Strategies
From a player’s point of view, a video game needs to be worth playing. The ac-
tions a player takes while in the game need to include gameplay that is some-
how rewarding to the player, especially if linked to progression within the game.
There are strategies available to the player that result in an enjoyable progres-
sion. Part of the fun of a video game is finding out what they are and perfecting
them.
From the player’s point of view, this is central to being taught how to enjoy
the game. To players, a game is an entertainment device over which they have
control. Learning how to manipulate this device in order to yield maximum fun
is imperative to good level design. To learn this, a player has to determine which
strategies of play yield the best results. A success strategy in that context is defined
as a strategy that produces a fun gameplay experience.
These success strategies can be formal in nature, requiring predescribed so-
lutions to gameplay questions. For example, this occurs when a player has to
progress through a level by following a specific path, determined by the level
designer.
Alternatively, success strategies can be informal, defined by the players
themselves. Good examples of this can be found in games where players can
devise their own enjoyable gameplay through nonprescribed interaction with
the game’s environment. Although less explicit than formal strategies, this kind
of freeform play can still be encouraged through good level design.
Teaching Mechanisms 31

Some Dos and Don’ts

Without trying to be exhaustive,13 I would like to highlight some typical dos and
don’ts related to this subject. They are not hard rules; use your own judgment,
but they generally should be considered.

Dos
Let’s start with some recommended approaches.

Teach by practical example

Always give players a chance to put the things they have been taught into prac-
tice as soon as possible. When a new gameplay mechanic is introduced to the
player, it is best to let the player try it out immediately. The best way of learning
for many people is by doing. And it is easier to put lessons into practice when
they are still fresh in the mind of the player.
This is a good habit to get into, and once you are aware of this method, you
will start to notice the principle in many other games. For example, many Nin-
tendo games do this consistently.14

Positive reinforcement
If at all possible, make sure players are actively rewarded when they pass a skill-
test or successfully progress through a challenging gameplay scenario.
When players are taught that there are very positive consequences for suc-
cessfully navigating the game, they will become eager to engage with it.

Prepare the player fairly

Make sure that players have the right tools and knowledge available to them be-
fore they are forced to perform a skill test that they can badly fail. There is a place
for lethal encounters in many games, but it is always necessary to make them fair.
Unannounced or unavoidable instant-death traps are generally to be avoided,
unless they are expected in the game’s genre.15
When possible, teach important skills in a safe environment first. A great
example of this philosophy is found in the Half Life 2 series:
13
Elements of this chapter will keep appearing in some form or another through the
book, and additional tips and danger areas will be highlighted per subject.
14
Zelda: Twilight Princess, to name one.
15
This will be covered in more detail later in the book, specifically in the chapters on
reward mechanisms and challenge.
32 Introduction to Level Design

Players discover that every basic skill is taught in a very unobtrusive

way, especially early in the game. This includes simple challenges, like
stacking the crates to get out of the windows in the Trainstation, to
more complex ones, such as the Antlion Pheropod “Bugbait” training
in the Coast section. The difficulty of subsequent tasks could then be
increased knowing that the player had been taught the mechanics they
needed to succeed.16
Make sure that players can comfortably learn the skills they are supposed
to pick up. There is no problem with ramping up difficulty later on and provid-
ing more difficult encounters, but when first teaching the player how to play the
game, the lessons should be forgiving.
Players should be introduced to new mechanics in non-frustrating ways.
This can mean that a skill can be taught in such a way that players cannot fail the
exercise and only need to focus on practicing the skill until they can perform the
tasks relevant to the teaching exercise.

Don’ts
We also need to be careful to avoid some problems. The following examples are
situations to look out for:

Don’t start with failure

Sometimes level designers are tempted to start a new level with a very serious
challenge. They want to begin gameplay on a tense note and immediately put the
player in grave peril. This may sound good on paper, but a game is not a book or
a script. If there is serious danger, there is a serious chance that the player will fail
the gameplay challenge. Imagine how frustrating it is for those players who don’t
pass this test and end up failing badly right at the beginning of a level. Rather
than being enjoyably tense, the experience is likely to be off-putting and tell the
player that the game will be a frustrating affair. In fact, it discourages the layer
from playing, and that is something we generally want to avoid.

Don’t taunt the player

It is strange that I even feel the need to write this, but this advice is too often
ignored. It is crucial that even when players are struggling to pick up a skill, that
they be encouraged in a positive manner, and not scolded for not being able yet
to master a mechanic or meet a gameplay challenge.

16
David Hodgson, Half Life 2: Raising the Bar, Valve/Prima Games, Roseville, CA, 2004,
p. 277.
Teaching Mechanisms 33

When a gameplay section is supposed to be about teaching, play mechanics

make sure the results are positive, even if the player takes a long time to master
the skill. Don’t shut off award paths or use taunting language or employ any simi-
lar tactics, as you will just put off those players who for some reason or another
struggle a bit more than others. You will end up punishing those payers who
have the most to gain from an encouraging approach. If that happens, you may
well lose them completely if they get fed up with the game and put it down for
good.

Theory Summary
There is a direct and useful link between level design and the educational aspects
of gameplay. Level design can be seen as a vehicle to teach the player among
other things, how to play and enjoy the game. Gameplay, and by extension level
design, functions as a teaching mechanism. The techniques and strategies that
can be derived from this conclusion cover almost every aspect of gameplay.
There are too many ways to employ this knowledge to focus on individually.
This chapter has identified a number of important ones, and future chapters will
explore even more.

Practice

Example 2.1: Teaching by Doing—

Mandatory Skill Gates
Summary
Generally the most enjoyable way of learning a skill is by “doing”: the player ac-
tually trying to perform the necessary actions, and adjusting and adapting until
he is able to put the lesson into practice. Because levels often make heavy use of
interactivity and player agency17 to a large degree, this method is very suited to
the form.
There are times when a level designer needs to be sure that the player pos-
sesses certain skills or is in possession of specific knowledge. A guaranteed way

17
The ability to act in the world.
34 Introduction to Level Design

of teaching the player something like this is by making absorbance of the lesson
a condition for progress.
The two aspects can be combined in a discreet level design scenario that is
very useful in many circumstances.

Game Genre
The technique is suitable for any game where progress can be halted if a game
mechanic is not yet mastered (either naturally, or by scripted means).

Goals to Achieve
• Showcase a natural way for the player to pick up skills through gameplay
actions in level design scenarios.
• Make sure the player cannot progress unless they have been taught how to
use the new skill.
• Teach new skills in a controlled setting.

Description
(Example type: Original/general)
If a level designer wants to make sure that players pick up a certain level of profi-
ciency when introduced to a new gameplay mechanic, he or she can consider the
use of a skill gate in a locked-off gameplay arena.
Such a setup is realized by establishing a dead-end area where the player
finds or is introduced to a new gameplay mechanic. This can be a new skill or
perhaps a new item that is to be used throughout the game from then on. Imag-
ine a player needing to jump down one-way drops (too high to jump back up)
all the way to the bottom of a ravine. Once at the bottom, the player is trapped
there unless they can find a way to backtrack along the path they just used. If
the player is then confronted with a new skill—for example, an increased ath-
letic ability—he can now practice that skill by using the new athletic ability to
travel back along his original route. This will only be possible if the player be-
comes proficient enough with the new athletic capabilities to be able to scale the
heights that were prohibitive before.
The designer can include extra difficult areas off the main path, filled with
rare collectables, to encourage the player to exercise and practice the new skills
even further. By the time the player has left the ravine, she will have learned the
new skill—or would not have been able to escape—and have been given the
opportunity to find extra rewards through extra efforts, showcasing that the new
skill is enjoyable and useful.
Teaching Mechanisms 35

Further Notes
This technique is extensively used in Nintendo’s “Zelda” games. Upon the intro-
duction of a new skill, the player is typically confronted with a use for that skill
close by.
There are many more ways that these skill gates can be part of a teaching
mechanism. A puzzle may need to be solved before the player receives an item
that unlocks the next area of the game. A creature needs to be defeated by using
a specific gameplay mechanic to its full potential. A locked dungeon may have to
be cleared of enemy creatures before it unlocks. Game history is full of further
examples that can be adopted for new level designs.
A further advantage lies in the fact that, since the player is guaranteed to
have learned the skill at the end of the gameplay scenario, the level design can
now take this into account.

Example 2.2: Teaching by Example

Summary
By showing players certain actions or outcomes in the game world, you give
them the opportunity to learn by observation and example. This principle is eas-
ily translated in custom-made gameplay lessons that are effective and appropri-
ate to the level setting.

Game Genre
The technique is suitable for all games that allow the level designer to create an
observable sequence of gameplay-related actions.

Goals to Achieve
• Introduce a new enemy into the level, at a safe distance from the player.
• Show conditions in which it becomes a threat.
• Demonstrate the severity and nature of the threat.
• Show to the player what tactics work against the enemy.
• Give the player a chance to practice this in a safe setting.

Description
(Example type: Original)
A classic scenario in level design is found in the introduction of a new enemy in
the game. If this encounter is of sufficient importance and the level designer has
36 Introduction to Level Design

time to turn the encounter into a mini set piece, it can be turned into a scripted
teaching mechanism. By showing how the creature reacts or how the creature
can be defeated through the actions of others, the player can learn and strategize
without being directly exposed to any immediate danger.
The trick is to find a natural way for the player to be separated physically
from the action but still be able to observe a scenario played out in his presence.
This can be done simply by providing a distance barrier (the player can’t get
close enough to the action before it runs its course) or a physical barrier that
still allows the player to see the action unfold. The example may play out on a
balcony the player can’t reach, or some similar restraint.
Let’s take, for example, an adventure game where the player is part of an
archaeological dig on the side of a mountain. The other members of the party
are human, as is the player character. When the player reaches a certain position
near some loose rocks an event is triggered. One of the party members slips on
the rocks and falls down onto a precarious ledge, which houses a big bird’s nest.
The nest belongs to a condor pair, and the mother can be heard squawking from
far away. The sound becomes louder, however, until the irate bird flies into view
and starts attacking the fallen archaeologist. It does so by either flying over the
NPCs head, trying to scratch him with its claws or by flapping its wings in front
of him trying to push him off the edge. The NPC takes initial damage, but even-
tually notices that crawling protects him from the flyover attacks, while throwing
rocks at the bird interrupts the flapping attack. Other NPCs get the clue and also
start throwing rocks at the bird, but are not able to hit it. The player, on the other
hand, is given the opportunity to do the same, and when he or she successfully
strikes the bird with a rock the bird is scared off. The NPC is subsequently res-
cued by his colleagues.
The player has learned a number of things:
• the existence of these condors in the level;
• the fact that they become aggressive if their nest is disturbed;
• aspects of their behavior: time to arrive on the scene, different attack modes;
• the amount of damage they can do;
• several strategies for coping with them;
• how to dispatch them.
All in all, that is a decent amount of gain from one custom-made lesson.
From now on, the level designer should be able to use the condor threat in sev-
eral circumstances. The player may be confronted with a path that is blocked by
a nest, or the player may be accosted by a pair of condors to up the challenge.
Many other scenarios can be imagined, but the player will at least be familiar
with key aspects of the creature.
Teaching Mechanisms 37

Further Notes
It is vital when employing this technique that the player is subject to the same
rules as those entities around them. To give a simple example, it would be unfair
to show a non-player character who is physically the same as the player character
to be more resistant to environmental damage than the player. So, for example,
if an NPC of the same abilities jumps off a high roof and survives, the same out-
come should apply to the player (unless there is a good explanation for a differ-
ent outcome).

Example 2.3: Formal Tuition—Overt and

Covert
Summary
Sometimes it is valid to teach a player something through a tutorial or training
sequence. The player is aware that he or she is being taught something about the
game and needs to pay close attention because of this. This example shows a way
to do this that does not break the game’s immersion.

Game Genre
This technique is especially suitable to games that require the player’s suspen-
sion of disbelief to stay intact.

Goals to Achieve
• Teach the player game related skills.
• Do this in a formal manner; the player knows they are being taught some-
thing.
• Show overt and covert ways of doing this.

Description
(Example type: Existing game)
Many games require at one point or another that the player go through a tutorial
in order to learn a new ability. This can, however, be done both in overt and covert
ways. Overt in this context means that the tuition is not hidden within the game
world. The player is literally told that they are being taught without pretending it
is part of the game proper. This is what happens with tutorial sequences outside
the levels of the game, a bit like an interactive manual.
38 Introduction to Level Design

Covert, on the other hand, means that the teaching occurs within the story
or within the reality and logic of the game world. The game stays “in character”
and does not break the fourth wall. This is a very useful technique as it has a
number of valuable side effects beyond the content of the tutorial. This kind of
teaching mechanism can achieve the following goals:
• teach the player a skill (or range of skills),
• maintain immersion,
• maintain suspension of disbelief,
• add to the level content.
Halo—Covert camera calibration tutorial. In Bungie’s Halo: Combat
Evolved, 18 we can see this technique utilized with great intelligence. The game is
played with a first person camera; preferences with regard to camera calibration
is one of the peculiarities of players: If you press the camera stick on the joy-
pad up, should the in-game camera look up or down? As it turns out, Microsoft
has studied this subject through extensive usability testing, and they found that
there is a 50-50 split of preferences on this subject. This meant that whatever the
default setting for the game camera was, it was going to be wrong and frustrating
to half the players. They decided, therefore, to let the player calibrate the cam-
era themselves in-game, in a covert camera calibration tutorial. This was a very
clever and useful solution to a problem that should not be underestimated.
The resulting level design solution was to place the player character in a set-
ting where the game could measure the player’s input when asked to perform a
task. In this case, the player is given a new combat suit and asked to “test” it for
optimal performance. The player is literally asked to look left and right, up and
down, and the resulting player choices tells the game what the player’s prefer-
ence is in this regard.
This solution shows that it is possible to maintain suspension of disbelief
even when teaching or calibrating very technical aspects of gameplay.

Further Notes
Other classic covert tutorial or teaching scenarios occur in games that use an
obstacle course for in-game training of operatives. This occurs, for example, in
the first Splinter Cell19 game, where the player is asked to finish an obstacle course
to assess if the player has received enough training.

18
Published by Microsoft Game Studios, developed by Bungie Studios, released Novem-
ber 15, 2001.
19
Published by Ubi Soft Entertainment Software, developed by Ubi Soft Divertisse-
ments Inc., released November 2002.
Teaching Mechanisms 39

Example 2.4: Teaching through Experiment

Summary
A very pure and natural way to discover abilities and other things is by experi-
menting with things. Playful experimentation is a cornerstone of education, not
only for children but for adults as well. (“What does this button do?”)
This method can be used to much effect as it tends to produce very enjoy-
able results.

Game Genre
The technique is very suitable for sandbox games. Additionally, it is a very useful
technique when there are multiple uses for a game object.

Goals to Achieve
• Create a situation that allows for self-taught gameplay skills.
• Create an environment that encourages experimentation.

Description
(Example type: Original)
There is a funny cliché associated with 2D point-and-click adventures that you
often need to combine the use of completely disparate objects in your inventory
to find a solution to some obscure problem—the (valid) complaint being that
there is no logical sense to many of these item combinations. Those games do
highlight, however, that it is fun and rewarding to find uses for things through
experimentation.
This principle can be expanded on in level design by making sure that there
are instances where the player can freely experiment within the interactive pa-
rameters of a level to learn new skills, or find new uses for objects.
To try to encourage this kind of experimental yet educational gameplay, you
may find it worthwhile to create specific areas where the player can indulge—a
“safe zone” with no loss of, say, expendable items.
A good way of doing this is by creating a situation where there are multiple
uses and outcomes built into the level design scenario from the outset. To do
that, we take a number of desired outcomes and characteristics and use them as
the building blocks for our level design scenario. Taking, say, the humble game
crate20 as a starting point for an example, we can list a number of ways in which
this can work. We can say, for example, that the crates can
20
Discussed in Chapter 8.
40 Introduction to Level Design

• block NPCs,
• be stood on,
• be climbed,
• be stacked on top of each other,
• be pushed over,
• be moved about.
If we take these basic abilities of the crate, add a player character’s abilities to the
mix, and combine those two with an environment containing aggressive NCPS,
collectibles that are out of reach and additional crates that the player can’t get to,
we end up with a recipe for great potential experimental fun.
The player could try to do things like
• create a stack of crates that can be climbed to reach high areas (to get to
collectibles),
• create a stack of crates to reach other crates to add to the mix,
• prepare traps where the crates can be shoved off ledges onto enemies,
• create towers of crates to be pushed over on top of enemies,
• create pens in which enemies can be trapped,
• create structures that are pleasing to the eye,
• stand on crates in order to review the environment form a high vantage
point.
Many other uses can be found depending on the level in question.
Look at all the things the player has the potential to learn—many ways to
find new objects, interact with enemies, manipulate the environment—and all
without any explicit tutorials. Anything the player learns will feel like something
they have earned through intelligent gameplay and as such is very rewarding to
the player.

Further Notes
There exists the danger that this technique does not aid the player in finding the
right solution. It is therefore best implemented in situations where the teaching
mechanic is optional—i.e., one very specific solution is not required to complete
the challenge.
Level Design Goals and
Hierarchies 3

J ust like in any other creative endeavor, it is important in level design to define
a set of clearly understood goals before any major work is done. However, this
is easier said than done, if there is no consensus on just what those goals may be.
Can we even speak of goals specific to level design, or are such goals related to
the game design as a whole? Do level designs for differing games share similar
goals? Should we even try to formalize a set of specific level design goals?
The longer we look at these questions, the clearer it becomes that they are
important and fundamental questions that need to be addressed early in the lev-
el design process. This chapter will address a number of level design goals that
are important to keep in mind.

Success Definition
What are you trying to achieve?
This is a question that should always have an answer. The why and how aspects
of the same question are addressed separately, but those are all moot if the what
question is not answered first. Being able to define clearly what it is you are trying
to achieve means that you can clearly describe a number of goals that need to be
met. That is to say that any goals that are set are subject to an overall success defini-
tion. This means that the goals are not arbitrary, but instead are set to achieve a
desired overall result, which is a definition of success. A goal is just part of a strat-
egy for achieving this successful outcome. This may seem too obvious to state,
but in my experience this most basic element of level design is often forgotten.
The success definition(s) can be game-wide1; indeed a clear definition
should exist at the top of a game’s hierarchy of goals. It can be level-wide, or can
1
Some game companies refer to such a game-wide goal summary as The X.
41
42 Introduction to Level Design

even occur on smaller gameplay scales. For example, a hypothetical level-wide

success definition could state:
“Train players’ combat skills throughout the level and build up the players’
level of confidence and ability in order to prepare them for the introduction of a
new enemy creature.”
A game-wide success definition might read:
“Create an action-horror hybrid that can be described as “Dracula meets
The Terminator.2”
In our case the success definition describes what the level designer needs
to achieve. This is something that should remain in focus as the ultimate goal
throughout the level design and level creation process. All other goals are by def-
inition subordinate to that overall aim. Whatever the final success definition(s)
incorporate, we need to be aware of what they are.

External Level Design Goals

There are all kinds of goals that may need to be taken into consideration when
determining success definitions. Some are independent from direct gameplay
considerations, insofar as they that they refer to goals outside of the gameplay
experience. The level design may have to
• be appealing to both genders,
• stay true to the brand image,
• sell X amount of units,
• push new engine technology,
• win a BAFTA award,
• receive a meta critic3 rating higher than 75%.
These are the overarching goals that are derived from the requirements of
the game’s development as a commercial or even artistic production. They de-
scribe the purpose of the level design components as much as they define the in-
ternal level design goals, which are derived from direct gameplay considerations.
It is a mistake to dismiss these external factors, as they are the ultimate reason
why a game is made in the first place.
Good examples of external goals like this can be found in the marketplace,
so let’s take a closer look at commercial considerations:

2
I have seen worse pitches.
3
See www.metacritic.com.
Level Design Goals and Hierarchies 43

Commercial considerations
It is important to note that, aside from alternative funding models and indepen-
dent game development, in most cases, games are supposed to make money.
There are many ways one can try to achieve this goal, but it should not be forgot-
ten that for many people it is the fundamental reason for developing a game in
the first place. It is wise to note this simple fact. In the case of publisher-funded
development, for example, a game that does not sell is likely to be seen as a failure
by the financing publisher. The most important reason that publishers fund game
development is because they think it is likely to make them money. I am not saying
this is bad or good, but merely stating a fact that offers another point of view on
the function of a game and, by extension, of its design. If the game design within
this context doesn’t support a commercially viable end product, it has failed to
fulfill its basic function from the publisher’s perspective. This is a possible interpre-
tation of game design that, even if unwelcome4 at times, nonetheless needs to be
understood.
Only after a game’s creation has been approved through these external goals
can the production go ahead. However, these goals do not in themselves deter-
mine fun gameplay, which is the realm of internal level design goals.

Internal Level Design Goals

As stated earlier, internal level design goals are derived from direct gameplay con-
siderations. Outside of the external macro goals described earlier, there will al-
ways be level design goals that are just part of the art form itself, driven by inter-
nal factors. They are the goals that contribute to the overall good gameplay. They
deal with requirements such as the following:
• Empower the player.
• Teach the player how to have fun with the game.
• Don’t break the player’s suspension of disbelief.
• Give the player a sense of achievement.
• Reward the player for exploration.
• Provide addictive, fun gameplay.
These are requirements that stem from supporting the game or the game-
play on a very pure level. They are as important to define as external factors.
It is important to never forget that it is the job of a level designer to imple-
ment, interpret, and exploit the game design to give it its best possible applica-
4
It is good to note that commercial considerations do not have to be crass or in conflict
with artistic merit.
44 Introduction to Level Design

tion. A game’s level design is not an isolated construct, just as a game design does
not function independently. They need to complement each other, or both will
suffer. (See Chapter 1, “Game Design vs. Level Design.”)
Internal level design goals are the bread and butter of our profession. They
give us very specific goals to aim for in order to provide a good gameplay ex-
perience. In a hierarchy of internal goals, the game’s overall gameplay experience
always comes out on top, and the level design needs to support this.
This also means we have to start putting our level design goals into some
type of hierarchical order.

Level Design Hierarchies

We looked at level design goals before and determined that they must support
and put into practice the game design. This is easier said than done, of course,
since the level design components will have to fulfill a number of requirements.
This is their purpose. Nonetheless, let’s assume that we have a clearly defined set
of goals, and that we have translated them into requirements.
So how do we know what value to assign to these goals and requirements?
What components do we focus on first when we start working on a new level?
Which elements are more important? What if it doesn’t work?
These are all typical questions that come up during a game’s development.
They are all related to the component’s importance and place in the project. To
determine these factors, we first have to take a step back and devise a way to as-
sign value to individual level-design components. We have to give them a place
within a logical level-design hierarchy.

Development Hierarchy
To create a situation where we can measure the value and implementation of
such things as AI encounters, item placement, puzzle design, and hundreds of
other level design components, it is helpful to define some kind of hierarchy. If
we can figure out where things fit in the scheme of things, we can make choices
on what and how to prioritize, and where dependencies occur.
In a game-design setting, here are a number of known stages and areas of
development. They tend to fit into a structure where one area supports another
in a certain hierarchical order.
Generally, we have to look at the following questions:
• Why are we making the game?
• What game are we making?
• How are we making it?
Level Design Goals and Hierarchies 45

Figure 3.1. Game-wide hierarchy.

When we have answered these questions and have logically placed our gen-
eral development components in this framework, we can see a hierarchy emerge.
For some games, this general hierarchy may look like the image in Figure 3.1.
In this pyramid shape, each segment is supported by the segments below it,
and were we to try to answer our why, how, and what questions with it, the fol-
lowing would emerge:
• Success Definition (Why),
• External Goals (Why),
• Internal Goals (Why),
• Game Design (What) – Level Design (What/How),
• Technical Design (How) – Art Direction – Audio Plan (How).
This diagram gives us a possible insight on where level design exists within
the overall context of the game’s development. I won’t go into too much detail
when it comes to production issues, as that is not the focus of this book. How-
ever, it is at least interesting, insofar as that it shows some relationships between
level design and the rest of the game’s production areas.
Things become more relevant when we try to find a similar approach within
the level design parameters for a game. We might ask those questions again:
• Why are we making the level?
• What levels are we making?
• How are we making them?
46 Introduction to Level Design

Figure 3.2. Level-wide hierarchies.

To answer those questions, a game’s level design hierarchy might be made

up of the elements shown in Figure 3.2.
And to repeat the exercise once more: in this pyramid shape, each segment
is supported by the segments below it, and were we to try to answer our why,
how, and what questions with it, the following would emerge:
• External Level Design Goals (Why),
• Internal level Design Goals (Why),
• Structure (What),
• Level Design (What),
• Per-Level Goals (How),
• Level Design Scenarios (How),
• Gameplay Moments (How),
• Gameplay Actions (How).
From this point on, we can go into real detail and start to assign things like func-
tion and resulting tasks to the broken-down elements.
Level Design Goals and Hierarchies 47

Dyson, a Case Study

To illustrate some of these concepts against a real world example, I am going to use the
game Dyson,5 designed by Alex May and myself, as a case study. Dyson was created in
a one-month competition for the independent games focused website and community
TIGSource. The game development competition was defined by the following rules:
The theme of the game is PROCEDURAL GENERATION, with an em-
phasis on generating new content every time the player starts the game.
Procedural generation, or procedural content generation, is a term to
describe the creation of content algorithmically “on the fly” (i.e. after
the software executes, rather than before). In video games it is generally
used to either generate “random” content (like graphics or levels), or to
reduce file size. This competition is focused on the former, but you may
choose to apply the theme however you see fit.6
With this background in mind, we can now examine the why, what and how ques-
tions and the level design hierarchy they imply. In the case of Dyson, all these ele-
ments are represented, as discussed below.

External level design goals

Dyson’s levels first and foremost had to show off the concepts of procedural gen-
eration, as defined by the competition rules. They had to do so in such a way that
they still ended up being enjoyable to play, without the procedural nature being
a gimmick. In other words, the procedural elements of the game needed to be
explicit in the levels and the gameplay they would hold.
All this had to be achieved within 30 days with a very limited amount of
time and resources. (The team consisted of only one programmer, one designer,
and one audio designer.)

Resulting strategy. To meet these goals, we decided that Dyson would be a game
that Incorporated scientific theories that show off procedural generation in a
gameplay format. Freeman Dyson’s Dyson trees and Astrochicken met John Von
Neumann’s self propagating machines, and everything was turned into a design
that allowed small seeds of data to grow into complex environments.

5
This is not meant to promote the game, but simply to provide a clear example of the
concepts we have just discussed. A fully featured version of the game is still being de-
veloped at the time of writing this book by Alex May and myself, with music and audio
being done by Brian Grainger. Have a look at it here: http://www.dyson-game.com.
6
“Procedural Generation Competition,” Procedural Generation Competition,
http://forums.tigsource.com/index.php?topic=1646.0, 2008.
48 Introduction to Level Design

To restate that in understandable English: Dyson was chosen to become a

game of space conquest. Flying robot seeds conquer territory and exploit re-
sources in an asteroid belt. Other seedling factions compete for the same re-
sources, and almost all the game art was to be procedurally generated, rather
than modeled.

Internal level design goals

To meet the external goals, we decided that the levels had to provide strategi-
cally interesting choices despite a very limited amount of resources at hand. This
meant that the levels needed to work as a replayable experience, where players
would get different content every time they play.

Resulting strategy. To provide enough variation within a limited number of lev-

els, we decided that almost all factors in the game had to be affected by a select
number of variables. To do this, a limited number of variables had to lead to a
maximum sense of interesting complexity.
To facilitate this in level design terms, we decided that the levels had to work
in two dimensions, and that all territory was governed by simple and recogniz-
able attributes. In the game, the seeds are planted on asteroids, and they turn into
trees whose genetic makeup is influenced by the asteroid’s variations in natural
resources. The trees produce seedlings, which are also affected in their abilities
by the mother asteroid, and they can in turn be used to conquer other asteroids.

Structure
Because of the need to make the most out of very limited resources, the lev-
els for Dyson had to be replayable. The level structure therefore had to support
open-ended strategic choices from the beginning of play, in order for multiple
playthroughs to feel very different, depending on the choices that were made by
the player early on.

Resulting strategy. To achieve this, we decided to create one very large level, with
interesting gameplay complexities introduced by AI agents that all vie for the
same resources as the player. Because the resources and layout of the asteroids
change every time the game starts, replayability is guaranteed.

Per-level goals
In this case, only one level was chosen to be in the game, but it had to have the
appeal and scope to keep the player interested for as many playthroughs as pos-
sible.
Level Design Goals and Hierarchies 49

Resulting strategy. For this to work, the gameplay goal of the one level we ended
up with for the competition had to be one of total conquest of a territory. The
win/lose conditions for a level had to be simple, so that players could easily un-
derstand the task at hand and try again if it didn’t work out.

Gameplay scenarios
The level in question could only maintain its long-term appeal if a number
of discreetly interesting gameplay scenarios would play out time after time,
without losing appeal. The level had to be balanced in such a way that these
kinds of gameplay loops would occur often enough and be enticing through-
out.
They had to be those kinds of moments that operate at a level where the
player is trying to find and perfect success strategies.

Resulting strategy. Out of the possible scenarios to evolve from the core game-
play elements, we focused our attention mainly on the following ones:
• Invest and grow: Choose to keep seedlings close by and use them to plan
more Dyson trees).
• Grow and expand: Try to create an army of conquest and take over other
territory.
• Choose units and attack enemy: Focus on the variations in seedling
abilities and pick seedlings best suited to defeat other seedlings in the
game.
• Defend: Take a defensive approach and try to hold territory while repelling
attacks from enemy factions.

Gameplay moments
To have lasting appeal, gameplay scenarios need moments of excitement or must
otherwise be engaging. These are the decision moments that involve players at a
deep level and have them watch the game unfold with real interest.

Resulting strategy. We tried to make sure that enough of these special moments
would occur to serve our other goals in the level-design hierarchy. We focused
on gameplay moments that included
• choosing a strategically strong asteroid to attack,
• finding the right moment to sacrifice seedlings and plant a new tree,
• setting up forces, to swoop to a heavily fortified enemy asteroid,
and so on.
50 Introduction to Level Design

Gameplay actions
Finally, the immediate actions the player can perform had to be balanced against
the overarching goals and had to fit the environment. Fundamental level-design
choices had to be made with basic actions like
• planting a tree,
• sending units to a target.

Resulting strategy. To keep things interesting for players, a number of restric-

tions, or conversely abilities, had to be considered with regard to those gameplay
actions. For example, planting a tree had to be balanced by an associated cost; in
this case a number of seedlings had to be sacrificed. (This had the added advan-
tage of making seedlings work as currency). On the other hand, sending units to a
target was made to be as easy as possible; all units’ seedlings orbiting an asteroid
could be sent off simultaneously.
To facilitate the use of these actions, and to serve several of the other goals
of the game, we also decided to include tutorial text that helps the player pick up
the game rapidly.
Note that it is surprising how easy it is to fit all the disparate elements of a
game and its level designs in a fairly sensible and logical hierarchy. The fit is not
100%, and some elements have been left untouched. For example, music and
audio play an important role. However, the example illustrates how it is possible
to construct a view of the level design process where each individual feature is
shown to support many others. The interrelationship within such a hierarchy
is useful and shows up important areas to focus on when designing and imple-
menting levels.

X Factor
Another way of looking at the concept of a success definition is by dubbing it
the X-Factor. I am highlighting this subject, as it often doesn’t get the attention
it deserves.
A good game is more than the sum of its parts. This can be seen very clearly in
some very famous example, for instance in Grand Theft Auto: Vice City,7 which
was a phenomenal success, both financially and creatively. However, if studied
in isolation, many of the individual game components were lacking in some way
or another.

7
Developed by Rockstar North.
Level Design Goals and Hierarchies 51

• The aiming mechanic was cumbersome.

• The third-person character controls were frustrating.
• The camera often couldn’t cope.
• The geometry and architecture were very low on detail.
There are many other examples as well, but the key point is that deficien-
cies or excellence in these areas do not necessarily make a bad or good gaming
experience.
Many people have tried to copy Vice City and have even improved upon
individual gameplay mechanics, yet have come away with less enjoyable games.
Apparently it is not that easy to copy gameplay mechanics and end up with some-
thing of equal quality. The answer lies in the fact that there is an art and a science
to games development, and that both have a strong impact on the final game. (I
am using the term in this context as something different from the art direction).
The science is the pure implementation of the required mechanics, but the art
comes into play by taking those gameplay mechanics and making them work as
effectively as possible within the game’s success definition (including the game’s
level design goals).
To go back to GTA: what the game does very well, however, is to place all
these individually lacking gameplay elements and put them in the service of the
game’s ultimate goals, both internal and external. Players forgive the frustrating
moments, the repetitive NPC dialogue, the camera issues, and so on, because
they are still successful in achieving those goals that the level designers and game
designers have put at the top of the hierarchy.
Additionally, what gives Vice City added appeal is how all its disparate ele-
ments have been made interesting by giving them cultural relevance. The music
in the game, the references to 80s pop culture, and the humor, can all be seen as
essential contributions to the final appeal of the game. It is also interesting to see
that they are all integrated in the level design of the game in a way that is deep
and practical.
In practical terms, the game successfully combines the art and science of
making games. And by applying all these things expertly in its level design, mis-
sion-based or sandbox, it manages to fulfill its game-wide and level-wide success
definitions and hit the target of the elusive x-factor that so many games strive
for.

A Modular Approach to Level Design

So now we know that is useful to try to place our level design components within
the context of an overall level design hierarchy. A good next step is to make sure
52 Introduction to Level Design

we can have level design components (modules) that are multi-purposeful and
effective.
The modular aspect of this approach comes from the idea that once you de-
termine what the function of a component is, you have more freedom in the way
you try to hit the requirements that this functionality implies.
A classic problem with inexperienced level designers is that they have insuf-
ficient grasp of the overall purpose of the particular level design task they are
working on. It is easy to get trapped in problems due to tunnel vision, I cer-
tainly have done so myself on many occasions early on in my career. When this
happens, the micro details of a particular level design component become more
important than its actual function in the grand scheme of things. Suddenly, that
ambush encounter you are working on HAS TO BE about using a particular
weapon or enemy, while it may be entirely possible that that encounter can fulfill
all its requirements while being staged against a completely different backdrop
or by using an entirely different class of enemies.
Realizing that this danger exists and always asking oneself, “What is the
function of this component?” goes a long way towards avoiding that trap.

Contingency Planning
Not only does this way of thinking help in designing levels in a more effective
manner; it also provides great scope for contingency planning. This is something
that within the high-pressure environment of commercial game development is
extremely important.
It allows us to create level design components that are scalable. For example,
if we know that a certain combat situation in a level is mainly included in or-
der to introduce a new gameplay mechanic to the player, we can of course try
to make the event as spectacular as possible by introducing complex scripted
events, incorporating intricate enemy behavior and pulling out all the bells and
whistles. This section will probably “read” very well on paper, especially early on
in the development cycle, when the level designer is assured that the creature
AI would eventually be more than up to the task of supporting such an amazing
encounter.
Now let’s fast-forward twelve months in time. We are now two months away
from having to ship the game, everybody is severely stressed and overworked,
and unfortunately, the AI of this particular enemy had to be scaled down several
levels due to time constraints and will no longer be usable within the original
plan for the level design. Oh dear, what now..?
This is an example where it is extremely important to realize why the encoun-
ter exists in the first place: to introduce a particular new gameplay mechanic. With
Level Design Goals and Hierarchies 53

this overall goal in mind, we find that it really does not matter through which AI
encounter this gameplay mechanic is taught and introduced to the player. It can
be substituted and made to work by using existing and working game assets, and
the final value to the game as a whole will be the roughly the same.
This may seem completely obvious, but it is very easy to get lost in insular
goals.
Anybody in games development is constantly bombarded with the follow-
ing question: “Wouldn’t it be cool if we… (Insert idea X)?”
As it turns out, idea X may well be cool, but does it really need to be there?
Can it be done in a different and more cost-effective way?
Those are the kind of questions you should ask yourself if you want to be
able to apply level designs with enough flexibility that you can react to unex-
pected circumstances.8

Application of Modular Hierarchy

If we take all that we have said before into account, we should now have a practi-
cal framework for our criteria and requirements that has great value. We should
be able to take any game component, find a place for it in this hierarchy, and be
able to ask the right questions regarding its place and value in development. This
is no minor thing and can save a difficult project from disaster if done correctly.
It means that there is much less dependency on specialist or bespoke solutions,
and that the developers can be much more flexible about the implementation of
the designs without compromising too much on quality.

Conclusion
Hierarchies can offer a very useful tool. They give us a way to break level designs
down into individual components and formulate a plan in which all of the level
design components fulfill their ultimate requirements. Furthermore, it allows us
to check the value and importance of individual components and refer to their
place in an overall hierarchy.
Viewing the level design process through hierarchical steps may not always
be a perfect fit. For one thing, one level design hierarchy may look different than
another, and your mileage may vary between different projects. But it is always
useful to try to find a way to define the why, how, and what questions within a
logical order or hierarchy that allows you to make judgment calls on the imple-
mentation of your levels.
8
You should always expect unexpected circumstances.
54 Introduction to Level Design

This gives us the practical means to assess what function any individual
component or element fulfills, what it relates to, and how important it is. This
has obvious uses that go beyond the need of the level designer; it also helps in
establishing clear goals and criteria that are understandable to anybody involved
in the development of the game.
Level Design Structure and
Methodology 4

A level’s structure and methodology are important early choices for a level
designer, and they can have a large impact on the actual creation of the
levels. We will look consecutively at structure and methodology, because the
choices of the former impact the likely choices for the latter.

Structure
Once we have decided on the type of content a level requires, and where it ex-
ists within the hierarchy of the game, we need to determine which structure to
apply to the level. In most cases this will be determined as part of the game
design, since that determines structure as a whole across the game. Nonethe-
less, it is often the case that at least some choices are left to the level designer, at
least on a smaller scale—for example, within the levels themselves. This kind of
structure is one of gameplay flow, which is very much in the hands of the level
designer.
Typically there are three main approaches to choose from in most game
types: linear, semi-linear and non-linear. Sometimes the distinction is not
clear-cut, and hybrids may occur. For example, a level may be divided be-
tween content that is 60% linear and 40% non-linear. Ultimately one of the
most important determining factors of structure is a game’s genre. A classic
shoot-em-up is much more likely to follow a linear structure than a freeform
RPG. But even within these genres, there is scope for differing progression
models.
Let’s look at the distinction between linear, semi-linear and non-linear in
more detail.
55
56 Introduction to Level Design

Linear Levels
Linear level design, as the name implies, is level design where the gameplay events
follow a strict line laid out for the player to follow.1 Sometimes this is referred to
by saying that the gameplay is on rails. Events unfold across this line in a strict
order that the player cannot deviate from. Progression through the level is only
possible if the player goes through the gameplay events in the order predeter-
mined by the designer. To look at a diagram of such a level structure, we can
envision something like the structure in Figure 4.1.
Although pure linearity is becoming less prevalent than it used to be, it still
has a place in video games. There are a number of advantages to the technique.
If they are appropriate to the gameplay needed for the game design, linear levels
can work well.

Figure 4.1. Linear level structure.

Advantages
The most important advantage to linear level design is the amount of control it
gives a level designer over the play experience. It is much easier to carefully de-
sign the experience of playing the level if the designer can determine the order
in which gameplay events occur. Since this type of overall control allows the
level designer to determine matters of pacing , consistency, story development,
learning curve, and many others, it is as close as a level designer can get to directo-
rial control. And in keeping with the analogy between directing a film, it can be
compared to being allowed final cut on a movie.

Disadvantages
The main disadvantage of linear level structure lies in the danger that it can make
players feel constrained in their gameplay freedom, possibly arbitrarily so. The
designer needs to make sure that the roller coaster ride is an enjoyable one, as
the player is not allowed to get off and find his or her own fun. If this is not done
correctly, there is a possibility that the player will start to resent the game, which
is of course to be avoided.
1
Generally this is done in advance by the level designer.
Level Design Structure and Methodology 57

Furthermore, if things start happening regardless of the player’s own actions, a

perception of futility may be created. And since the linear structure of a level is
not the most flexible one, it is fairly common for certain gameplay events to be
forced onto the player. This can be a dangerous approach. If the player’s actions
don’t matter, than why act at all? If this is allowed to get out of hand, the experi-
ence of the entire game will be tarnished or even ruined.

Implementation Strategies
In linear level design it is vital that the designer has a good understanding of
pacing and play psychology.2 Most players will accept this kind of directed experi-
ence as long as they are directed with a sure and steady hand and the gameplay is
always rewarding or interesting. If the player does not have time or inclination to
question the direction, the level designer is successful. Unfair gameplay challenges,
or dreary boring stretches, must be avoided if possible, as the player does not
have a choice in engaging with these gameplay sections. Since these events are
forced upon the player, they have to be worthwhile for them not to grate.

Some Typical Examples

There is no hard rule on when to choose linearity in level design, although there
are times when it best supports the game’s structures or general internal goals.
For example, some puzzle games work best when the next puzzle will only be
presented if the previous one has been solved. Tetris would not be very compel-
ling if it were a free-roaming affair where the player would not feel the pressure
that its linear approach enables. There are also story-driven games that rely on
the progression of the story in a strictly linear fashion. These kinds of games may
be becoming less fashionable, but if expected to create levels for them, we have
to be prepared to rise to the challenge.

Semi-Linear Levels
As a compromise between linear and non-linear gameplay the level designer can
opt for a hybrid form. Semi-linear gameplay allows players to direct their own
experience in some instances, but it requires that players follow a script in other
instances. This can be done by a system where players can perform a certain
number of gameplay tasks or follow a number of paths of their own choosing,
but eventually are led to a bottleneck. This bottleneck can be a physical one such
2
All covered in great detail throughout the book.
58 Introduction to Level Design

Figure 4.2. Linear level structure.

as progress to a next area via a single door, or it can be a conditional event. The
event can be anything from reaching a certain number of experience points to
hitting a time limit or having collected a certain amount of items. There are plen-
ty of possibilities available within most game genres. (See Figure 4.2.)
Semi-linear gameplay progression is widespread in games, as it fits many
game types and allows sufficient control over the experience.

Advantages
Semi-linearity, if used correctly, can offer the best of both worlds. It gives the level
designer a reasonable amount of directorial control over the events the player
will experience, but leaves enough freedom in the hands of players for them to
feel they are authoring their own experience.
Furthermore, adroit level designers will do their best to create the illusion
of full player freedom, to maximize players’ involvement in the experience and
deepen their immersion. This is an important aspect of level design and one that
will be further developed in several chapters later in the book.

Disadvantages
On the other hand, semi-linearity can represent the worst of both worlds. If
applied in a lazy manner, it can arbitrarily lead players by the nose when unin-
spired, while leaving them to their own devices at inappropriate times. When a
level designer starts to give the player some freedom, more freedom will be ex-
pected throughout the level. If they are allowed to perform certain gameplay
actions at one point in the game, it is only natural that the players expect this
to be possible at other points. If the level designer does not carefully think this
Level Design Structure and Methodology 59

through, a logically inconsistent world can be the painful end result. This can
completely undermine the expectations the player has of a level and can destroy
trust in the fairness of a game. At least with strict linearity, players know what
they can expect.

Implementation Strategies
It is important in this kind of level design to observe consistency. Players will
be acutely aware of when they are treated unfairly, and because of this, the level
designer should make sure that restrictions or player direction does not feel too
arbitrary.
Where possible, techniques should be used that hide as much of these re-
strictions as possible, and that successfully create an illusion of player freedom
and choice. If players have limited progression choices, but are made to believe
they have many, the best of both worlds has been achieved. The level designer
has in that case invisibly guided the player’s experience.

Application
The semi-linear approach is probably the most popular one in level design circles.
It avoids the limiting constrictions of pure linearity and steers clear of the logisti-
cal problems of freeform gameplay. Multiple play styles can be incorporated so
the levels cater to a wider variety of game players, contingencies can be included
for when players find it hard to progress, and the world can be made to feel more
responsive to players’ choices, even if in fact the choices are limited.
These are serious advantages, and they make it easy to see why so many lev-
els follow this kind of approach.

Non-Linear Levels
In a non-linear level, the order of gameplay actions is mostly left to the player.
This is one of the reasons why non-linearity is linked to sandbox design and to
emergent gameplay.
True non-linear gameplay is very rare indeed, but some games and game
types come close to it or feature moments where it does occur. Often, this type
of gameplay goes hand in hand with interactivity; in a game where it is viable
to create one’s own gameplay, a high level of interactivity can be very helpful. A
good example of this is games where a physics system allows players to manipu-
late their environment on their own terms, but within the restraints of the phys-
ics system, thus leading to non-linear and non-scripted gameplay events.
60 Introduction to Level Design

The clearest example of non-linear gameplay lies within the realm of some
multiplayer games. Since it is the other players, and not the level designer, who
provide the bulk of gameplay, multiplayer games can be highly non-linear.

Advantages
Because the players themselves exert maximum control over the gameplay experi-
ence, they are likely to take responsibility for their own failures. Since the level design-
er does not directly dictate the progression choices, players can experience a sense of
ownership over gameplay. This is generally perceived as a positive outcome.
A game that allows players to manufacture their own gameplay experience
produces an added bonus of free content. After all, if automatic generation of
gameplay occurs, it does not require work from the level designer. It is gameplay
content created by somebody else, which in the busy schedule of professional
level design is a very good development.

Disadvantages
A non-linear gameplay world is much harder to test and is therefore harder to
design in a robust way. It may be impossible to test every single permutation of
player action that may occur, which is inherently dangerous. Unforeseen player
actions or tactics may “break” the game in unforeseen ways, possibly by allowing
some players to dominate others in an unacceptable way, or by finding loopholes
in gameplay logic that allow the player too much power.

Requirements
In non-linear levels, the level designer has to make sure that players have enough
tools to play with the game world. A requirement for non linear gameplay is that
players must be allowed to write their own gameplay story. To do this, a certain
amount of interactivity is required. There is a need for deep and engaging game-
play resulting from the player’s own actions, rather than prefabricated scenarios.
Where possible, the conditions for this must be provided by the level design-
er.3 This can take the shape of strategic depth, for example in layout design in
multiplayer games, or of purely physics-based interaction in single-player games
where the player treats the world as a toy.

3
Also, in this case, extra responsibility is laid on the shoulders of the game designer to
allow for this. The level designer can only implement what he or she is given in the game
design.
Level Design Structure and Methodology 61

Methodologies
Earlier in the book I mentioned that this text is not meant to be about level con-
struction; instead, it is focused on level design. But this does not mean that we
should not look at methodology. When eventually faced with technical construc-
tion issues, it is good to have spent time researching a number of methodologies.
This allows us to structure a level well in advance of level creation and gives us
a chance to prototype on paper. Potential problems can be found early, and the
level creation process can be scheduled easier.
This part of the book will not attempt to be a detailed listing of all possible
level design methods, but will rather focus on a number of general methodologies
that may provide hints on how to proceed in other unique cases. The main goal is
to give examples of various approaches that a level designer can adopt, depend-
ing on the situation he or she is in. It is up to the designer to determine what to
include in the level design documentation.
After having looked at structure, methodologies are a logical next area to
look at. A number of useful methodology types are shown below.

Annotated Maps
Most people are familiar with the concept of annotated maps. This is a method
widely used throughout the game industry, where the level design is described
through a gameplay breakdown of all the physical spaces in a level. This is often
done by creating a map on paper, providing a legend and numbers that indicate
the “rooms” or other gameplay areas. This can be extremely effective, as it allows
not only the level designer, but also other team members, to retrieve useful data
from the level design document.

Example of Annotated Level Design: Stolen,

Level 4
Annotated maps are commonplace in professional level design and is an ap-
proach I have had to take myself many times. To show how this may look,
the following example details a section of a level heavily weighted for stealth
gameplay, made for the game Stolen.4 The game’s player character, Anya, needs
to traverse a heavily guarded area without being spotted, as shown in Fig-
ure 4.3.

4
Developed by Blue52.
62 Introduction to Level Design

Figure 4.3. Level design detail: Satellite Array Level. (Source: Stolen, developed
by Blue52.)

Here are some annotations for the map:

02 Lower compound (links to 3 and 1)
Anya will be in an outside area where she can clearly see the building
that houses the cable car docking bay.
The player has to approach the cable car dock without being spotted,
using stealth mechanics. A route is available via the roofs of several
trucks and a number of poles but the final jump cannot be made.
In order to finalize the route the player needs to get near a forklift truck
and raise its forks that are holding a pallet. After this the player will be
able to get on the roof of the docking bay, and enter through a vent
Several guards patrol in key positions.
Anya can get to the main entrance of the docking by evading the guards
and sticking to the shadows, but there will be no way in from the ground
floor.
An alternative route takes her around the edge of the compound. She
has to evade the guards patrolling and somehow get past a guard at the
checkpoint of the area.
Level Design Structure and Methodology 63

03 Docking bay
The player finds the cable car inside, but it is not reachable from the
ground. A little tower/building gives access to it but is locked. A gantry
connects with the cable car but is too high to be reached. This is the
destination in this area.
A ledge route along the walls of the room leads the player to a vent
system high up, and can be reached by a ledge jump. Eventually the
player lands on top of gantry that triggers a real-time cutscene.
The guards holding her equipment leave the building to board the cable
car. Anya quickly jumps on top of it to stay out of sight. The cable car
starts to move up towards the mountain.
As you can see, the method can become quite involved. A very large amount
of gameplay information is recorded in a useful manner. Other people than the
level designers can also benefit. Take the following addition of information for
game area 02. It is meant to provide notes on required game assets or other re-
lated information useful to other members of the development team:
Special Assets Needed/ Notes:
Code
Interactive Objects/Other:
Forklift truck needs to have operable forklift.
Two different collision states are needed.
Art
General:
Forklift model.
Special:
None
Production
Sound:
Forklift engine.
Forklift raising its forks.
Story Events/Cutscences:
Forklift raising its forks.

Concept art
Even more information can be added to the level design. For example, concept
art can be very useful to provide the reader with a better visual guide of the envi-
ronment and its content: see, for example, Figure 4.4.
All of this builds a picture of the final shape of the level design even before
construction begins.
64 Introduction to Level Design

Figure 4.4. Concept art. (Source: Stolen Level Design, Blue52.)

Dangers
When so much information is collected into a limited amount of documents,
there is a real danger that the resulting level design may become too cumber-
some. A monolithic, unwieldy, and inflexible level design is problematic for all
parties involved. It is hard to keep up to date and may provide too much informa-
tion for some, but too little for others. If a level design document like this is ap-
propriate, it still may require further explanatory documents or visual support,
some of which will be shown below.

Flow Charts
To combat some of the clunkiness of annotated maps, we can express level de-
sign progression through flow charts. There are many positives attached to this
method. They can be created and modified quickly, which is very important if
the level design has to be flexible. They are much better able to express condi-
tional data and, in doing so, provide a logic check for the level designers. (This
can also be very important if any amount of scripting is involved.).
Here is an example of conditional level design data expressed through the
flowchart in Figure 4.5:
A player enters a room and activates a floor switch that locks the room
and releases two guards. Only when the two guards have been defeated
Level Design Structure and Methodology 65

Figure 4.5. Conditional combat situation.

66 Introduction to Level Design

Player Holds
Both Keys?

Target Area
Reached?

Figure 4.6. Level flowchart.

and two objects they were carrying have been collected5 can the player
leave the room. A cut scene is played to signify this.
This kind of information is hard to capture in annotated maps, no matter
how complete the annotation is. Flowcharts are very good at capturing specific
information and showing how it relates to other elements. This makes them very
useful indeed to create an easy-to-read and coherent overview.
Figure 4.6 shows a flowchart applied to a whole level. Note how this kind
of flowchart shows progression by providing questions and answers that signify
gameplay or game states.

5
I am sure you can figure out why this is so.
Level Design Structure and Methodology 67

Dangers
The large degree of concision and logical overview comes at a price. There are
several dangers and problems that may occur when using his method. Out of
many, here are some typical examples:
• There is no information on the level’s environment.
• Emotional impact is not represented.
• There is no temporal data (outside of chronology).
• There is no room for unexpected results.6
These and other dangers make it doubtful that a flowchart can express a
level design fully. (But I have personally seen people attempting to do just that.)
However, this is an extremely powerful method to be used in conjunction with
other methods. The use of flowcharts, for example, is very complementary to
annotated maps, described earlier.

Level Arc: Freytag’s Pyramid

In 1863, Gustav Freytag published an important book on dramatic structure
called Die Technik des Dramas. The book aimed to provide an overview of dra-
matic components as seen in classic plays, namely Greek tragedies and Shake-
spearian drama. Freytag argued that all these components follow a specific or-
der, one component leading to the next one. The well-known visualization of
this underlying structure is called Freytag’s Pyramid. (See Figure 4.7.)

Figure 4.7. Freytag’s pyramid.7

6
This can also be an advantage.
7
“Freytag’s Pyramid,” Wikipedia, http://en.wikipedia.org/wiki/Dramatic_structure,
2009.
68 Introduction to Level Design

Freytag’s Five-Act Progression Breakdown

As Figure 4.7 shows, Freytag divides drama into a number of consecutive steps:
exposition, rising action, climax, falling action, and dénouement or catastrophe, each
represented as an act within a play.

Act 1: Exposition
In Act 1, the basic story elements are introduced for the first time. We learn
enough about the setting and about some of the characters to form a framework
in which the drama takes place. This exposition ends with an inciting moment.
The inciting moment creates the catalyst necessary to allow the unfolding of an
interesting drama. Without this, there would be no reason to tell the story to
begin with. It often takes the form of a conflict of some kind.

Act 2: Rising action

After the story has been kick-started by a central inciting moment, further depth
and drama is added during Act 2, rising action. This is done by introducing fur-
ther layers of drama (possibly including new conflicts, hindrances, and obstacles
to the protagonist’s goals) and by introducing new characters and other compli-
cations.

Act 3: Climax (turning point)

The rising action leads naturally to a climactic moment. This is a turning point to
the story: a key moment where a change occurs that is fundamental to the story.
This change can be for better or for worse, depending on the type of play. Typical-
ly, in a comedy the change is for the better, while in a tragedy the opposite occurs.

Act 4: Falling action

The conflict between the protagonist and antagonist reaches a critical stage dur-
ing the falling action part of the play. The drama will progress towards a resolution
that serves either protagonist or antagonist. The final direction of this resolution
can be kept ambiguous for a certain amount of time to create suspense.

Act 5: Dénouement or catastrophe

The final resolution of the play comes in the form of a positive conclusion (dé-
nouement) in case of a comedy, or in the form of a negative ending in the shape
of a catastrophe. Both conclusions show a progression in the experience of the
protagonist. In a dénouement, the protagonist is better off than at the beginning
Level Design Structure and Methodology 69

of the play, while a catastrophe leads to the opposite conclusion. Either way, the
drama gets resolved.

Five-Act Progressions in Level Design

Nobody suggests that the above description of a five-act classical play or drama
is representative of all plays. It is merely a description or analysis of a particular
kind of drama that has often been used in the past. Judging by how many plays
have followed this progression, it must have been very popular with a great many
writers. Perhaps this is because it creates a recognizable format in which to tell
the story: a frame of reference that an audience can relate to. In truth, the five-
acter is still very much alive and is used throughout the arts, not in the least in
Hollywood cinema.
If we extrapolate from this, it may be possible to identify similar principles
within level design. Let’s take Bungie’s Halo as a test case:

Halo as a five-act level design progression.

Halo is a good subject for this experiment, because despite a certain amount of
freedom in the way the game can be played, the story progression through the
game is fairly linear. In other words, many of the methods are up to the player,
but the gameplay goals are predetermined.

Act 1: Exposition
In Act 1, we are introduced to the protagonist, Master Chief. Through exposi-
tion, we learn much about the background to the story and the story’s environ-
ment. We learn that the protagonist is a soldier and that the current action takes
place on board a space ship. The inciting moment that functions as a catalyst that
escalates the story takes the form of an alien attack on the space ship.

Act 2: Rising action

Act 2 serves as a vehicle for introducing elements to the game that deepen the
experience. The player has landed on alien structure Halo and has to learn how to
combat the resident hostile alien forces. In the meantime, we learn more about
the background story of the conflict.

Act 3: Climax (turning point)

In the climax of Halo, the player learns that a far greater danger than the alien
protagonist exists. It is a viral life form known as the Flood. It infects and takes
70 Introduction to Level Design

over other life forms in such a way that the Flood ends up corrupting and con-
trolling its victims. The Flood have started to infect both the aliens and the hu-
mans, and their threat lies in infecting entire planets.

Act 4: Falling action

Much of Act 4 is defined by the player’s epic struggle to contain the threat of the
Flood.

Act 5: Dénouement and catastrophe

In the final act, we already know that Master Chief has been successful in his
main goal, stopping the Flood from spreading by destroying Halo. The final out-
come of the story centers on the personal survival of the hero.

Problems Associated with Freytag’s Pyramid

Although there are many uses to the pyramid, and there are many examples of
dramatic creations that fit its definition, there are a number of criticisms that can
be aimed at Freytag’s pyramid that are worth investigating.

Oversimplification
It can be argued that to create such a neat overview, the subject must be over-
simplified beyond the point of being useful. One could argue in the case of Halo
that the story features several turning points. Should we ignore some of them in
order to point at one specific climax? If we don’t, does it mean that Act2 some-
times appears after Act 3?

Restriction
Restriction is the enemy of creativity. At least, arbitrary restriction is. What if we
want to start with a catastrophe and a climax, and through exposition and falling
action slowly tell a compelling story? We may want to write a comedy with a fatal
ending or a tragedy with a happy ending.
All of these criticisms are valid to a point. As a device for forming an
interesting dramatic construction, the pyramid is hopelessly restrictive and
oversimplified. But the pyramid’s value is not one of dramatic creation;
rather, it is a tool for dramatic visualization and analysis. After all, Freytag
intended to analyze and show the dramatic structure of classical plays.
Considering the ubiquity of his work to this day, he was very successful
in achieving this goal. There is no need to slavishly follow this particular
Level Design Structure and Methodology 71

method, but it can inspire us to devise our own methodology to suit our
own needs.

Applications for Level Design

If we extrapolate from this, we should be able to devise similar techniques that
successfully aid us in level design analysis and visualization. This is something
that in the hectic and often stressful reality of professional level design can
be of immense benefit.8 There are several techniques and approaches avail-
able to level designers, but the ones most commonly used are based on event
diagrams.

Event Diagrams
An event diagram9 is a graphical representation of the content of a level, poten-
tially including representations ranging from emotional impact to duration, or
combinations of several factors. There are many elements that can be included in
these diagrams, but typically they will show one of the following things:
• type of event,
• duration of event,
• chronology (if applicable),
• impact (color-coded).
Take the example in Figure 4.8.

Figure 4.8. Event diagram.

8
In fact, it is beneficial in all circumstances.
9
Admittedly, an entirely made-up term.
72 Introduction to Level Design

As we can see in this example, the following elements contribute to a clear

picture of the level:
• type of event (legend),
• event Duration (column width),
• chronology (column order),
• impact Value (column height).
These elements can each make a valuable contribution to the level design.
Using this method, it is easy to identify and read them.

Type of event
The type of event can consist of any gameplay event intended by the level de-
signer to occur at a specific moment. This can range from the very impor-
tant unique moments (Set Piece X) to more generic or modular occurrences
(Scripted AI Encounter Y). In the above example, other gameplay events in-
clude: new skill introduction, boss encounter, pickup cache, and master key found.
Alternatively, or as a complement, less specific event descriptions can be used,
like combat, platforming or puzzle, which still provide a good overview of a
level’s content.

Event duration
It is unlikely that this data can ever be exact, but it is still possible to convey
helpful information. For example, it can be useful to show duration in relation
to other events, or one can show patterns in gameplay that may be problematic,
like a sequence of long, but low impact, gameplay. This may indicate a section of
gameplay that the player may find boring. If precise information is known, this
can be included alongside less specific descriptions, such as short, medium and
long.

Chronology
As noted earlier in this chapter, chronology is partly dependent on linearity.
Nonetheless, even in very open freeform games with a non-linear structure,
things still often happen within a certain sequence. Skill trees may have to be de-
veloped, some areas may only unlock if a certain item has been found, defeating
a specific creature may unlock a new skill, or the collection of a specific amount
of items may allow the player to buy new gear.
In the above example, the boss encounter does not occur until after the master
key is found. This kind of information makes it easier to maintain continuity and
avoid many bugs.
Level Design Structure and Methodology 73

Impact value
Emotional impact is a very vague term. How do we measure it? There is no sat-
isfying answer to that, especially in instances where one player’s subjective ex-
perience may be different from the next player’s experience. But this does not
mean that we should not try at all, especially in the type of levels where we are
able to author the general experience of all players. Let’s say, as in the above
example, that we have included a set piece event in the level. Naturally, we as-
sign a high value to it, most likely higher than for any other events in the level.
If later playtests show that its impact on the player lies well below other events
in the level, we know that we need to do something to raise the impact value of
the event.
Instead of a general value like emotional impact, it can often be a better idea
to represent a more specific element. In a survival horror game, this can be a scare
factor, or in an action adventure it can be an action quota. Use whatever is right
for the game you are working on.
How this is represented in the diagram is also open to many choices. Typi-
cal choices are color coding and column height, the latter being the one shown in
our example. A line diagram can be used, or even a pie chart; it is up to the level
designer to determine what is appropriate.

Event Diagrams Summary

Event maps provide an often-useful visual overview of important level design
data. This allows the designer to quickly assess the level’s overall tone and makes
it easy to give others on the team a quick summary. However, in most cases it
should not be seen as an alternative to a detailed and thorough level design doc-
ument; instead, it is more effective in conjunction with one. This is especially
true when it comes to highlighting specific elements within a level, as shown in
previous examples.
It is of course entirely up to the level designer to decide which data to in-
clude and what type of visual representation to use.
The methodology is easiest to apply to linear level designs. Nonetheless, the
methodology has applications even for completely open level designs.
Note that his methodology is not to be confused with beat maps. These pro-
vide a graphical overview of the player’s “heart rate”10 at specific moments in the
level, but this is almost never a sufficiently useful technique.

10
Typically by displaying a number that signifies heartbeats per minute.
74 Introduction to Level Design

Object-Oriented Level Design (OOLD)

As already touched upon in Chapter 3, “Level Design Goals and Hierarchies,”
it is helpful to think of level design components as modular objects. Another
way of describing this methodology is as object-oriented level design. This term is
based on an idea that originated in the world of computer programming, known
as object-oriented programming. Object-oriented programming is:
a modular approach to computer program (software) design. Each
module, or object, combines data and procedures (sequences of in-
structions) that act on the data; in traditional, or procedural, program-
ming the data are separated from the instructions. A group of objects
that have properties, operations, and behaviors in common is called a
class. By reusing classes developed for previous applications, new applica-
tions can be developed faster with improved reliability and consistency of
design.11 (Emphasis mine.)
One does not have to be a programmer to realize that from a design per-
spective, reuse, improved reliability, and consistency of design are major advantages.
These are things that as level designers we always should try to achieve. If we
look at object-oriented programming (OOP) in greater detail, we find further use-
ful principles. There are a number of fundamental principles associated with
OOP. These include these are: encapsulation, inheritance, and polymorphism. All
of these should be of interest to a level designer.
These terms have been defined as follows:
Encapsulation refers to the creation of self-contained modules that bind
processing functions to the data. These user-defined data types are
called “classes,” and one instance of a class is an “object.” For example,
in a payroll system, a class could be Manager, and Pat and Jan could
be two instances (two objects) of the Manager class. Encapsulation
ensures good code modularity... [I]nheritance allows the structure and
methods in one class to be passed down the hierarchy. That means less
programming is required when adding functions to complex systems ....
The ability to reuse existing objects is considered a major advantage of
object technology....
... Polymorphism in the object-oriented approach refers to the ability
of a programmer to treat many different types of objects in a uniform
manner by invoking the same operation on each object. Because the

11
“Object Oriented Programming,” The Columbia Electronic Encyclopedia, Sixth Edition,
http://www.bartleby.com/65/ob/objecto.html, 2001–2007.
Level Design Structure and Methodology 75

objects are instances of abstract data types, they may implement

the operation differently as long as they fulfill the agreement in their
common contract. 12 (Emphasis mine.)

Practical Examples of OOLD

In Chapter 3 we discussed modular level design. OOLD is a method that takes
the principle of modular level design and applies it in practice, ensuring modu-
larity and other similar concepts. The previously described elements of encapsu-
lation, inheritance, and polymorphism can be easily incorporated into level design.
Let’s go through them one by one.

Encapsulation
In ID Software’s game Quake, the player can often traverse a level with the aid of
teleporters. The player just walks into one and is teleported to a different loca-
tion in the level. This can be considered a class in level design terms. Once the
basic elements are in place (such as trigger areas, destination areas, and three-di-
mensional models), we can have a class called a teleporter. Individual teleporters,
as objects, can be one-way teleporters, or might feature different 3D models and
have unique destinations and starting points. But they will always be teleporters
and belong to the teleporter class.
Let’s take a less literal example next: encapsulation applied to gameplay sce-
narios. Let’s say that a level designer has spent a significant amount of time set-
ting up a combat situation that makes use of specific aspects of the environment.
(For example, cover, line of sight, and height difference.) If this particular situa-
tion works well in the game, the level designer may decide to replicate it by keep-
ing the essential elements in place, but then providing different-looking environ-
mental details, such as different AI opponents. The encounter’s integrity stays in
place, but the original benefit of the gameplay is encapsulated in each instance.
(Or the environmental detail may stay the same and the combat can change.)

Inheritance
Early in the level design process for a game, we may want to create a generic
object called a treasure chest, scripted in such a way that when the player opens it,
a random item of limited wealth is generated (once). This may require a certain
amount of scripting from the level designer, but once it has been set up, this
treasure chest should be useful throughout the game. Say that play-testing now

12
“Object-Oriented Programming,” The Free Dictionary by Farlex, http://encyclopedia2.
thefreedictionary.com/Object-oriented+computer+programming, 2009.
76 Introduction to Level Design

shows that players highly enjoy these moments of excitement where a random
award is received, and the game team decides that this type of occurrence should
be more prevalent in the levels. To avoid littering the level with treasure chests,
the level designer decides to create a number of new objects, such as cupboards,
filing cabinets, desks, or any furniture or similar object that can be opened and
closed and used as a container. Players can now find treasure all through the level
by opening up these new objects, without the objects feeling out of place.
Although these physical objects are new classes (in programming terms),
when programmed as the same type of object, they can inherit much of their
function from the behavior of the original treasure chest object.

Polymorphism
When a level designer can incorporate polymorphism into his or her work, a
huge amount of flexibility is gained. Let’s go back to our earlier example of trea-
sure chests and the creation of other classes with overlapping functionality. Say
that the game is nearly finished, the designer has placed hundreds of these chests
and cupboard and drawers, vases, you name it, throughout the level. The pub-
lisher however now demands that a cash register sound should play whenever
the player receives treasure. If polymorphism has been factored in the designer
should be able to associate this sound to any object that makes use of this be-
havior simultaneously. This means that making the change once, to the behavior
applied to all of these objects, not to all the objects individually.

Dangers
If we go back to our earlier example of Halo, we can find a good example of a
persistent danger of this technique: if not careful, the level designer risks creat-
ing highly repetitive gameplay sections. In Halo this took the form of the now
infamous “library levels.” The library levels were created in such a way that they
reused environments and gameplay in a very thorough manner. This was done in
such a manner that the player is confronted with arbitrary repetition. Even if as-
sets and gameplay sections are re-used, they still have to feel individual and worth
playing. In the library levels, the player literally has to play the same thing over
and over again, which is more akin to dreary work than to rewarding gameplay.

OOLD Summary
Applications of this kind of thinking, of borrowing these programming concepts
and applying them to level design processes, are widespread and far reaching.
Level Design Structure and Methodology 77

They range from simple puzzles to extended gameplay sections and even whole
areas of level geometry. The examples themselves are not important; it is more
about locating moments when this kind of thinking can be applied to anywhere
that benefits from it.

Dense Level Design

An example of a level design methodology that combines elements of previ-
ously discussed methods can be found in dense level design. We speak of dense
level design when a larger-than-normal amount of gameplay is incorporated in
the physical environment of the level. This type of density can be achieved in
many ways but generally occurs through reuse of the environment in time. New
gameplay elements can be introduced to the level when a player revisits a sec-
tion.
Another aspect of dense level design is density of gameplay space. All as-
pects of the gameplay space are exploited if possible. This can mean that a level
will make much deeper use of vertically layered gameplay, incorporating all ar-
chitectural features. For example, many doors can be opened and rooms can be
entered, even if they serve no direct gameplay purpose.

Advantages
There are many clear advantages to this approach. So much so that in most cases,
I advocate the use of this technique as one of the first things to look at in many
level design tasks. Let’s look at some of the advantages.

Reuse of assets
In most cases, as level designers we are expected to make the best use of the
assets we have. If interesting gameplay can be replicated over several iterations
without changing the assets needed, the resulting gameplay comes at a lower
cost to the game’s development.

Reuse of gameplay
This reuse of gameplay has many advantages, but an extremely important one is
that it allows the level designer to teach players how to enjoy the game more eas-
ily. This is the case because the designer can take initially simple gameplay sec-
tions and slowly upgrade their difficulty and skill level in a controlled manner,
without alienating the players. In this case, repetition and familiarity are your
friends.
78 Introduction to Level Design

Immersive effect
If an environment is used to its fullest, if all of its features, be they interactive,
artistic, or physical, are incorporated consistently in the level’s gameplay, it will
create a much more immersive experience. It gives the environment a sense of
reality. Even if set in a non-realistic framework like fantasy or sci-fi, the environ-
ments still have to conform to the logic of that framework.

Dangers
There are several dangers associated with this technique, and some of them are
hard to avoid.

Spatial confusion
Game levels are experienced through a two-dimensional interface. In most cases,
this is a computer screen or television monitor and some kind of game control-
ler. While this method is fine in principle, it does require certain concessions
that in normal life do not occur. A good example of this is the way gameplay
spaces in levels are often simplified to facilitate orientation.

Lack of gameplay readability

Most level designers are weary of reusing the same environment too many times
without serious modification, as it is much easier to give a level area a specific
purpose that is understandable to the player. If the player revisits an area and the
gameplay has changed, it may need some re-evaluation on the player’s side. This
is not necessarily a problem, but there needs to be a good reason for this, or the
player will feel ambushed. Something can only be done so many times before it
becomes annoying.

An example: MINERVA—Metastasis 2
MINERVA is a modification for Half Life 2, consisting of a number of episodic
levels designed by Adam Foster. The MINERVA mod actively reinterprets the
gameplay of Half life 2 by introducing a much stronger emphasis on density of
play.
Instead of relying on horizontally-sprawling, immense maps that stress
the engine’s area-capabilities to its max, MINERVA maps are incredibly
small. This is because of Foster’s ground-breaking idea to utilize every
possible area to its maximum potential, and instead of expanding hori-
zontally, he expands vertically. Rather than leave large areas wasted with
inaccessible buildings, “fake” corridors and rooms to give the impres-
Level Design Structure and Methodology 79

sion of an immersive, realistic environment, Foster makes every area

accessible. This doesn’t mean that one can simply travel all over the
maps in any manner one chooses—but instead through the use of very
creatively-placed barriers Foster is able to funnel the player around the
maps in a spiraled fashion.13
All throughout this level, it is notable that the gameplay is extremely layered. The
player almost constantly revisits areas that have been traversed before, but which
are updated with new gameplay features. This is especially true for combat with
the game’s enemy soldiers (Combine soldiers), presumably because they can use
the same paths and doors as the player. So it is not illogical for them to make full
use of the environment as well. The end result is that the action comes and goes,
but the environment stays consistent. In most cases this is good enough, as the
available tactical variation is strong enough to support this kind of repetition.
Furthermore, it also contributes to the environment feeling very real, as if archi-
tecturally properly designed.
Where the level fails, though, partly because of the use of dense level design,
is in readability. Quite often, players find themselves lost or unclear where to go
now, a danger that is especially great if routes are backtracked several times. A
player who has already been down a known route several time now has to won-
der at every corner: “I’ve been here already, but has something else opened up?”
Nonetheless, through the consistent use of dense level design, the final experi-
ence is very powerful. The player develops a real bond with the environment,
which is very important in level design.

Further Examples
There are, of course, many other methodologies applicable to level design, some
more obscure than others. Some level designers like to start their design by first
working on key elements of the levels, to make sure that all the impressive fea-
tures are worked out in time. Others work in a completely opposite manner,
sketching out all the lighter detail first to make sure that a functioning outline of
the level exists early on.

13
“MINERVA: Metastasis 2,” MINERVA: Metastasis 2—Planet Half-Life, http://plane-
thalflife.gamespy.com/View.php?view=HLMotw.Detail&id=7, 2009.
Single-Player vs. Multiplayer
Considerations 5

O ne of the approaches taken for this book is to make the text indepen-
dent of genre and technology. Wherever I can, I try to explore con-
cepts and theories that have as much range as possible and lend themselves to
multiple applications, without being dependent on things like existing tech-
nology or game engines. If at all possible, I try to avoid too much focus on
specific genres to make general points, although I do examine aspects that
are unique to specific genres to make sure their pitfalls and idiosyncrasies are
understood.
This book applies the same philosophy to multiplayer and single-player as-
pects of level design. When possible, it tries to provide material that is useful
to both game modes. However, at times, chapters will deal with specific and
custom-made elements on a case-by-case basis.
Before we reach those specific aspects of multiplayer and single player level
design issues, I would like to make a number of general observations, which can
function as a useful background against which we can place more specific is-
sues in upcoming chapters. This chapter will highlight some of those specific
elements.

Single-Player
We speak of single-player gaming when the gameplay occurs in isolation from
other players. In basic terms: when the player can enjoy the game without
input from other people. This does not mean a game cannot have both single-
player and multiplayer content, just that it has to be able to provide an experi-
ence that allows a single player to enjoy it. This is sometimes referred to as solo
play.
81
82 Introduction to Level Design

Single-Player Level Design Considerations

There are a number of very clear considerations that are relevant to single-player
level design, sometimes aiding the level designer, sometimes providing obsta-
cles. Here are some typical examples.

Control
A major advantage of designing for a single-player experience is that it gives the
designer more control over what happens to the player. There are no other play-
ers involved whose input may cause unpredictable outcomes, or whose input
can counter the themes and moods the designer is going for. Imagine a dark
and atmospheric survival horror game filled with perverted Freudian overtones
meant to disturb the player on a psycho-sexual level. Now imagine that all the
time, two co-players in the background are constantly giggling at those dark sex-
ual themes. This would simply devastate any efforts to slowly build up a carefully
crafted experience.
Single player levels don’t have these problems at all. Designers can do their
work in providing a directed experience without fear of third parties ruining it
for the player.

Balancing issues
A potential negative lies in the possibility that the designer may misjudge the
mindset or experience or skill level of players and create bottlenecks that need-
lessly frustrate players, or that even stop them from progressing.
This is not a rare occurrence unfortunately. A level designer should be care-
ful when it comes to implementing such harsh bottlenecks. The fact of the matter
is that the level needs to provide a good experience for all kinds of people, and
that this is harder than often imagined.1 Generally, it is a bad idea to try to please
everyone, but that does not mean that we shouldn’t try to balance the game for
those people who are deemed the target audience for the game. Later chapters
go into much detail on this matter. For example, see the sections on flow theory,
dynamic difficulty adjustment, and pacing later on in the book.
Either way, single-player levels require great empathy for the player in order
to create a balanced game experience

Narrative possibilities
Although multiplayer levels provide plenty of opportunities for narrative con-
tent, it can be easier to aim narrative elements at a single person, rather than at

1
Sometimes it is even impossible.
Single-Player vs. Multiplayer Considerations 83

a group of players. The game can check for certain player acts or events to occur
and can be certain that they have been triggered and experienced by the same
person, while this cannot always be easily determined in a multiplayer context.
This mostly applies to games where it is important that information be experi-
enced linearly and absorbed in such a way that the game can rely on the player
having certain knowledge. This can occur in text adventures or point-and-click
adventures, for example.

Multiplayer
We speak of multiplayer gaming when some of the game content or experience
is provided via interaction with other players. Although most video games are
designed with a single-player component, multiplayer video gaming has taken
a real step forward with the advent of online play. Some of the most successful
games2 ever created are mostly multiplayer affairs. One cannot be a fully rounded
level designer without taking into account the needs of a multiplayer audience.
Let’s look at the most typical multiplayer modes.

Multiplayer Cooperative Play

Sometimes it is fun to help each other and tackle challenges together. It can
be fun to solve a murder mystery with a friend, or go row-boating together.
Video games are no different, and some of them provide this specific type of
multiplayer gaming. This type of gameplay has become much more normal
with the advent of Massively Multiplayer Online games, like World of Warcraft,
where groups of players band together to tackle the game’s obstacles coop-
eratively.

Multiplayer Adversarial Play

A very common type of multiplayer gaming consists of people competing
against each other as adversaries or opponents. There are more subcategories of
this kind of play than we need to go into, but some of the more popular ones are
the following:
• Coop: This is the traditional cooperative mode where players can jointly
take on the single-player game’s challenges, although often with enhanced
difficulty and optimized content.
• Versus play: Chess, backgammon, table tennis, beat-em-ups.
2
Blizzard’s World of Warcraft provides a good example.
84 Introduction to Level Design

• Deathmatch: Players fight each other in arenas and score points by annihi-
lating each other.
• Capture the flag : Two teams fight each other by trying to steal the other
team’s flag and returning it to their own base.
• Competitive sports gaming: Just as in real sports environments, video
game versions of sports games can offer human opponents.
Many other examples exist; it is a very popular way of playing games.

Multiplayer Level Design Considerations

Depending on the type of multiplayer game, there are some very interesting dif-
ferences than those found in single-player level design. In one form or another,
level designers working on these kinds of games are likely to have to deal with
content that features the following characteristics.

Enhanced routes and challenges

In games where player characters traverse an environment, the multiplayer com-
ponent allows the level designer to create routes that make use of this extra di-
mension. Routes and challenges can be devised that require players to cooperate
in order to progress. A player may need to stand guard and fight off marauders
while another player picks a door lock to open up the escape route. Two players
may need to throw a switch simultaneously. It is easy to think of content in this
regard. Possibilities abound, and that is very useful for level designers, as it is
almost always enjoyable to solve game challenges with another player. It auto-
matically grounds players’ actions into the game world, and this really deepens
the game’s immersion and enjoyment.

The human element

Another bonus to multiplayer games is the opportunity to include human ca-
pability in the game’s experience. What I mean by this is that the content of
the game can be directly enhanced by other people in ways that programmed
or animated or modeled content cannot. In a deathmatch game, for example, it
is much more exciting to play against human opponents than against artificial
intelligence opponents, normally known as bots. Not only are humans better at
coming up with completely unexpected strategies; they are also able to adapt to
those of others.
Additionally, other players can introduce elements of real emotion into
gameplay, such as a passion to win a match, empathy for other players, or just the
ability to converse with others within the game.
Single-Player vs. Multiplayer Considerations 85

Where possible, this needs to be enhanced within the level designs. It is use-
ful to give players well-defined moments of human-to-human interaction within
a level. Place people in predefined face-off situations. Provide players with co-
operation dilemmas: can I trust the other player or not? How to engineer those
moments depends entirely on the game and genre in question, but it is always
good to keep this principle in mind.

Clearly defined competitive challenges

When it comes to competitive play, nothing is simpler than creating a situation
where two or more players compete with each other on the same task to see who
is better at it.
To a degree, this can be emulated in single-player games. But the human face
of an opponent is often the difference between a pastime and a serious competi-
tive challenge. For example, this really comes to the fore in deathmatch- or team-
based adversarial play. The level designer needs to do no more than provide an
environment that is balanced for the competing teams, and then allow the teams
to fight it out in whatever game modes are applicable. There is no need for fancy
extra special challenges; these will be provided by the players themselves.
In capture the flag, for example, it is quite normal for a level to consist of two
halves that are roughly mirrored in order to guarantee that both sides have the
same advantages or disadvantages. In deathmatch levels, we often find simplified
environments that act as smooth flow in arenas, with very few bells and whistles.
The gameplay is largely enjoyable because the level simply gives the players a
good excuse to compete against each other. The same goes for multiplayer racing
games, and for too many other genres to mention.3

Single-Player and Multiplayer Hybrids

Although rare, there are examples of games that manage to fuse both single-
player and multiplayer mechanics. This sounds like a paradox, after all; how can
a game be both a multiplayer game and a single-player game? Isn’t it true that
the moment more than one player is involved, the game type automatically has
to be seen as multiplayer? Well … maybe not. What if a player is not aware of
other players? Is it possible to design a game where players can choose to experi-
ence the gameplay as if it were solo play, while still surrounded by other players?
These may well be inconsequential questions, but some of these types of games
have emerged in recent years and may be worth studying.

3
There is easily enough material there to fill another book.
86 Introduction to Level Design

Simulated Multiplayer
This is a relatively rare phenomenon, but it can occur when human opponents
are absent, but their roles are played by AI opponents. Good examples of this
can be found in multiplayer games that provide tutorials where players can prac-
tice their skills against computer AI. Additionally, we see it in games that are
meant to have humans playing each other, but that provide a simulated setting
for when no other people are available. As previously mentioned, bots in First
Person Shooters (FPSs) are often used in this context to provide opponents to
simulate human players.
Special mention needs to be given to Alternative Reality Games, which often
blur the line between real players and simulated players. This will be explored in
more detail in Chapter 12.
Towards a Shared Grammar
for Level Design 6

L anguage is often used as a metaphor when describing aspects of art. We talk

about how a painting speaks to us, or how a movie has much to say on a
particular subject. We speak of an artist’s vocabulary or comment on an artist’s
voice. There are plenty of further examples relating to almost all art forms, so it
is understandable that this is such a useful metaphor. Everybody understands
statements such as, “Artists want to speak to an audience, to get a message across
that hopefully, the audience wants to hear.”
There is no reason not to employ the same metaphor when discussing level
design. Unfortunately, people are not yet used to looking at our field in such a
way, perhaps because level design is such a young profession. Nonetheless, re-
gardless of how many people are aware of it, the field is slowly developing its
own voice, language, syntax, dialect, or whatever you want to call it. It is impor-
tant that level designers understand this and try to become fluent within the
language of their own field.
The next part of the book is mainly about level design concepts, theory, and
practice. But hopefully, as a consequence of studying these things, level designers
or other interested parties will be aided in understanding, or contributing to, a
shared grammar of level design. Hopefully, we can work to gain a better under-
standing of the language of our art form to allow us to get our message across
loud and clear.

Conventions and Language

When a new art form develops, it starts to recognize that there are certain tech-
niques available to it that are more useful than others. Artists start to learn how
to effectively communicate with their audience, while the audience becomes
better at recognizing which techniques the artist is using. There is a dynamic at
87
88 Introduction to Level Design

play here, where both the audience and the artist develop together on an indi-
vidual level, as well as scaling up to a societal and cultural one. There is in fact
a constant dialogue of sorts between artist and audience. This dialogue is only
possible when artist and audience can communicate, at least to a small degree, in
a mutually understood language. Before we can develop and better understand
our own language for level design, we need to understand a bit more about how
an art form’s language develops in general. And to do this, we need to look at our
understanding of conventions.

Conventions
When artists, or any entertainers, for that matter, try to make a connection
with an audience, they use certain techniques they know to be effective. Some
techniques come from natural reactions; a loud bang will make you jump, slow
movements are non-aggressive, etc. Other reactions, however, are conditioned
ones. Those techniques are based on repeated usage through time; the audience
recognizes them from previous encounters. For example:
• The scarred man is the villain.
• The hero gets the girl.
• The villain ends up dying a terrible death.
• The comedy will have a happy ending.
• Lassie will come home.
These are events that we have come to expect from many Hollywood stories,
pulp novels, and other creative genres, to the point where they have become cli-
chés. They may not be the best thing for an original story, but they are something
that the audience will easily recognize as genre staples. They are part of a formula.
These genre staples can also be described as conventions.
The idea of conventions refers to nothing more than an accepted truth or
an accepted way of looking at something, and it covers many more things than
entertainment clichés. Conventions in general occur when a culture or signifi-
cant group of people accept a certain set of rules that govern something they are
exposed to. If this sounds needlessly vague, let me give you some examples to
show the scope of this concept.

Fashion
Most people conform to accepted conventions of dress and fashion. It is ex-
tremely rare for men to wear miniskirts, wear pajamas outside the house, or wear
a suit to work that features a giraffe motif. Some people flaunt these unspoken
rules; but they are seen to be eccentric, or indeed unconventional.
Towards a Shared Grammar for Level Design 89

Social Behavior
There is a large amount of behavior that is conditioned and based upon a mu-
tual understanding of conventions. We don’t speak with our mouth full, throw
change on the ground, pinch a stranger’s nose, or steal each other’s property
without it being frowned upon. We decided upon these conventions a long time
ago for diverse reasons, but we now agree to stick by them.

Literature
In most books, it is common to gain insight in the motivation of the protago-
nists. Most stories feature some kind of conflict, and events occur in such a way
that they make chronological sense. It is often asserted that only a certain limited
amount of stories can exist, and that all stories fit within these kinds of arche-
types. Books that flout these rules are seen as experimental, difficult, or indeed
once again, unconventional (a term that is revealing in itself).

The Language of Art

As we have seen so far, there are many conventions based on natural and condi-
tioned reactions, and it is important to be able to see the distinction between the
two. What matters most, however, is our ability to translate these conventions
into a coherent framework. This is important in social situations, so that we all
agree on how to interact with each other. But it also governs other aspects of our
lives, including how we dress or even how we experience art.
Artists are often most effective if they can translate existing conventions,
or develop new ones, and integrate them in their art in a way that enriches the
work. Playing with conventions directly touches the experience for the audience,
as it plays with their expectations and conditioned responses. These conventions
become part of a toolset for the artist, or put in a different way, they become part
of the artist’s vocabulary.
When a set of artistic conventions has existed long enough that it has be-
come stable, it can develop into something deeper. The conventions start to take
place within a coherent context, or they form a prism through which the medi-
um as a whole is perceived. For example, this has happened with conventions in
film. People almost universally understand the use of certain filmic conventions
even when they are quite abstract in nature, like screen distortion in a flashback
scene, or complicated camera moves that nonetheless are accepted by the audi-
ence. These kinds of integrated and coherent sets of conventions are part of what
is often described as the language of the medium.
90 Introduction to Level Design

The Language of Level Design

Just like any other art form, level design is subject to the same principles of con-
ventions and the medium’s language. Game players can read many aspects of
level design these days. Many conventions already exist and contribute to an
evolving and ever-richer level-design-specific language.
• We know that a health pack lying in a sealed-off area implies that access to
this area can be gained.
• It is understood that a cut scene showing a lever and a door explains a rela-
tionship between the two.
• It is accepted that in most 2D platform games, jumping on a creature’s head
is an acceptable action to perform.
• It is clear that searching behind a waterfall may yield the discovery of hid-
den loot.
At one point, not so long ago due to the youth of our profession, all these
conventions were new and fresh. We need to recognize these achievements so
that we can use them in our own work, or even develop our own conventions.
The most important aspect to integrate in our understanding is that we are all
constantly using and working on developing a coherent language of level design,
complete with its own syntax and grammar.
Note that when dealing with these elements of conventions and a medium’s
language, it is extremely important to be consistent. Even if new conventions are
established within the game the level designer is working on, it is vital that unless
there is a very good reason, the convention has to apply all through the game.
If this rule is not followed, the player is basically taught that the game world is
arbitrary, unreadable, and ultimately unfair. This kind of betrayal of a consistent
set of rules is a perfect way of alienating the player.

Level Design and Art

Over the course of this book, I will often use the terms fun, entertainment,
and art. The terms will frequently be used to describe level design goals, or
desirable outcomes. The terms are not interchangeable, and all three can be
attained in a level simultaneously. Quite often they are connected or even in-
terdependent. This is especially true for art and entertainment, which form
a complex relationship with each other. There is no consensus on where one
thing starts and the other ends. To make matters worse, the meaning can mu-
tate and change over time as one era’s entertainment can form the next era’s
art.
Towards a Shared Grammar for Level Design 91

As every art student knows, it is pointless to try to create a shared definition

of art. In fact, it is a sure-fire way to start a heated argument, unfortunately one
that can have no resolution. Nonetheless, I often speak about level design as an
art form. Although this notion of level design as art is disputed by many people,
I see it as an irrefutable fact. The fact that most games are not art may be true, but
this does not mean that games can’t be art. This argument is true for many other
creative fields, like music and film. Level design can create art, and therefore in
my opinion it is an art form. (Nobody thinks film is not an art form because of
the overrepresentation of formulaic box-office creations.)

Conclusion
What more established art forms have in common with each other is that there is
a language assigned to them that allows for critical analysis and discourse. They
have developed a shared grammar of the medium that applies to all their dispa-
rate elements, and this is something I feel we should also aim for. Sir William
Thomson (Lord Kelvin) said: “If you cannot measure it, you cannot improve it.”
I would like to add to that the notion that “if you cannot discuss it, you cannot
make it better.”
It is important to grasp the significance of a medium’s language and study
it, as it teaches us how and when communication occurs in the medium. Just as
with a language in a linguistic sense, it is much easier to express oneself in a lan-
guage when the underlying culture is studied as well. Level design as an art form
also features a shared artistic language. Some elements are related to other art
forms, like film, storytelling, or photography, to name a few. But it also features
elements that explicitly belong within the field, based on conventions and ideas
developed specifically for level design.
It is this combination of elements, if properly studied and understood, that
can turn our designs into a powerful message spoken with a clear voice.
Emotional Feedback II
Systems

O ne of the ways in which the success of an art form or an artistic expres-

sion can be measured is by gauging its emotional impact. Can it move us
emotionally? Can it cause anger or sadness, create hope or explain despair? If
we look at other art forms, this question is easily answered; many examples can
be quoted from films, books, photos, musical compositions, etc. All of these art
forms have found a way to touch us in some way or speak to our inner self. They
have learned to reach an emotional core and can become part of our inner life.
What happens if we look at level design, and its attempts to reach the inner
life of a game player? Can we honestly aspire to similar achievements on an emo-
tional level as other more established art forms?
I think we can, although we have far to go when it comes to fine tuning our
proficiency in this matter. Nonetheless, it cannot be denied that level design has
created its own set of emotional feedback systems. We have developed tried and
tested methods for creating and feeding emotional responses from our audience.
As level designers, we have a strong advantage over most other art forms when
it comes to his kind of engagement. Our audience will actively participate and
contribute to the experience. This willingness to engage with our work is a key
aspect of understanding the power of level design. Because of this participatory

Simplified gameplay feedback loop.

1

93
94 Emotional Feedback Systems

aspect of the art form and the final control we have over content, we can predict
many of the actions and emotions of the player. This means we can take these
player contributions and enhance them, alter them, reward them, and reflect
them. Such reactions provide the underlying principles of the use of emotional
feedback systems in level design, based on an understanding of the player’s own
emotional investment and contribution.
The following chapters will look at a number of these systems that are avail-
able to us and provide clear and practical examples that show how we can best
use them. Ultimately it is vital that as in other art forms, we can to engage with
the player’s inner life. It is important that we can create a real emotional response
through our work. The good news is that we have many ways of doing so.
Reward Systems
7

T he carrot or the stick?

While the answer to this age-old question depends on the situation, it is gener-
ally accepted that one is meaningless without the other. If carrots are the norm,
they will not be seen as anything special, just a fact of life, no more special
than other omnipresent elements. Alternatively, if the stick is the only thing
available, it ceases to have any motivating power; it just offers abusive coercion
and does not foster any innate motivation. Both examples are extremes that
unfortunately do happen in real life and in most cases have been shown to be
deficient. If you work somewhere where your boss only screams at you and
never praises your achievements,1 chances are that you will leave. If you are a
runner and never lose a race, there is not much incentive to train to become
even faster.
There is, however, a third way. If somebody who is holding all the carrots
beats you with a stick, it would be very satisfying if you were to wrestle the stick
away from your tormentor and make him give you all the carrots. I know which
way I would prefer.
The following chapter will look at some fundamental principles of reward
systems and positive or negative motivators. Some of these occur within an ev-
eryday context, while others are more specialized or unexpected. Eventually we
will seek ways in which to use them in our levels in a manner that is effective and
entertaining, rather than there being something as crude as being hit repeatedly
with a stick over some measly carrots.
1
Not unheard of in the game industry.
95
96 Emotional Feedback Systems

Concept

Reward Systems and Schools of

Behaviorism
Like most mammals, human beings are fairly predictable in their reaction to re-
wards and punishments, insofar as that they have been studied extensively in this
regard, and behavior tends to fall into a number of typical2 responses. Without
wanting to go into a detailed exploration of behaviorism and its many schools
of thought, I would like to suggest that it covers a lot of ground that relates to
level design. A key factor in level design is the use of rewards and punishments
to enforce (or reinforce) or discourage certain behaviors. This is a massive sub-
ject, and one that is laced with controversy and difficulties. I suggest that readers
find their own path through subjects such as operant conditioning, intermittent
reinforcement, Pavlovian reactions, and other related subjects.3
For the purpose of this book, I think it is enough to look at a number of
reward systems that apply directly to its subject matter. As a preamble to their
uses in level design, I would like to shine a general light on the following reward
systems.

Escapism and Wish Fulfillment

There are many reasons why we turn to art or entertainment: enlightenment,
a need to expand boundaries, a quest for an alternative point of view, or other
equally valid motivations. However, sometimes all we need is a certain amount
of wish fulfillment, sometimes referred to as escapism. Generally these words are
not used as terms of appreciation; indeed, they are often used as terms of ridi-
cule. “That book is just an escapist fantasy” or “The movie has nothing to offer
but flights of fancy and empty wish fulfillment.” It is a big mistake to do so. Crit-
ics who employ this kind of language may well be right about the artistic merit
of the work they are reviewing, but they tend to make a grave error by mistaking
the effectiveness of the artist with the method employed, or even the goal of the
work. In other words, the problem is not that the work is escapist (one can insert

2
Insofar as this is possible when it comes to the often irrational behavior of people.
3
“Behaviorism,” Wikipedia, http://en.wikipedia.org/wiki/Behaviorism, 2009.
Reward Systems 97

other similar concepts); the problem is that the artist has not been successful in
providing enjoyable (or artistic) escapism.
Nobody would criticize a person for going on a long walk in the countryside
or reading a travelogue (clear escapism). There are no problems with somebody
listening to music from another country and in the process being transported
there (obvious wish fulfillment.) Who begrudges somebody reading Tolkien’s
The Hobbit (a flight of fancy)? Tolkien himself had the following to say on the
exact same subject:
I have claimed that Escape is one of the main functions of fairy-stories,
and since I do not disapprove of them, it is plain that I do not accept
the tone of scorn or pity with which “Escape” is now so often used: a
tone for which the uses of the word outside literary criticism give no
warrant at all. In what the misusers are fond of calling Real Life, Escape
is evidently as a rule very practical, and may even be heroic. In real life
it is difficult to blame it, unless it fails; in criticism it would seem to be
the worse the better it succeeds. Evidently we are faced by a misuse
of words, and also by a confusion of thought. Why should a man be
scorned if, finding himself in prison, he tries to get out and go home?
Or if, when he cannot do so, he thinks and talks about other topics than
jailers and prison-walls?4
The fact of the matter is that sometimes the here and now is not a desirable place
to be and can even be punishing, and that we can reward ourselves by engaging in
creative expressions that recognize this and provide an alternative. This has been
the motivation, and even the subject, of some of the most enduring pieces of art
in the history of mankind and should not be sneered at. To do so risks inviting
cynicism, closed-mindedness, and contempt for other people. Not only can es-
capism provide wish-fulfillment for the subject involved, it can also constitute a
form of rebellion: as a way of taunting the world by saying: “You can’t touch me
while I am in here!”

Submission and Release

There are times when we decide that it is all right for us to relinquish control
over a situation or even (mostly temporarily) our lives. This occurs when we
decide that somebody is able to teach us or help us in some way and we are told
that this can only occur if we follow the other’s lead. Although human beings are
naturally social creatures, this does not mean that they are naturally submissive.

4
J.R.R. Tolkien, Tree and Leaf, Allen & Unwin, London, 1964, p. 53.
98 Emotional Feedback Systems

There are of course many times when people espouse a group mentality, but that
is not what we are discussing right now. I am rather talking about those times
when people willingly, in exchange for some kind of reward, submit to others, or
at least take somebody else’s lead.
There are countless examples, both humble and significant. We can see this
principle at work in something as fanciful5 as ballroom dancing. There can only
be one person in that situation who leads. It also occurs in a very formal context,
for example in education. This does not only apply to children, who are naturally
much more submissive, but also to adults. For example, driving instructors or
music teachers must be listened to, or they can’t do their work.

Consensual (Temporary) Submission

In the case of most adult examples, the submission we are talking about is
purely consensual. Nobody is forcing you to submit to some arbitrary figure
of authority, yet it happens frequently and willingly. There are no punish-
ments for not submitting in this manner, unlike (for example) in the case of
submission to the law, which if refused, would lead to serious punitive conse-
quences. Instead, these are moments of subordination or submission that we
actively seek out, which is an interesting fact, considering that we would never
behave like this in other circumstances. If your music teacher told you to do
his or her shopping, you would laugh in that person’s face. This is because it
is a ridiculous request that yields no discernible reward to the person asked
to submit to this “authority.” It is ridiculous because there is no agreement in
place to do so.
In almost all similar cases there is an agreement, written or otherwise, that
exists between the parties involved. This agreement is essential and generally
quite clear in scope. We don’t need to see in a detailed contract that explains all
the legal parameters that state that the person blowing the whistle in a soccer
match is to be listened to.
You don’t run away from your guide in the jungle because he or she gets
cross with you for picking up some venomous lizard. We submit to this kind of
authority on a regular basis because there is an agreed-upon reward. We learn
how to play an instrument, enjoy a fair and enjoyable match of soccer, or enjoy
a safe and fascinating trip through the jungle. After the activity ends, temporary
submission is abandoned and authority is released.
Not only does this kind of system yield its own rewards, it can also be closely
tied in to other systems, and can in fact be the reason why we would even engage

5
Perhaps not the best word to describe it.
Reward Systems 99

in them. A good example of this will follow later in this chapter and discusses
catharsis.

Challenge and Empowerment

It is often asserted that hard work rewards itself, not just in a sense of achieve-
ment, but also in more practical ways. If we take off our cynical hat for a mo-
ment, we will realize that this is true in many situations.
There are positive challenges. If you go running for an hour every day, you will
become very fit. If you study a language diligently for ten years, you will become
fluid in its use. If your work is better than that of your colleagues, you will get that
promotion. (Well, you should, anyway.)
And there are negative challenges. Stand up to that bully and he will stop ha-
rassing you. Deal with the tragedy of a loss for a long enough time and it will hurt
less and less. Resist smoking for a year and your cravings will be gone.
The main point to recognize is that life is full of challenges, both positive and
negative, and that meeting them, sometimes even just attempting to meet them,
gives people a sense of control over their own life. This is a massively important
principle, which is often referred to as empowerment. The word is overused at
times, and it is often coated in sugary, feel-good language. But at the basic level, it
is nothing more than finding the personal power to take control of your own life.
This is a huge motivator for anybody, and many belief-systems, books, self help
tapes, motivational speaker’s careers, and other things have been based on it. The
reward of empowerment is so strong that it can even lead people into unhealthy
behavior patterns. For example, addiction is often, paradoxically, about control.6
Some people try to feel empowered by trying to control other people’s lives. But
in general, it is widely understood that reacting to life’s challenges positively, or
seeking positive challenges to master to ultimately gain a level of control over
life’s experiences, is a very positive endeavor.

Fairness
This principle only works however when life treats people fairly. To put our cyni-
cal hat back on: life isn’t always fair. The notion of fairness is a very interesting
one, as it is a strictly human concept. The wind isn’t fair, nor is gravity. Fire cares
nothing for fairness. Yet within our own lives, we attribute huge importance to
it. There are few things as annoying as someone jumping ahead of you in line, a
6
For example, gambling is often said to give addicts a sense of control by letting the ritual
control other aspects of their lives.
100 Emotional Feedback Systems

lazy millionaire underpaying hard-working employees, or losing your pension in

some financial corporate debacle through no fault of your own. Unfairness is a
form of disempowerment, and it should never be underestimated how much of a
negative impact this can have.

Investment and Payoff

Related to the expectation of fairness is a reward system, which has its roots in
financial principles. When we put our money in the bank, we expect to receive
interest payments as a reward for letting the bank hold it. If we invest in a com-
pany through buying shares or in some other way, we expect a return on our
investment in the form of a profitable payout. There are many other examples,
but these should suffice to recognize the principle. We invest our money in such
a way that we expect to receive some kind of reward at the end. This is a fairly
basic principle of financial systems. It extends to further principles, for example
to buying property in a country with a growing housing market, or starting a
business in a country which is experiencing economic growth.
These examples are of a purely financial nature. But the principle is equally
recognizable in other areas. For example, we can invest time into something and
expect something in return. Or we can invest effort and expect some kind of re-
sult. This can overlap with the principle of empowerment, but it doesn’t have
to. For example, we can invest time and effort to solve a difficult cryptic puzzle,
which yields a reward in providing a satisfying answer. Large investments offer
the expectation of large rewards. A scientist may dedicate his or her entire life
to finding a cure to a particular disease or other scientific problem. Although
it might not necessarily be unfair, it would be unfortunate if these investments
went without rewards.
Sometimes the rewards are immediate, (the activity itself is so enjoyable
that it is a reward in itself), or sometimes the rewards arrive over time, or it is
reached after sustained effort or investment. Sometimes both can occur within
the same activity.

An Example: Treasure Hunting

Almost every culture on earth understands the concept of treasure hunting. This
is clear from representations in books like Treasure Island,7 films like Pirates of
the Caribbean,8 legends like the Lost Gold of the Templars, and the iconic image of
7
Robert Louis Stevenson, 1883.
8
Published by the Walt Disney Company.
Reward Systems 101

the classical Pirate. Furthermore, in the real world there are thousands of people
who engage in amateur treasure hunting with a metal detector, as well as pro-
fessional treasure finders who organize well-researched expeditions in heavily
funded commercial projects.
Investment in the context of the treasure hunt can contain examples like
these:
• Treasure hunting investments:
■ financial investment in an expedition,
■ effort of researching a subject,
■ spending time waving a metal detector around,
■ hard physical work during an expedition.
• Immediate returns on investment:
■ enjoying the company of other expedition members,
■ gaining knowledge,
■ enjoying the outdoors,
■ feelings of excitement.
• Long-term returns on investment:
■ treasure, of course!
Romantic as this all sounds, there are real world treasure hunters who are
very successful. A famous example is Mel Fisher and his company Salvors Inc.,
whose most famous expedition led to the discovery of the long-lost Spanish gal-
leon Nuestra Señora de Atocha,9” whose wreck was the home of an enormous
hoard of treasure:
Among the items found on the wrecks are a fortune in gold, silver
bars, and coins destined for the coffers of Spain; a solid gold belt
and necklace set with gems; a gold chalice designed to prevent its
user from being poisoned; an intricately-tooled gold plate; a gold
chain that weighs more than seven pounds; a horde of contraband
emeralds—including an impressive 77.76 carat uncut hexagonal
crystal experts have traced to the Muzo mine in Colombia; religious
and secular jewelry; and silverware.10
This shows that treasure hunting is an excellent example of an activity with a
multi-layered inbuilt reward system. Although treasure hunting is in many ways
an obscure example, everybody can understand the reasoning behind it. It is an
activity that can act as a nearly perfect metaphor for the principle in general.
9
Our Lady of Atocha.
10
“1622 Feet,” Mel Fisher Maritime Heritage Society and Museum in Key West, http://
www.melfisher.org/1622.htm, 2008.
102 Emotional Feedback Systems

Isolation and Social Reinforcement

People are social creatures. This may not always seem obvious, considering how
badly people can treat each other, but nonetheless, almost nobody can enjoy
extended periods of social isolation. We really do need some minimum amount
of social contact to successfully experience our humanity. This principle is so
strong that it is represented in fundamental instincts. Take, for example, human
childbirth: it is well known that it is beneficial for child and mother to bond as
soon as possible after birth. This process has been linked to the hormone oxy-
tocine in women, which facilitates this bonding.11 Regardless of the scientific
explanation of any hormonal grounding, it is hard to argue that social reinforce-
ment isn’t one of the most important driving forces in our lives.

Social Reinforcement
The need for social systems does not stop with the child-mother bond. When
we grow through life stages, we feel the need to be part of a family unit. The re-
wards are clear and easy to recognize. Through social bonds, we receive support
or companionship and guidance. If this cannot be found in the family, people
sometimes join another group to make sure they receive it nonetheless, like the
army, or an ideological group of some sort. At times this need is so strong that
people will join groups that are clearly abusive, like some cults. This shows how
powerful the need for social reinforcement is.
In addition to these examples, we often join clubs, go to communal events,
start our own families, and form complex social networks. And, most relevant to
this book, we seek out enjoyable activities that are based on or reliant on social
interaction. Have you ever been to a sci-fi convention? It is remarkable how so-
ciable huge groups of people can become when they acknowledge that they have
something in common. This can be seen in events where people just enjoy each
other’s company, when they share a common interest12 or goal, or when they
want to compete against each other in an enjoyable way, like at sporting events.
It is hard to deny the existence and the plentiful awards that all of these social
interactions can offer. It is therefore not surprising that they are all around us and
we engage in them on a regular basis.

11
Catherine West, “Level of Oxytocin in Pregnant Women Predicts Mother-Child
Bond,” Aps Observer 20:10 (2007), available at http://www.psychologicalscience.org/
observer/getArticle.cfm?id=2245.
12
Perhaps I should start a level design appreciation society?
Reward Systems 103

Temporary Isolation
This reward system can also work inversely, strange as that may sound. Who
hasn’t heard the grim saying, “Hell is other people”? Although not a very nice
thing to say, it touches upon an important truth. Many of the advantages of a
strong social network can also cause serious distress. Sometimes all the social el-
ements in our lives can become oppressive. Too many opinions vying for domi-
nance, attempts at peer pressure, social control instead of social support, or even
just the fact that sometimes we don’t get along with some other people.
At such moments, we may decide that the best way to improve our situation
is to engage in activities that enforce temporary13 solitude, or at least minimize the
number of people we have to interact with.
When we need “peace and quiet” we may go on a long walk in the country-
side. It allows us to focus on our own thoughts, without the din of others disturb-
ing them. There are many other examples, like painting or writing a book. We
may seek an activity that, in solitude, lets us focus on the voice or on the ideas of
others. Listening to music or watching a film, and even doing a crossword puzzle,
come to mind. As prevalent as social interaction can be, we enjoy a comparable
number of solitary activities that produce their own vital awards.

Adversity and Catharsis

It is an unfortunate but nonetheless potent truth that some rewards can only
come from dealing with negatives. This is complicated at best and covers a wide
range of topics. I have dedicated a whole chapter to the principle of using nega-
tive emotions (Chapter 9),14 which goes into some detail about using negative
emotions in a positive way to strengthen level design. A number of specific tech-
niques will be discussed and illustrated by clear examples. Within this chapter,
however, I want to spend some time on the principle of catharsis as a reward
system.
Catharsis
• Noun the release of pent-up emotions, for example through drama.
■ Derivatives cathartic adjective & noun.
■ Origin Greek katharsis, from kathairein ‘cleanse’.15

13
Almost nobody is interested in permanent solitude. That would be throwing out the
baby with the bathwater.
14
Chapter 9, “Negative Emotions.”
15
From the Compact Oxford English Dictionary. Available online at http://www.askox-
ford.com/concise_oed/catharsis?view=uk.
104 Emotional Feedback Systems

Catharsis is a special case reward system because it is often one that we

wouldn’t seek out intentionally, because it cannot be achieved without also ex-
periencing negatives beforehand. The concept is almost always described as hav-
ing something of a purging nature, all the way back to Aristotle’s early use of the
word in his famous Poetics, where he describes catharsis as being one of the main
characteristics of tragedy, where it serves as a purgation of pity and fear.
Furthermore, there are definitions of catharsis that are just as valid but come
from psychology, religion, or medicine. Whatever the interpretation, though,
a defining feature of the principle of catharsis is that its reward can only come
about after one suffers a serious amount of pain or adversity and feels the purg-
ing nature of the subsequent emotional release. This is why it is an interesting
reward system, since it feels counterintuitive to seek out negative experiences
to receive a positive reward. Indeed, in many cases we don’t do this and in fact
try all we can to avoid it. But on further inspection, we find that in most societ-
ies there is actually a deep-rooted respect for doing exactly that. In the English
language, this can be seen in phrases or sayings like “character building” or in
the slightly more testosterone-tinted observation that some kind of adversity
“makes a man out of him.”
All over the world, people engage in initiation ceremonies, or in ritualized
rites of passage. An almost uncomfortable number of coming-of-age ceremonies
are associated with painful or difficult tasks or quests. Ultimately we find that
the basic principle of coming through a painful or difficult experience to learn or
experience something positive pervades society on all levels. Uncomfortable as
it may seem, negative experiences can provide rich rewards. This knowledge is
formalized throughout all aspects of society. Catharsis is the basis for all kinds of
reward mechanisms, some of them formalized and accepted, others only recog-
nized after the fact or even coerced. Thank goodness that the latter is more and
more frowned upon, although it still exists in practices like initiation ceremo-
nies. Other examples can be found in activities that take serious sacrifices from
the participant, sometimes in ways that seem extreme to others, but that give
rewards that otherwise cannot be attained. People have spoken of life-changing
experiences after extreme challenges like mountain climbing or spelunking16 ex-
peditions, extremely long periods of meditation, or extended fasting.
Plenty of other examples exist, but they all require a real extended effort
from the participant. It is this extended aspect of enduring hardship that in most
cases is linked to the positive outcome, the sense of catharsis. It can only occur,
however, if the participant is tested in such a way that failure to endure is a possi-
bility. And therein lies the danger. Paradoxically, catharsis as a basis for a reward

16
Cave exploration.
Reward Systems 105

system can only work well if there is this element of danger of failure or at least
of real challenge. Otherwise, the emotions released at a positive or negative17
outcome will not be strong enough to trigger a real sense of catharsis.

Concept Summary
What most of the reward systems and structures described above have in com-
mon is that they demand a certain amount of work, sacrifice or investment from
the subject. Something is given and something is received. It is not necessarily true
that more effort equals a larger reward (although that is often the case). It is
more a matter of making a task worthwhile and meaningful because seeing one’s
efforts rewarded is fulfilling and enjoyable. This sometimes means that a very
small amount of effort can reap a large reward, or a grueling task may end in a
small symbolic victory, yet the feeling of fulfillment can be equally strong. Often the
joy simply comes from what amounts to solving a life puzzle or conundrum, or
it revolves about learning and applying the right skills for a task.
Unfortunately, there is a flip side to all this, insofar as unfulfilled reward sys-
tems provide negative emotional feedback. At times this can be of comparable
impact to the expected positive outcome. It can also demotivate people to en-
gage in important aspects in their lives, if through thwarted reward systems they
have been taught not to do so. What use is love if it is likely to end in tears? Why
work a hard job if the promotions only go to people who cheat? Why play a game
when any effort can be instantly wiped out by the level design?
One of the many things a level designer has to understand is the dynamics of
all these reward systems, as well as being able to effectively use them in practical
situations, in a fair and even-handed way, if possible. The following chapter will
attempt to take these concepts of diverse reward systems and examine them in
the light of level design theory.

Theory
If somebody holding carrots beats you with a stick it would be very
satisfying if you were to wrestle the stick away from your tormentor and
make him give you all the carrots.

17
Catharsis can still occur even if the participant fails at the attempted task. It is the final
release of emotions that shapes the sense of catharsis.
106 Emotional Feedback Systems

The previous chapter discusses many of the possible ways in which reward sys-
tems and structures can work, and, as importantly, what expectations people
have from life. In this chapter we will examine some of them again, but within
the context of level design theory. We will derive or construct some game-spe-
cific principles, as well.
I would like to make an important initial point before we do this, however,
What has to be clear from the outset is that as authors of a level’s content, and
therefore of much of the player’s game experience, we are completely within
our rights to manipulate the player’s reward expectations in ways that are unex-
pected, but that are ultimately more rewarding to the player. Even though it falls
squarely within our responsibilities to do so, this is often forgotten, or worse,
ignored. As long as we don’t violate other important principles of level design,
we have the power to give the player the opportunity to finally take away the big
stick from the universal tormentor and run away with all the carrots, and feel re-
ally good about it.

Escapism and Wish Fulfillment

Level designers are lucky that they work in a medium that has an audience that
is already willing and able to submit itself to feelings of escapism and wish fulfill-
ment. It is perfectly acceptable for a gamer to crave these things; it is even (er-
roneously) at times expected that a gamer will exclusively crave them. This makes
us lucky, because it gives us an audience made up of people who are willing to
travel with us and gives us a whole range of techniques that we can use to get
them there.
We can do this by taking them somewhere that doesn’t exist in this world
and then delight them with amazing new sight and sounds. Fantabulate!18
Or, we can take them into our interpretation of an enjoyable activity based
in real world concepts. Simulate!
Both have many areas where they overlap, as covered in many chapters else-
where in this book. But they also have important differences that need to be
examined. (There is also a third option that will be examined later.)

Fantabulate!
A level is a virtual construct. It may have its own set of rules, logic, physics sys-
tem, ecology, and other internal systems, but they all take place in a virtual set-
ting. The disadvantage of this virtuality is that somebody needs to design and
18
If this is not a real word, it should be; and I am claiming it now.
Reward Systems 107

implement all of these things. The advantage for level designers is that in this
need, or put to put it differently, in this license to do so, lies a huge amount of
freedom and power. In game levels with a non-realistic setting, the level designer
has the license to fantabulate.

New rules
Within a fantastic virtual construct, we are free to create many things or situa-
tions that simply would not work in the real world, all with the approval of our
audience. Not only may we invent these new rules; we are positively encouraged
to do so. We have already concluded in Chapter 2, “Teaching Mechanisms,” that
part of our responsibility lies in teaching the player the rules of the available
gameplay activities. In a fantastic setting this is especially important, as the rules
may be unknown in real life.
For example, we may have to teach players that summoning a fire elemental
is an extremely effective way of deterring packs of ice wolves from attacking19.)
The real fun lies in the fact that players who enjoy these kinds of things, and
there are many, many millions of them, also really enjoy learning about this new
world they find themselves in. Within an escapist mindset, experiencing new
fantastic concepts is an attraction in its own right. If we go back to our earlier
example of Tolkien, we see a work of fantasy that partly excels because of its
sheer scope of invention. The book creates a very deep sense of wonder, partly be-
cause it consistently and thoroughly showcases a new world with an extremely
detailed and well-thought-out set of rules. This applies to almost anything in the
world, including its history, its ecology, and its magic system. Reading about all
of these things is a large reason for the success of the book.
Level designers have to do the same thing. They need to interpret the new
rules of the world and teach them to players in such a way that it creates a great
sense of wonder, as well as teaching them how to play the game.

New environments
Hand in hand with new world rules come new environments; and once again,
a great amount of work and a great amount of freedom for the level designer. A
great amount of work because within this virtual construct somebody has to do
the constructing. This does not mean that the level designer has to create all the
environments solely by himself or herself, although at times this is feasible or
necessary. In most cases, this work must be done in partnership with the art de-
partment. But the level designer does need to design all the gameplay space, and
the way it is used. This gives level designers a great amount of freedom because
19
Yes, I picked this cliché on purpose, for illustrative clarity.
108 Emotional Feedback Systems

they are the authors of a new gameplay environment, and to a large degree, of a
whole new gameplay world. This is one of the reasons why level design is such an
enormously fulfilling profession; it literally gives a designer the power to create
worlds.20
So far, so good. In fact, this is no different from most other forms of level
design.
Where level design featuring themes of escapism and wish fulfillment in a
fantastic setting differs from more reality-based design is in some of the intrinsic
goals. A big reason for the existence of the levels is to present a gameplay envi-
ronment, and a virtual environment that appeals enough in its own right that the
player wishes to engage with it and spend time sampling its content. This means
that it is reasonable to include enjoyable areas that don’t feature much gameplay
(but aid in escapism), or to go further and assert that exploring these areas is
part of the gameplay appropriate to the goals of such a level. A big mistake that
many people in game development make is to assume that all gameplay spaces
must feature active challenges and encounters. It is actually important to also
include gameplay space that celebrates escapism through the medium of explo-
ration, or other ways that the player can just enjoy the world. These are some of
the reasons that so many games feature a fantasy, sci-fi, surreal, or otherwise fan-
tastic setting. For many reasons, these genres are especially suited for this kind
of design.
Providing the player with many level design scenarios to achieve these goals
is an important way to allow for deep and interesting elements of wish fulfill-
ment and escapism. Level designers should always ask fundamental questions
about the scenarios they create. In the case of a fantastic setting, these questions
can include:
• Can I reach that strange but beautiful area?
• How do I study that new dangerous creature?
• What else can I use this artifact for?
• Who built this structure?
• How do I reach that floating fortress?
These are just a few random examples, but each one shows that interesting level
design scenarios are just around the corner. And answering questions like these
goes very far in providing the player with what he or she wishes for, and consti-
tutes an effective use of a powerful reward mechanism.

20
If you are interested in level design and that doesn’t appeal at a very basic level, you
may ask yourself some questions.
Reward Systems 109

Simulate!
A completely different approach to escapism and wish fulfillment is found in
simulations, and before we discuss level design theory and simulations, we should
actually look at what is meant by the word.

Simulation and imitation

Normally, when we speak of simulation, we are talking about modeling a real-
world system or situation in order to learn something new. This could be for
scientific reasons; for example, a simulation and study of hunter predator cycles
could be used to warn when a particular species becomes overhunted and may
become endangered. It could also be for financial reasons; a simulation of a par-
ticular economic system may predict which factors contribute to inflation. In
any of the examples we can find of simulations, it is generally the case that there
is a need for accuracy in order to correctly extrapolate from the data that the
simulation produces:
[A] simulation results when the equations of the underlying dynamic
model are solved. This model is designed to imitate the time-evolution
of a real system.21 (Emphasis mine.)
Most games are not like that at all. (The exceptions will be noted shortly.)
Games are all about enjoyment. When we play games, we play them for all kinds
of enjoyable reasons: to have fun, to exercise our brains, to have a meaningful
artistic experience, and so forth. Simulation games are no different and exist
to provide an enjoyable experience, in most cases by providing players with a
chance to engage in a real-life activity they normally would not be able to enjoy.
A game can offer a player a chance to be a soccer manager or a train conductor
or a theme park operator.
These are great examples of games based upon wish fulfillment as a reward
system. If we look closer, we find that they aren’t games of simulation at all, but
games of illusion and imitation. The game imitates real-world activities only to
the degree that their fun aspects are replicated for the enjoyment of the play-
er. This kind of imitation is, unlike practical simulations, not concerned with
accuracy at all, but with the appearance of accuracy. The games would quickly
become extremely tedious if they tried to accurately simulate all aspects of the
activity in question.

21
Stephan Hartmann, “The World as a Process: Simulations in the Natural and So-
cial Sciences,” http://philsci-archive.pitt.edu/archive/00002412/, 2005.
110 Emotional Feedback Systems

Accuracy only needs to be observed as long as it supports enjoyable game-

play. An actual racing game simulator (SIM) would be far too difficult for most
gamers to enjoy. And what is the point of playing a grand prix SIM if the player
cannot win? It would be accurate and realistic, but not much fun, especially be-
cause it fails at the first hurdle and doesn’t provide the wish fulfillment element
of the game’s reward system.
Strangely, although the use of the word simulation is suspect, I still advise
that we adhere to its usage in games. It is simply too confusing to do otherwise,
because as a description of genre, it is too widespread to change. It is essential,
however, that level designers know what simulation games are really about: en-
joyable imitation.
This is not semantic nitpicking, but a fundamentally profound difference
that causes much debate and conflict. Almost every level designer, on a regu-
lar basis, will have to argue this point against somebody who insists on making
gameplay decisions that fail this test of enjoyable imitation, solely based on the
argument that the game needs to correctly simulate a real world event. There are
times when simulation and imitation go hand in hand, perhaps when a particular
sport’s league is implemented, or when the correct dimensions of a vehicle need
to be followed. But even in those circumstances, it needs to be clear that these
implementations still serve an enjoyable imitation of a real world activity.
This has tremendous impact for the level design of these games. Instead of
being at the mercy of real-life rules and physics, the level designer now has the
role of illusionist. The tracks in a realistic racing SIM now only have to feel like
they are correct; as long as they are fun, the job is well done. The lack of accuracy
in a wildlife photography game’s terrain means nothing as long as it produces
expected results that don’t break immersion. It is all smoke and mirrors.

“Serious games”
There are only few exceptions to this rule, mainly in the area of so-called serious
games and educational games. They are noteworthy because although they can
display many of the characteristics of other video games, they are fundamen-
tally different. There is no formal definition of what exactly constitutes “serious
games”, but it is fair to say that their main focus is that of teaching some real-
world application or education. This can be a commercial focus, for example a
driving game for a driving school, or a scientific one, for example a game that lets
students identify certain plants as part of a biology lesson.22 As already noted in
Chapter 2, games are extremely suitable as a teaching tool, since we are already
trained at a very young age to engage in gameplay in to learn all kinds of diverse
22
Serious games are nearly always educational games.
Reward Systems 111

skills. This and the ever-improving technological sophistication of commercial

video games have led to a proliferation of serious and educational games that
recognize this principle.
For example, see the Serious Games Initiative, which has done much work
in this arena:
The Serious Games Initiative is focused on uses for games in exploring
management and leadership challenges facing the public sector. Part of
its overall charter is to help forge productive links between the electronic
game industry and projects involving the use of games in education,
training, health, and public policy.23
Because the defining aspect of these games is that of real-world application,
they are always expected to produce tangible results, or they will have failed in
their basic function. And this result has to be realistic or accurate at all times, or
its real-world application will be ruined. Because of this, the player has to be able
to trust the game to produce teaching material that is trustworthy and cannot
just be an imitation or an illusion. A serious game teaching somebody how to fly
an airplane in real life has some serious responsibilities in real life to live up to.

Fantastic Simulation
On rare occasions, a hybrid game form appears that tries to provide fun game-
play through “realistically” simulating an activity in a fantastic setting, or even a
fantastic activity in any setting.24 This is a strange beast indeed and initially is not
easy to quantify, but some clear examples exist. Take for instance Startopia,25 a
game that expects the player to successfully run a spaceship colony, balancing
the needs of all the diverse onboard species. Another example can be found in
the famous Tamagotchi brand, where the player is expected to take care of a fan-
tastical creature in a realistic manner.
These games still contain the key elements of a rewards system based on
escapism and wish fulfillment, but it is up to the level designer to decide where
to fantabulate and where to simulate. However, the question can be asked: how
can a game simultaneously be both fantastic and realistic?

23
“The Serious Games Initiative,” The Serious Games Initiative, www.seriousgames.org/
newswire/.
24
Typically a fantastic setting, however.
25
Mucky Foot Productions, 2001; published by Eidos.
112 Emotional Feedback Systems

Staying “in character”

The answer to this question lies in the assumption that a level should stay in
character. Like an actor, the game cannot acknowledge the world outside of its
own fiction. If this happened, it might not be strange for a player to take slow in-
cremental lessons in hover board control to perfectly learn the nuances needed
to enter the Martian Circular Race.26 The level designer needs to be aware that
although there is room for imitation and illusion, the levels cannot cheat the
in-game rules at any time. I will leave this topic for now, before it all becomes
too metaphysical, but I would like to advise any level designer working on such
a game to treat the fake rules of the game as if they were real.

Some Further Notes on Wish Fulfillment

In most of the examples and cases discussed so far, wish fulfillment has been
linked to giving players the freedom to engage in activities they probably can’t
in real life. This can be to shine in a career as a formula one racing driver or to
captain a star ship. The activity itself is the wish being fulfilled. The principle
goes much further, however, sometimes in unexpected ways. The player may be
confronted with a fast vehicle, leading to a wish to drive it, or the player may spot
a castle on the horizon, leading to a wish to reach it. Many of these kinds of sce-
narios are actually in the hands of the level designer. In wish fulfillment, we have
an immensely powerful tool to entertain the player through our level designs.
In this context, wish fulfillment means adding gameplay scenarios that create a
desire and eventually give the player the means to satisfy it.

Avoiding clichés
A well-known criticism of wish fulfillment is that it panders to simplistic desires
and that is “too easy,” leading to cheap entertainment that doesn’t challenge or
engage the audience enough. This danger certainly exists, but it is no more a re-
sult of wish fulfillment than elevator music is a result of making music accessible.
If used well, wish fulfillment is a powerful technique that can be used to reward
gameplay, deepen immersion, and to challenge the player’s conception of what
a desired outcome is. It is up to the level designer to decide how to implement
these principles, and what clichés to avoid. There are no hard and fast rules and
what constitutes a cliché can be entirely dependent on the game’s genre or ex-
pected audience.

26
I made that up.
Reward Systems 113

Investment and Payoff

Players of massively multiplayer online role playing games (MMORPGs) know all
about this one. In fact, they have a rather negative term for when a game becomes
too much of a chore — “grinding.” The grind describes a painful and boring slog
where players perform repetitive gameplay tasks27 that very slowly increase their
experience points, eventually allowing them to level up,28 which gives them more
and more of the power needed to make a real impact in the world.
Sounds terribly boring, doesn’t it? Yet millions of people do this day after day
without fail, simply because there is a payoff for all this investment. They may
hate doing it, but they do it anyway if they perceive the payoff to be worthwhile.
However, we can do much to reduce the tedium of this kind of gameplay with
smart level design. I will focus more on avoiding the grind later. But first, let’s
look at the principle of investment and payoff in level design.

The Agreement between Designer and Player

Throughout this chapter, and indeed throughout this book, I have made it a re-
curring theme that artist and audience strike a deal. The player of a video game is
happy to face the challenges the game offers — if there are rewards for doing so. It is
tempting to leave it at that; but because it is such an important principle, let me
try to illustrate the concept in a little more depth. As in most other level design
techniques, it is important to get the balance right. In the dynamics of invest-
ment and payoff, the two sides of his bargain need to be balanced. If the reward
is not high enough in relation to the investment, the player will feel cheated. If
the reward is too high, the player will become spoiled and will expect too much
throughout the game afterwards. If the challenges are too hard, the player will
become frustrated; if they are too easy; the player will become bored. So, a few
principles need to be respected:

Proportional rewards
Rewards need to be proportionate to the effort expended in obtaining them. Or
at least, the player needs to know that the potential awards can be proportionate
to the effort needed to obtain them. There are exceptions, but they are excep-
tions to the rule. If we ask players to abide by certain rules, they need to be able
to trust the contract and the fact that they will receive a reasonable reward for
27
Like vanquishing small furry creatures or picking berries for days of in-game time on
end.
28
Attain higher levels for their character.
114 Emotional Feedback Systems

their efforts. Even when there is an element of chance in play, this principle stays
intact, as the potential award needs to be high enough to compensate for those
instances when the player receives a lesser reward, or none. But in that case, the
player needs to know that chance is a factor.

Avoid boredom if possible

Try to avoid situations where the player ends up repeating the same boring task
over and over for small incremental awards. I have already mentioned “the grind
earlier in the chapter, but it can occur in other ways as well. A good example
can be found in enemy placement. For example, in a first person shooter, if the
same enemy guards every pickup, the task of dispatching that enemy to gain the
pickup becomes very boring, very quickly. Constantly repeating the same task
for the same gain is a sure-fire recipe for boring the player witless. Instead varia-
tion needs to added, by diversifying the challenge, even if the basic components
stay the same. The challenge can be repeated but possibly with variations in
• enemy numbers,
• enemy arms,
• strategic positioning,
• player arms,
• pickup location,
• time limit.
And so it goes. There are uncounted ways of adding variation of this kind. Doing
so allows the level designer to scale content and difficulty level to a large degree.
This can go quite far and can allow the player to be taught different ways of ap-
proaching the same problem. (How to get the award.) This makes it possible to
stave off player boredom with a repetitive task, while at the same time teaching
new gameplay mechanics.

Avoid frustration
Don’t promise the player a fat juicy carrot and then just end up repeatedly bash-
ing that player with a stick. This is another instance where rewards need to be
balanced against the effort needed to obtain them. It is frustrating if it is too hard
to get a gameplay award. And there is a point where any reward is too little for the
enormous effort needed to obtain it. It is worth noting that many level designers,
especially those just beginning, err badly in this department. Challenge levels
don’t scale indefinitely; in fact, quite early on, difficulty starts to become a turn-
off, unless skill keeps pace with it. If your level design is based upon a concept
of elite Yoda-master level skills, you have effectively turned off the majority of
your players. These are players who will probably never play your game again, and
Reward Systems 115

they will tell others not to buy it. It is simply not in the contract between level
designer and audience that the audience can only enjoy the game after unbeliev-
able skill levels have been reached. Or at least this is the case in most games.
There are exceptions, like certain types of shoot-em-ups where this difficulty is
part of the core gameplay. And even in these games, there is still an acceptable29
level of difficulty and an unacceptable one. Generally, it is better to think of inter-
esting challenges, rather than of difficult ones. Difficulty is just one of many ways
to keep a challenge interesting.

Submission and Release in Level Design

Earlier in this chapter, we saw that at times people are happy to submit to an
authority they normally would not recognize.

Application
So far so good, but does this have anything to do with level design? Well, sur-
prisingly, it does. This agreement between parties to temporarily allow a reward
system of submission and release is clearly recognizable in the contract between
artist and audience. Don’t we submit to the whim of the writer or the wishes of
the movie director? Even in the non-passive context of a video game, we still
submit to the same principle, because there is the promise of a reward on the
other end. Sometimes we can be held or a long time, without complaining, in a
state in which we normally would not find ourselves, because the artist is assured
and strong in his or her craft. I will return to this principle of an agreement, or
contract between artist and audience, several times in this book, as it can lead to
useful applications in creative expression, and particularly in level design.

Trust in the Machine

The most important aspect of this dynamic is the fact that players are happy to
be led by the level designer if they trust the design not to betray them. This can
be overt — players know and acknowledge that they are submitting. Or it can be
covert — players are not aware that a direct effort in this regard is being made.30

29
Chapter 8, “Immersion,” will explore in greater detail the required balance between
difficulty and skill levels through an examination of “Flow Theory.”
30
Other than the fact that they are aware that they are playing a game.
116 Emotional Feedback Systems

Overt submission
In the case of overt submission, trust is lost if the situation does not result in some
tangible bonus or reward. (The mission needs to end in some form of success,
or at the least in an honest chance to succeed. A tutorial needs to teach, rather
than punish; and collecting all 100 gems of infinity needs to really pay off.31 In
many ways this is the same principle as discussed earlier in the area of investment
and payoff, but it is still a useful alternative way of looking at reward systems. In
this particular case, the player is aware that the game offers specific awards for
specific behavior, and related gameplay tasks are actually represented as such.
It would be a huge mistake to allow a situation where the player thinks that a
certain outcome is guaranteed, only to find out that this no longer holds true.
From this point on, trust is lost, and the player will not be able to make informed
gameplay decisions.

Covert submission
Covert submission to authority is more widespread than most players realize. A
lot of it is planned out early in the level design stage, since it is intended for cer-
tain things to occur, no matter what the player decides.32 If players do not realize
this, or if they agree not to acknowledge the fact that they are forcibly led to
certain conclusions, chances are they won’t resent it.
For example, a level designer may wish to teach the player how to use a cer-
tain play mechanic and therefore have devised a number of artificial encounters
that act as a tutorial. Players may think they are following the request of a villager
in need, while in actual terms they are gaining enough experience points to be
able to wield the sword they will receive at the end of the mission. In order to
present formal gameplay challenges like this in such an informal way, the chal-
lenges have to be presented through the voice of the game itself, without ac-
knowledging the formal real-world goal of gaining X experience points or some-
thing similar. It is generally a good idea if the game “stays in character” and does
not tell the player what the real reason for the task is.
Covert submission is less direct than overt submission, and therefore it is
less easy to define what a betrayal of trust means. There are some definite danger
areas, however. It is often a bad idea to present the player with any of the follow-
ing situations:

31
Perhaps by providing a nice porcelain dog? (I am kidding.)
32
This is why, although it sounds paradoxical, it makes sense to classify this as submis-
sion. Chapter 10, which covers topics like immersion and suspension of disbelief, goes
into great detail on how this works.
Reward Systems 117

• Don’t repeat the same covert task too many times. The player is sure to
catch on sooner rather than later.
• Don’t arbitrarily use this technique. It is much easier to pull off when there
is a good in-game reason for doing so.
• On completion of the task, don’t inform the player that it was all just a ruse.
The player will feel manipulated and resent it.
These are just some general examples. There are many others, all of which de-
pend on genre and conventions.

Empowering the Player

Unlike real life, games provide us with an opportunity to face challenges without
penalties that go beyond the limits of the game itself. This is a real advantage,
since it allows the player of a game to experiment with different strategies for
overcoming challenges, or repeating attempts at solving the same problem. It is
entirely possible for a player to face initial impossible odds, yet through repeti-
tion or experimentation, manage to finally overcome them. As level designers
we are uniquely placed to provide the player with interesting challenges, as well
as the tools and means to successfully overcome them. In other words, we can
design levels that are geared towards player empowerment.
As a reward mechanism in real life, this is hard to beat, and it is no differ-
ent within the virtual setting of a game. The difference in a game setting is that
as level designers we can shape the actual world and the contained gameplay
scenarios. We can lend a helping hand by rigging the game in favor of the player,
eliminate much of the fairness inherent to real life, and provide players with mul-
tiple chances to overcome the same challenge. And we can determine much of
the final reward offered to the player, sometimes including the type of empow-
erment that the experience rewards. To be able to wield these godlike powers
wisely, we should examine these advantages a bit farther.

Shaping the World

One of the most fundamental acts that a level designer undertakes is that of cre-
ating the environment. It is literally a matter of world building and is therefore an
area that has far-reaching consequences for gameplay. Crucial to this act of cre-
ation is the design of the player’s role in the world. How does the player interact
with the world’s dynamic systems? How much control is given to the player to
affect the environment? How effective are the powers that the player wields? The
answers to these questions paint a unique gameplay experience for every indi-
118 Emotional Feedback Systems

vidual game, from Tetris to Half Life, and what they all have in common is that
they can empower or disempower the player in a variety of ways.

Rigging the game

Life’s challenges tend to be fairly random. They just pop up, and we have to cope
with them, regardless of our ability to do so. As level designers, however, we can
tailor the players’ challenges, and we can have much to say over their ability to
deal with them. In most cases, we can and should make sure that the fix is in, and
that the game is slightly rigged in favor of the player. The player finds just the
right keys to open just the right doors. The player has just enough training to be
able to just win the tennis tournament. The ambush area features just enough
cover for the player to make a fighting escape. Often, we can engineer these mo-
ments in such a way that the player makes the key decisions, or is tricked into
thinking so,33 thus deepening the sense of empowerment.

Keeping it fair
Most people are painfully aware of the fact that life is not inherently fair. We want
it to be, and sometimes the outcomes of our life’s challenges are pleasingly posi-
tive, but just as often they simply aren’t. This is why we often seek out challenges
that can guarantee a fair outcome. Video game challenges fall in that category. As
designers, we can facilitate players in this wish by making sure that we keep the
challenges they face firmly in the realm of fairness. We have the power to create a
world where taking action does mean taking control of one’s destiny.

Allowing the player to cheat

Although we can add a sense of fairness to gameplay, on the other hand, there is
nothing wrong with us allowing the player to cheat. I don’t mean cheating in the
sense of breaking existing rules, and we can’t let the player rewrite code. But we
can present situations, such that the player can approach them in ways that the
real world doesn’t allow. For example, say we create a jumping puzzle where fail-
ure means that players will fall a long way down, normally to their death. How-
ever, nothing prevents us from placing a trampoline at the bottom of the drop,
which bounces the player back to the start of the jumping sequence. We might
engineer a fight sequence to take place in a location where the player is less vul-
nerable and can approach the fight from a position of strength that is normally
absent, perhaps by giving the player access to a shielded vehicle. We can even
engineer situations where the player can attempt to do something like stealing

33
I will return to this theme of benign deceit several times throughout the book.
Reward Systems 119

an item from a museum and make the price of failure much lower than in reality.
In this case, tripping an alarm can be temporary, and the museum security would
reset after five minutes, no matter how many times the player trips the alarm.

The final reward

The sense of empowerment after overcoming an interesting challenge is reward-
ing in and of itself. This alone is sufficient grounding to make it a useful reward
mechanism. Additionally however, we can reward the player with enhanced
means to exercise power, or with an environment that itself is easier to control,
as a form of literal empowerment. These kinds of awards can be quite literal. The
player may receive an upgraded weapon or gain access to a level’s security sys-
tems. Subtler methods may be employed as well. For example, the player’s rela-
tionship with non-player characters (NPCs) can change for the better, or some
player skill can be enhanced from now on, making it easier to deal with future
challenges. Whatever the reward may be, in this particular reward system, it is
important that the player ends up feeling empowered in some way that matters.

Social Dynamics in Level Design

In the earlier section on isolation and social reinforcement, we have seen that
these social principles can both reward and punish, depending on the context
in which they are used. As in real life, in level design this is something that is not
always clear to the player. Sometimes these principles can be quit subtle, but
that does not necessarily take away from their impact. In other circumstances
this principle can be the central one of the design and may be clear to anybody
involved. Either way, social dynamics can be of the utmost importance to level
designers and can be used as strong and important reward systems.

Multiplayer Aspects
In multiplayer games, the social dynamics we have discussed cannot help but
come to the fore. By their very nature, multiplayer games need to be condu-
cive to social reward systems. Therefore, the level design has to incorporate this
where it can, to support the game’s design. This can be done in diverse ways and
on many levels of sophistication. The question, as always, has to be “what are my
level design goals?” In this case, some of the answers may lie in areas dealing with
issues as diverse as spatial considerations, fostering group interactions, or converse-
ly, by enforcing social bottlenecks. (The latter case shows that even in multiplayer
games, temporary isolation can be a positive factor.) All of these examples are
120 Emotional Feedback Systems

elements that can occur in real life and are easily adopted or translated into level
design theory. Let’s take a closer look.

Spatial considerations
Have you ever seen a group of school children enter one of those mazes made of
hedges, which are often built in the garden of some castle or ancient mansion? It
is hardly a coordinated and cohesive affair. In fact, if the children are too young,
it may lead to some rather distraught scenes. What only moments ago was a nice
group affair is now a disjointed and confused scene, filled with individual ac-
counts of fear and dismay.34 The maze, while conducive to individual explora-
tion, or at least to use by small numbers of people, ceases to be enjoyable if the
group becomes too large and is expected to stay together. Imagine what would
happen if the group consisted of hundreds of people!
If we translate this scenario into a typical gameplay one, we get a similar
need for spatial consideration. Depending on the size of the game, it is likely that
the gameplay space itself needs to take special steps to support the game’ social
dynamics. For example, a game that requires large groups of people to explore
old ruins can be made much more enjoyable if the ruins themselves allow for
this by providing wide and high corridors and large rooms. This means groups
can travel together and intermittently take stock of the situation by being able to
gather en masse in a single room, without anybody being left behind or outside.

Fostering group interaction

A classic way of strengthening communities is by involving them in some kind
of group task that benefits the whole. This is especially true if the group can per-
form a task much more effectively than a collection of individuals. Take for in-
stance a support group for parents with children who suffer from diabetes. All
the individual parents may have much knowledge of some of the issues and have
developed coping mechanisms that are helpful to them some of the time. Task
them to improve things collectively, and it becomes clear that the support group
however allows them to pool the information and knowledge and make it avail-
able to all other parents. They can set up a network that can cover for parents
with specific needs, set up an information network, lobby the government as a
pressure group, and so forth. Not only is this group much more effective than
a collection of individuals; it can also forge real bonds of social reinforcement
because others in the group understand the individual needs and problems and
can react to them.

34
Ok, perhaps it is not quite that bad, but it could have been!
Reward Systems 121

If social reinforcement is a desired result of the level design, it makes sense

to foster similar types of situations, where the whole can be more effective than
the individual parts. Since we can author specific challenges and scenarios with-
in the level design, it is within our power to create challenges that require group
solutions or at least are more effectively handled by people working together.
Furthermore, this can be done in such a way that it strengthens an online com-
munity.35 This can happen in multiplayer games ranging from team-based first
person shooters like Counter Strike,36 which is best played on levels where a well-
trained clan can cover big parts of the map strategically by communicating well
and assigning supporting duties, or in games like Gears of War,37 where the level
design is geared towards a separation of tasks that are vital in order to achieve a
common goal. In both cases the level design is crucial in fostering positive social
reinforcement by providing gameplay scenarios that require or encourage this.

Social bottlenecks
Sometimes in a busy multiplayer environment, the constant human interaction,
or worse, constant attacks or abuse, can become too much. Sometimes it is fun
to be a loner, to save the girl as a unique hero, or to just collect one’s thoughts. Or
sometimes it is better to serve the group through individual actions. This makes
the act of breaking away from the group or the masses a positive and can be seen
as a reward in its own right.
This does not mean that it cannot be done while in the service of the greater
communal good. The level designer can use social bottlenecks, for lack of a bet-
ter term, that provide gameplay opportunities for individuals. A sniper may be
able to find a lone perch on top of a building and hold off an advancing group
of enemies while his friends make an escape. The sniper is sure to enjoy a great
deal of social reward when he or she rejoins the group later on when they are
safe. A scout can forge ahead unseen by the enemy and report on the best route
forward. Plenty of other examples exist, but the main point is that it helps to
think about these principles early on, as they can be just as rewarding as group
interactions.

Single-Player Aspects
It may seem a bit odd to talk about social dynamics and single-player games.
Aren’t these games played in solitude, i.e. the opposite from a social dynamic?
35
However, this tends to happen as a consequence anyway.
36
Developed by Valve Software.
37
Developed by Epic Games.
122 Emotional Feedback Systems

This is true to a degree; the player is not with any other people while playing the
game. Nonetheless, this does not mean that social dynamics and social reward
systems cannot have an impact in the absence of people. Two important topics
present themselves: the player can still be exposed to non-player characters
(NPCs) and the player can still be rewarded by, or subjected to, rewarding situ-
ations derived from solitude.38

NPCs
Just because the other people in the game are digital doesn’t mean that they can-
not have an impact, or more to the point, that they cannot be subject to the sys-
tems we described in multiplayer games. This means that spatial considerations,
fostering group interaction, and social bottlenecks are still techniques and areas
that can be explored. NPCs may not be as intelligent or many-layered as real
people, but that is in many ways beside the point. What matters is that they still
need space to maneuver, can work more effectively as a group, and can be subject
to social bottlenecks.
Where NPCs differ from human characters is that the level designer can
control them. While in a multiplayer situation the level designer needs to cre-
ate an environment that is conducive to the people themselves creating socially
rewarding interactions, in a single-player game, the game the designer is able to
determine or predict many of the actions of the NPCs. The designer therefore
can design specific gameplay scenarios that are socially rewarding. This is an im-
portant advantage to have as a designer, as it means the design is not dependent
on the fickle nature of real people.
The player can even form relationships with NPCs. It is possible for a level
designer to foster player bonding with non-player characters, in the absence of
real people to bond with. NPCs can at times provide a similar function to real
players, sometimes with surprisingly strong results. One of the best ways to do
so is by letting the player invest emotionally in the non-player character, as well
as giving the NPC a direct gameplay function. Giving the player the proverbial
puppy to care for can provide direct gameplay gains when the puppy grows up
to be a fiercely loyal guard dog.

Solitude and isolation in single-player games

Sometimes the absence of a thing makes it more powerful in the mind of a per-
son. Think of becoming homesick or missing a loved one while at work. Or in a
slightly more ominous scenario, imagine being locked up and awaiting an inter-

38
Often presented as a contrast to social interaction.
Reward Systems 123

rogator. Imagine traveling through the ruins of an ancient culture, or through a

city mysteriously devoid of any occupants, even though there are signs of recent
habitation everywhere. These are all examples of social dynamics being in play in
situations of complete solitude, a paradoxical but real situation.
Even in single-player games that are normally filled with NPCs, there are
very direct advantages to providing opportunities for solo gameplay. Solitude
can provide breathing space that allows the player to form new social strategies.
Eventually it can even engineer a situation where the player misses the company
of others, thus creating a mechanism that reinforces bonding the next time the
player meets an NPC.

Catharsis and Level Design

We have discussed earlier when people come out of extended periods of enduring
some kind of hardship, or engaging in an extreme challenge, this can lead to a very
strong release or purging of emotions. Together with the resulting positive feel-
ings, for example a sense of renewed purpose, empowerment, and revitalization is
often referred to as catharsis. This cathartic effect is a recognized outcome in many
situations and can even form the basis of formal reward systems. This makes it an
interesting subject for level designers, since we may be able to use some of these
principles to create emotional feedback mechanisms for our levels.
Before we continue examining how this can work, I would like to show an
example of how the principle has been successfully used in filmmaking and see
if the use in another art form can give us knowledge applicable to our own craft.
The best example I can think of is found in the film language developed by Al-
fred Hitchcock throughout his career. Hitchcock is famous for many reasons,
but chief among them is his particular use of suspense.

Alfred Hitchcock and Suspense

Hitchcock was a past master at manipulating an audience’s experience, and his
influence also reaches other chapters in this book, including Chapter 1339 with
regard to set pieces. But in this chapter, we are going to look at the rather nu-
anced fashion in which he used tension and release in a way that was revolution-
ary in its time.
A key aspect to the success and effectiveness of a Hitchcock movie is the use
of suspense to captivate the audience. Suspense in filmic terms is the technique
whereby the director creates a large amount of tension in the audience, but then
39
In the section on “Making Your World Memorable.”
124 Emotional Feedback Systems

waits a certain amount of time before allowing the tension to be resolved. This
state in which the audience is waiting for a resolution is what we call suspense.
While in this state of suspense, the audience will be highly focused and receptive
to onscreen actions.
Compare these two scenes:
Scene 1:
A train is racing across the tracks. We can see that it is filled with
passengers. The camera zooms in on a family scene, two parents and a
child sitting around a table in a carriage. The parents are talking while
the child is playing with a toy train, making it crash and providing the
appropriate sound effects himself. The couple now seem content to look
at their son playing with the toy, they smile at each other. Clearly, they
are happy. Suddenly, the carriage lurches, people scream loudly, and
mayhem ensues. All of a sudden there is quiet, the camera zooms out,
and we see that the train has stopped on a bridge that seems partially
collapsed. It is hanging over the edge of the broken bridge, precariously
balanced!
This scene may be effective40 as it stands and cause fright out of a sense of sur-
prise. I even included a small amount of foreshadowing through the child’s play
with the toy train. Let’s see what happens if we introduce the element of sus-
pense:
Scene 1 v.2:
A grim-faced man is connecting wires to a small box and sweating
profusely. It slowly becomes clear that he is setting up a timer, one that
is linked to a large amount of explosives. Cut to: a family scene, two
parents and a child sitting around a table. The parents are talking while
the child is playing with a toy train, making it crash and providing the
appropriate sound effects himself. Cut back to: the man with the bomb.
We can see more of the environment around the man. He is attaching
the timer and the charge to a large pillar. We can now see that there
are many other charges connected to the same timer, attached to other
pillars. Cut to: the family we have seen earlier. The boy is still playing
his train crash game. The camera reveals that the family is seated within
a train carriage. Cut back to: the man with the explosives. We see he is
ready with his work. He checks his watch and looks at the timer.
The camera zooms out to show that the explosive charges and timer
have been attached to a very high and large train bridge. Furthermore,

40
Although perhaps not a masterful piece of script writing.
Reward Systems 125

we can see the plume of a train appearing further down the tracks.
A train is approaching. Cut to: the family in the train. It is now clear
that the family is heading for disaster, together with everybody else on
the train! Cut back to: the man with the explosives, who is now hiding
behind some rocks with the timer, watching the approaching train.
A figure approaches slowly from behind (a policeman) and suddenly
lunges for the timer, trying to wrestle it from the bomber. A struggle
ensues. Crosscut several times between the fighting men and the approaching
train. The policeman is losing the struggle and the bomber is nearly able
to retrieve the timer. Cut to: a large explosion; debris flies through the
air. Cut to: the approaching train.
Suddenly, the carriage lurches, people scream loudly, and mayhem
ensues. All of a sudden there is quiet. The camera zooms out, showing
that the train has stopped on a bridge that seems partially collapsed. It
is hanging over the edge of the broken bridge, precariously balanced!
The bomb exploded early and the train was able to stop just before
plummeting down the now-broken bridge. Cut to: the family in the
carriage, clearly shaken but unhurt. The boy is still clutching his toy
train.
Despite the laughably clichéd content of the second scene, it illustrates
clearly how much extra tension, depth, and meaning can be added through the
use of suspense. For a long period, the film is able to heighten the tension in the
audience, drawing its complete focus onto the onscreen action. This audience is
finally rewarded for enduring the suspense by a strong and enjoyable resolution
to the tension. This is a clear example of a practical reward system employed to
achieve greater artistic impact.

Including the audience

This was by no means a new technique, even when Hitchcock was mak-
ing his films. But what made much of Hitchcock’s work special was that he
added an extra dimension to the use of suspense by giving the audience more
information than the protagonist onscreen, and extending the duration of sus-
pense. Since the audience cannot alter the events onscreen, this allowed for a
large amount of tension to be created, leading to a subsequent powerful reso-
lution.41
Hitchcock was very aware of the power of this technique and made it a vi-
tal part of his filmmaking. He was notably successful in incorporating audience

41
This resolution does not have to be a positive one. If it is rewarding for the audience, it
may well be a disturbing or frightening one, or whatever the artist finds appropriate.
126 Emotional Feedback Systems

knowledge into a scene, meaning that the audience knows more about the on-
screen situation than the protagonist, which creates even more tension and sus-
pense. A typical Hitchcock classic where this occurs is Rear Window,42 where at
a key moment the viewer is allowed to see things unfold while the protagonist
is asleep.
This principle has become so well known that scenes or entire films are now
is referred to as being Hitchcockian, and many famous directors have taken the
old master’s lessons and incorporated them into their own style. Brian de Palma
is a director who comes to mind who has often used Hitchcockian suspense in
his own movies.43
What this kind of suspense shows us is that it is possible to use the principle
of catharsis, the release of emotions after enduring a lengthy and serious chal-
lenge, to create sophisticated artistic techniques. If it can be done in film, it may
well be possible in level design.

Suspense, Catharsis and Level Design

Suspense works as a technique of a cathartic reward system by using a key aspect
of it, namely the endurance of a challenge over time. It actively sets out an ex-
ceptionally challenging44 situation and artificially extends its duration, suspend-
ing the outcome until the director feels that maximum emotional impact can be
reached in the moment of release. The other main components are challenge and
release. Let’s examine these three in level-design terms.

Exceptional challenge
Catharsis works differently from other reward mechanisms in level design be-
cause of the nature of the challenge necessary to produce the desired emotional
release at the end. The challenge has to be one that goes beyond the expected
and actually tests the player’s ability, and his or her will to cope with something
on a deeper level than regular game mechanics. This is an extremely difficult
thing to pull off well and one of the more dangerous techniques in level design.
The problem is that is easy to create something that is extremely challenging, but
not so easy to make that challenge one that players are willing to finish. Get the
balance wrong, and the player will just stop playing in disgust and never come
back to the game. Why would players have to put up with fighting some unbe-

42
Paramount, 1956.
43
His film Body Double comes to mind, and in fact it can be seen as an ode to Hitch-
cock.
44
To the audience.
Reward Systems 127

lievably strong opponent that just kills them when even one small mistake is
made? Whatever the level designer’s plan is for challenge, a number of consider-
ations have to be taken into account. It helps if the player is motivated to take on
the challenge, the challenge itself makes sense, and it looks at the beginning of the
challenge like there is a chance of success.

Motivation. Why cross a desert if it is known that it will be a gruesomely hard

experience? Why single-handedly take on a whole gang of dangerous criminals?
Why decide to go into the awful dungeons that are clearly haunted? If all of these
challenges deliver on the seriousness of the challenge that they imply, a good
answer to these questions is needed. The level designer can answer them by
providing a good motivation for the player to adopt. The stronger the motivation,
the more willing the player becomes to accept an exceptional challenge. A player
may not want to put up with a long and extremely dangerous trek through a
miserable wasteland, but will become totally committed to doing so if it is to
stealthily follow a group of thugs who have kidnapped the player’s favorite in-
game companion.

Nonarbitrary and focused gameplay. Once a degree of focus is attained through

providing the right motivation, the player still needs to be confronted with
actual challenging gameplay. To keep that focus in place, it helps if the gameplay
is immediately linked to the motivation of the challenge itself. It helps if the
actions required are somehow logically linked to the main task at hand and not
blatant arbitrary hoops to jump through, put in place by the designer.

Viable strategies. Although moments of despair and futility can be allowed to

creep into the challenge, it is vital that at the start of the challenge the player is not
overcome with a sense of hopelessness. The aim of the exercise is to lead players
to catharsis, not to immediately put them off. This means that it is important
to make sure that the player thinks it is worth proceeding. To do that, the level
designer has to provide the player with viable strategies for progress. This doesn’t
have to be a strategy for success; in fact, it can be as humble as allowing the player
to follow an overwhelmingly strong foe and ponder what can be done.

Suspended resolution
Now that an exceptional challenge has been created and the player is commit-
ted to taking it on, the level designer needs to make sure that the challenge plays
out over a sufficient amount of time. It is of no use creating a difficult situation
that can be resolved in ten seconds, as this does not provide enough of a test.
The player will either feel lucky or simply not challenged after all. Instead, just
128 Emotional Feedback Systems

as in Hitchcockian suspense, the challenge grows in meaning and impact if ex-

tended through time. Feelings of panic and doubt can come into play during
this ordeal, but they can be taken away again by a sense of progress, the addition
of new viable strategies, and small successes. This resulting extended gameplay
works as a pressure cooker where the player’s need for a resolution starts build-
ing up and provides the energy for a sufficient emotional response at the time
of release.

Release (Catharsis)
Finally, now that the player has been guided to a moment where a release of ten-
sion can occur, it is very easy to forget that it has to be a rewarding release of ten-
sion. This can be done in several ways, but it is vital that it is done, or the whole
exercise would end up feeling futile and depressing to the player.45 Rather than
risking this, the level designer needs to make sure that the rewards and the ac-
companying feelings of catharsis are strong enough. This can be done by making
the method of release really satisfying, for example by providing a particularly
enjoyable gameplay scenario that allows the player to overcome the challenge.
It can also be helped by providing an extremely positive outcome, for example
by giving the player a long-coveted item, or restoring an emotional bond with a
previously lost companion. There are countless ways of rewarding the player, but
each level designer must make sure that this indeed happens. If chosen correctly,
the resulting emotional release will create a game moment that the player will
cherish for a long time to come.

Associated Dangers
Catharsis is a dangerous technique to use in level design, yet it is a tempting
one because the potential reward is so high. What the level designer needs to
understand is that the technique easily breaks one of the level design fundamen-
tals, namely making sure that the skill level and challenge are in balance. Frus-
trate players long enough with difficult gameplay and they will leave the game
for good. And the nature of the extended challenges we are discussing here is
by definition frustrating, as the resolution desired by the player is purposefully
suspended by the level designer. Tread carefully!
Furthermore, as in all art forms, the artist has a certain amount of responsi-
bility to the audience. In this case the responsibility is not trivial, as we are talk-
ing about highly manipulative techniques used to elicit an emotionally strong
response. This does not invalidate the technique, but it should at least prompt
45
Another chance to lose a player forever.
Reward Systems 129

the artist to check to make sure that no ethical lines are crossed. It is not up to me
to suggest where these lines lie, but I do think that the question should be asked
by level designers who find themselves in this situation

Practice

Example 7.1: Investment and Payoff—

Awarding Exploration
Summary
A basic reward mechanism that all levels should feature is one of investment and
payoff. The player should not only be rewarded for their efforts, but the rewards
should be proportional to the amount of effort required to do something. This
applies to many aspects of gameplay, including exploration.

Game Genre
Most games where the player can explore an environment.

Goals to Achieve
• Encourage explorative gameplay.
• Provide subtle direction.
• Provide proportionate rewards.

Description
(Example type: Original)
Abandoned house. When choosing a setting that rewards exploration game-
play, it is worth taking a moment to think of something that really speaks to the
imagination of the player. This is a nice general goal to maintain, but it is essen-
tial in the case of exploration gameplay, as we have to create an environment in
which the player’s imagination leads them to explore.
An example of such an environment can be a grand abandoned old house.
The concept immediately puts certain images and desires in the mind of the
player: images that can be incorporated into the level design, based on explora-
tion gameplay principles.
130 Emotional Feedback Systems

House areas. We can list areas in the level that are subsequently more and more
off the beaten track, but can yield bigger rewards, both in terms of in-game items
that can be found and of new interesting areas that the player can discover (the
latter can function as a reward in its own right):
• corridors,
• main rooms,
• secondary rooms,
• locked rooms,
• hidden cupboard,
• out of reach attic (pull down ladder),
• locked basement,
• secret passageway,
• secret garden.
These are all easily incorporated into subtle and imaginative level design sce-
narios. Faint footsteps on a dusty floor can lead to a hidden cupboard. Locked
rooms are a clear invitation to explore and find a master key. A locked basement
demands to be accessed one way or another. A secret garden in the middle of a
folly maze can yield real revelations.
Each of these examples can reward the player in one way or another, but
they are all subject to explorative gameplay. If the player is willing to put in the
effort, then he or she can uncover deeper and more exciting secrets and be re-
warded in the process.

Further Notes
It is always useful when thinking of a setting for this kind of gameplay to take
this into account. Is it easy to provide deeper and deeper layers of explora-
tion? If the answer is yes than the level design process should be significantly
easier.

Example 7.2: Escapism—Safe Haven

Summary
Providing escapism can operate as a strong reward system in its own right, and
can appeal to people at a very pure level. Few things are as human as the oc-
casional wish to get away from things, to escape from the troubles of our lives.
Although games are often wrongly derided for this, we should celebrate the fact
that game levels can be a fantastic vehicle for achieving this escape.
Reward Systems 131

Game Genre
The technique is suitable for use within specific levels or a dedicated level that
functions as a hub for all other levels.

Goals to Achieve
• Provide a reward mechanism that uses escapist desires to be effective.
• Incorporate this directly into the level design.
• Tie this into other uses, like providing a practice space for the player.
• Incorporating exploratory or other rewards into the general setup.

Description
(Example type: General)
The extendable safe haven. If a level can feature a safe house or an otherwise
safe area (for example, one that can function as a hub), the level designer has a
good opportunity to create an area that fulfills the desire for escapism. This is
done by creating an area the player returns to regularly, at will or otherwise, in
which no harm can befall the player. Instead, it is a safe place in which the player
can indulge in exploration, strategize, practice skills, store loot, and enjoy a re-
warding environment. Other uses can be added as well.
The safe haven—let’s take a fenced off forgotten industrial area as an exam-
ple—can be made into an extendable safe haven by treating it as a playground and
a home base that features rewards that get slowly unlocked. In this example, the
player can start in a single building at ground level where they can store loot and
decide what to do next. Subsequently, throughout progressing through the level
(or through a hub for all levels), the player will acquire additional skills, abili-
ties and equipment, which can be used to extend the safe haven. New athletic
abilities can make the player reach high places that were previously out of reach,
including a new building with new secrets to discover. New equipment, like a
blow torch or a fence cutter, can open up previously locked areas. The more the
player progresses, the more environmental rewards they may achieve, like great
panoramic views, finding new creatures to interact with and other rewards. As
long as the whole area provides an entirely safe playground for these principles,
the chances are that it will feel like an escapist retreat or playground.

Further Notes
This is a setup that has been used in a number of games, but is especially well
implemented in the Tomb Raider46 games, where the safe house is actually Lara
Croft’s in-game house.
46
Published by Eidos Interactive.
132 Emotional Feedback Systems

Example 7.3: Giving the Player What

Whey Want—Wish-Fulfillment
Summary
Similar to escapism, wish-fulfillment identifies a desire in the player and fulfils
it. This desire can be created, however, which makes it a viable technique to use
in level design.

Game Genre
This technique is suitable for most games.

Goals to Achieve
• Create a desire in the player.
• Build the desire up to a degree that it can become a real reward if fulfilled.

Description
(Example type: Existing game)
The most effective way of providing wish-fulfillment in a level is by creating and
then building up a desire or by strengthening an existing one. In this example we do
this by confronting the player with something desirable all the way through a level,
and making sure that that desire is granted only when this has the greatest impact.
“Wouldn’t It be cool if?” Every gamer has moments when they wish they
could do something, or experience something in a level, that they can’t for one
reason or another. This technique is based on the idea that the player is led to
think this, only to be extremely pleasantly surprised when they can do it after all.
So, the object of this technique is to create a desire that may seem out of
reach but can be fulfilled after all. The exact subject of the desire is completely
fluid and depends on the game and the specific level.
Some great examples are found in a few very famous games (no coincidence,
in my opinion). In Half Life 2: Episode 247 the player is confronted with an ex-
tremely powerful AI companion, a robotic sidekick named Dog. Dog is shown
early on in the game to be extremely strong. (This goes back to a previous install-
ment of the game.) The creature helps the player out on a few occasions, show-

47
Published by Buka Entertainment, developed by Valve Corporation, released October
10, 2007.
Reward Systems 133

ing prodigious strength and ability. The player slowly starts to wish for DOG
to be able to do more than provide sporadic help, and a subtle wish enters the
player’s mind that it would be really cool if DOG were pitched against one of
the game’s super powered opponents. This never happens, however, and it stays
a wish in the player’s head every time they see DOG. Until suddenly, this does
happen when the AI creature suddenly reappears when the player really needs
him and gets into a fight with an enormous robotic tripod creature, known as a
Strider. The resulting fight is pure wish fulfillment.48

Further Notes
Including moments like this can be combined with the need for set pieces and
memorable moments in a game. Get a few of these right, and players will fondly
remember the game.

Example 7.4: Social Reinforcement

through Codependency
Summary
Cooperative multiplayer games rely on groups of players enjoying a game to-
gether. Some of this can occur naturally, but there is much we can do to make
our levels foster social interactions. A strong example of this comes out of en-
gineering a level environment where codependency leads to positive social
reinforcement.

Game Genre
This technique is suitable for cooperative multiplayer games.

Goals to Achieve
• Reward cooperative play.
• Provide a level design set piece based on cooperative principles.
• Teach the player the value of communication.

48
This moment was so successful that it has been turned into a poster available from
Valve’s online store: http://store.valvesoftware.com/productpages/prints/product_
HL2DogvStriderPoster.html.
134 Emotional Feedback Systems

Description
(Example type: Original)
A cooperative multiplayer game’s core game design should be such that social
interaction is rewarded, and in most cases this goal is met in basic gameplay deci-
sions. Players may share resources, commands can be given to a whole group, or
group attacks can be subject to damage multipliers.
This is all valid and on its own can do much to reach the game’s desired
goals, but the level design needs to provide a context for the cooperative actions
and enhance them or instigate them when possible. This can be done by provid-
ing the players with challenges that are only solvable, or are better solvable, by
players working together in concert. The idea is to create a sense of codepen-
dency where players need to be aware of each other and fulfill roles that support
the group as a whole.
Codependent group defense. In this scenario the players need to defend a
specific area—let’s say a base—against an attack or invasion of sorts that is com-
ing in waves from all directions. Codependency is guaranteed if the base can-
not physically be defended by a collection of isolated individuals, but requires a
group working together, in constant communication with each other.
Imagine that the players can defend their base through the use of turrets,
which are mounted with the only weapons strong enough to stop the invaders.
There is, however, only a limited number of turrets available, too few to cover
the entire base, and they are very slow to be redeployed. The only way that the
defending players can position the turrets in time to repel the waves of attack is
by receiving advance notification of the directions from which the attacks are
coming. To do so, a number of players will have to leave the base, scout out ad-
vancing enemies, and report their locations back to the base.
To create an even deeper sense of codependency, we can make the scouting
job too dangerous for one player, and instead require a guard to go along and
protect the player while he do his scouting work.
When all the waves of attack have been stopped, all players involved will
have individual stories on how they helped the group as a whole, which in turn
will yield a large sense of accomplishment both on an individual level and on a
group level.

Further Notes
This is a specific set piece that requires the level designer to plan and orches-
trate the gameplay to a certain degree and make sure that players understand
what their roles are. Easier implementations can be construed that employ the
Reward Systems 135

same principles but are less dependent on polish and extra level design work: for
example, a t-junction where two players need to cover each other’s backs, or a
bottleneck where players need to perform various diverse tasks simultaneously
in order to progress.
Immersion
8
T he concept of immersion constitutes an interesting metaphor. It brings
to mind the sensation of being immersed in water or some other fluid.
When immersed we are completely enveloped, and in many ways it speaks of
an all-encompassing experience. We speak of being immersed in a story or be-
ing immersed in work. When this occurs, we temporarily exchange our wide
view of the world around us for one that employs a very narrow gaze or focus.
There are many ways this state can be reached, but the one that is of most in-
terest to us within the context of this book is the state of immersion reached
through artistic constructs, although I will touch on other forms of immersion
as well.
We cannot speak of this kind of immersion and immersion techniques with-
out looking at the concept of suspension of disbelief, which lies at the core of these
matters.

Concept

Suspension of Disbelief
George Orwell famously coined the term double think in his seminal dystopian
novel 1984. The concept of double think is fascinating, if initially difficult to
comprehend (by definition). This is the description offered in the book1:
the power of holding two contradictory beliefs in one’s mind simultane-
ously, and accepting both of them.... To tell deliberate lies while genu-
inely believing in them, to forget any fact that has become inconvenient,
and then, when it becomes necessary again, to draw it back from oblivi-
1
George Orwell, 1984, Penguin Books, New York, 1968, p. 171.
137
138 Emotional Feedback Systems

on for just so long as it is needed, to deny the existence of objective real-

ity and all the while to take account of the reality which one denies—all
this is indispensably necessary. Even in using the word doublethink it
is necessary to exercise doublethink. For by using the word one admits
that one is tampering with reality; by a fresh act of doublethink one
erases this knowledge; and so on indefinitely, with the lie always one
leap ahead of the truth.
Perhaps with the exception of politicians or the mentally ill, it is hard to un-
derstand how human beings can have this capability of holding two contradic-
tory beliefs simultaneously. Yet without realizing it, we engage in this behavior
pretty much every day of our lives. We do it when we go to the cinema, watch a
play in the theater, read a book, engage in a role-playing session, or play a game.
During all of these activities we readily, actively and often wholeheartedly believe
in something we know to be untrue. We know that the people on the silver screen
are actors, or that the book describes a fictitious world, yet we let ourselves be
convinced of this fantasy in order to experience things we otherwise would not.
We suspend our disbelief.
Suspension of disbelief is no less an extraordinary concept than Orwell’s
notion of double think. Both concepts rely on a human being’s capacity for make
believe, to believe in an artificial reality of our own construction, while knowing
it to be false.

Cognitive Dissonance
At times this ability can be so strong that it takes us to too far, into areas where it
isn’t appropriate. An example of this is when it occurs within the realm of a per-
son’s worldview. In other words, it can occur outside the realm of make believe,
fantasy or other accepted forms of artificiality, which is of course very danger-
ous. How can we make correct judgments of the world around us if we submit
to contrary beliefs? The simultaneous belief in contrary facts, when exposed to
evidence that threatens this belief, can lead to strange results. It can cause mental
friction and conflict, which need to be resolved. When this happens we speak of
cognitive dissonance, which has been defined as
anxiety that results from simultaneously holding contradictory or
otherwise incompatible attitudes, beliefs, or the like, as when one likes
a person but disapproves strongly of one of his or her habits.2
2
American Psychological Association (APA): cognitive dissonance (n.d.), Dictionary.
com Unabridged (v 1.1). Retrieved June 02, 2009, from Dictionary.com website: htt p://
dictionary.classic.reference.com/browse/cognitive dissonance.
Immersion 139

What is essential to understand about cognitive dissonance is that people

can force themselves to maintain the contrary beliefs by going to extraordinary
lengths to reconcile their worldview with their expectation of it, which may not
be factual at all. This need to fix the facts of the world around a belief system is
a need that can be manipulated, and history shows us that this is often done for
exploitative reasons.

The Need for Self Deception

What all this points at is that human beings have a strange capacity, and some-
times a strong need, for self-deception. (Sometimes this reaches the point of
detrimental self-delusion, as cognitive dissonance shows.) Some of the origins
for this phenomenon have already been discussed in Chapter 2, where it is con-
cluded that this ability is needed to allow us to construct artificial mental arenas
in which we can learn lessons, or experience environments in a safer context
than reality allows us. In other words; sometimes self-deception can be a survival
skill. This need, however, can be appealed to in ways that constitute mental ma-
nipulation.
Throughout history, but increasingly in the last century to the present, peo-
ple have understood this phenomenon and used the knowledge to manipulate
other human beings, sometimes with catastrophic and even genocidal results.
Here is a passage of an interview with Hermann Goering:
We got around to the subject of war again and I said that, contrary to his
attitude, I did not think that the common people are very thankful for
leaders who bring them war and destruction.
“Why, of course, the people don’t want war,” Goering shrugged. “Why
would some poor slob on a farm want to risk his life in a war when the
best that he can get out of it is to come back to his farm in one piece?
Naturally, the common people don’t want war; neither in Russia nor in
England nor in America, nor for that matter in Germany. That is under-
stood. But, after all, it is the leaders of the country who determine the
policy and it is always a simple matter to drag the people along, whether
it is a democracy or a fascist dictatorship or a Parliament or a Commu-
nist dictatorship.”
“There is one difference,” I pointed out. “In a democracy the people
have some say in the matter through their elected representatives, and
in the United States only Congress can declare wars.”
“Oh, that is all well and good, but, voice or no voice, the people can al-
ways be brought to the bidding of the leaders. That is easy. All you have
140 Emotional Feedback Systems

to do is tell them they are being attacked and denounce the pacifists for
lack of patriotism and exposing the country to danger. It works the same
way in any country.”3
In this case the manipulation occurs by appealing to the human need to feel
protected, through the use of propaganda. Other examples can be given where
manipulation occurs by appealing to a need to protect, nurture, mate, or one of
many other natural human urges.

Positive Uses
It is important to understand that the mechanisms described so far, although
open to malicious manipulation, are not negative in themselves. There are of
course many legitimate ways to use this knowledge. Techniques to enhance im-
mersion and encourage suspension of disbelief have been practiced and honed
for millennia by storytellers and other artists across many disciplines.
There is nothing sinister about a skillful storyteller enthralling an audience
around a campfire. We don’t resent a filmmaker for using meaningful camera
angles to add strength to a scene. Cognitive dissonance, for example, can be
used to create tension in a story or to engineer an audience’s need to resolve
within a fictional setting. Artists and entertainers rely on manipulating the au-
dience in order to be effective, and the audience is willing to subject itself to
this kind of positive manipulation. Crucial to these positive examples is that
the manipulation occurs with the consent of the manipulated. (As opposed to, for
example, political propaganda.) We could describe this process as one of benign
deceit.
Pianist Hal Galper describes this implied understanding between artist (or
entertainer) and audience as follows:
A mutual social contract is unconsciously agreed upon between artist
and audience. The conditions of this contract must be fulfilled by both
performer and listener and entail emotional risk for both parties. Con-
trary to popular belief, most people are afraid of freedom, of “letting
go.” Suspension of the sense of self is a rare, pleasurable experience that
is not an everyday occurrence in most people’s lives. Audiences gravi-
tate toward live performance situations because they offer a safe way
to let go. However, letting go can suggest the possibility of emotional

3
Gustav Gilbert, interviewing Hermann Goerring during the Nuremberg Trials in April
1946. Later published in the book Nuremberg Diary (G. M. Gilbert, 1961). Available at
http://www.snopes.com/quotes/goering.asp.
Immersion 141

risk because the listener’s defenses are let down. Listeners, by showing
up paying and paying an attendance fee, have willingly entered into an
agreement with the performers. They feel comfortable with the fact
that, along with other members of the audience, they’ve agreed to open
themselves up to any influence the performers may exert upon them. As
suspending one’s sense of self creates a unique sense of freedom, the lis-
tener can then derive pleasure from this experience allowing themselves
to be swept up by a musical performance without emotional risk. The
audience has, for that moment, put the state of their emotional well-
being into the hands of the performer, a responsibility the performer
must accept with care.4
There is an underlying understanding between artist and audience that this
manipulation is legitimate as long as it is not abusive. There are of course many
instances where the boundary between abuse and legitimate artistic expression
is blurred, but the principle stays intact. The disagreement between artist and
audience in that case is one born of execution, not of principle.

Other Forms of Immersion

There are also non-manipulative factors at play when we study immersion. Hor-
mones can play a great role in focusing the mind, not to speak of other chemi-
cals like endorphins, testosterone, or adrenaline. There are many psychotropic
or mind-opening drugs that can immerse a person in all kinds of immersive
experiences. Crucially, these kinds of mechanisms take control away from the
subject.
Then there are many other activities that can all contribute to an immersed
mind. On a regular basis most of us are
• immersed in work,
• immersed in a moral conundrum,
• immersed in sport,
• immersed in beauty.
There are many more, so many that it is pointless to try to name all genres or
areas where immersion can occur. More so, it is beside the point. What matters
is that it is completely clear that this capacity to be immersed is deeply ingrained
in the human psyche and that it will often occur, especially in situations where this
is beneficial to us.

4
Hal Galper, “The Social Contract: Presentation And Creativity,” Down Beat Magazine,
http://www.halgalper.com/13_arti/social_contract.html, December 1994.
142 Emotional Feedback Systems

Concept Summary
A human being’s capacity for immersion and suspension of disbelief is all-perva-
sive and deeply ingrained in all aspects of our lives. In many cases this is a benign,
even necessary phenomenon. It occurs on a daily basis on a natural and often
subconscious level.
One should be aware, however, that it can also be manipulated, especially
by those who understand what underlying mechanisms are at play. This can be
done in amoral and illegitimate ways (without explicit approval), or this can be
done with the consent of the user to create a positive experience (as in the arts).
If it is done well, the various elements that allow an audience to suspend its dis-
belief contribute to a general sense of immersion. In effect, the artist creates a
coherent virtual environment or experience in which the audience can believe.
We can find this immersion in artificial environments through suspension of dis-
belief all throughout culture. Examples are rife and include books, films, music,
and video games.
Further factors that can contribute to immersion but are independent of
suspension of disbelief are also at play. These are too wide-ranging in scope to
cover in one book, and they often occur outside the context of art and enter-
tainment. Nonetheless, it is important to study them to create a deeper under-
standing of immersion and immersive factors, since this knowledge can help by
improving or discovering new immersive techniques. What all these concepts
tell us is that human beings are subject to, and have inbuilt mechanisms that allow
them to enjoy, deeply immersive experiences.
A solid understanding of a broad scope of immersive techniques lies at the
core of most aspects of level design. As level designers we have the capacity to
fulfill our end of the unwritten contract between artist and audience by making
sure we employ all the knowledge we have of mood manipulation without be-
traying the trust of our audience.

Theory

Immersion and Level Design

It is legitimate within the arts to engage in a benign deceit to create a positive ex-
perience for the audience. The audience knows it is being deceived but chooses
to suspend its disbelief willingly, as it is part of the process of enjoying art. Or at
Immersion 143

least this is what happens if the artist is skilful enough in the necessary manipula-
tion techniques. This is extremely relevant and applicable to level design because
video games often operate as artificial worlds. We want players to believe in those
worlds so that they accept as a consequence the game’s formal rules.
If level design acts as a teaching mechanism, it helps if the pupil is willing to
believe the teacher.
It is not enough for the level designer to just focus on gameplay mechan-
ics in an isolated way. We need to be able to bind the player to the gameplay
world in a manner that is not objectionable or overly coercive. We need to be
able to charm players, engross them in a new universe, and make sure they want
to spend time in it. This can be done in a positive way, by providing a beautiful
or engaging environment; or just as valid, in a negative way by realizing a night-
marish dreamscape. Key to this is that we make sure that the player is willing to
suspend disbelief in what clearly is an artificial reality. (In most cases, immersive
gameplay will trump other elements of immersion.)
There are hundreds of ways to achieve the goal of immersion, and through-
out this book we will cover a large number of basic techniques that can be adapt-
ed by any level designer. This chapter will look at a number of general principles
and allow for ways of extrapolating to practical theory and application.

More on Suspension of Disbelief

I have so far spent a considerable amount of time on the topic of suspension of
disbelief, although with regards to immersion it is not the only factor of impor-
tance. This partly comes from the knowledge that to play a game is to suspend
one’s disbelief. If this voluntary self-deceit doesn’t occur, either the game isn’t
being played or the player is delusional and doesn’t understand that the activity
is in fact a game. This is the original conceit that is necessary to even begin talk-
ing about games, just the same as a reader knows the difference between fiction
and fact, and a moviegoer or theatergoer knows the difference between drama
and reality. This is more important than it may seem on the surface, where it
appears as just common sense. It is important to explicitly contemplate this con-
cept since, as shown in the previous chapter, it brings to the fore a number of
important underlying psychological processes, all of which have direct bearing
on level design.
However, there is another layer of immersion in play, which has to do with
the actual content of the game. The in-game actions, events, environment, etc.,
are all subject to their own processes of immersion, including (but not limit-
ed to) suspension of disbelief. This type of suspension of disbelief is covered
throughout this chapter, especially in the section on game logic, and will be ex-
144 Emotional Feedback Systems

trapolated upon when appropriate. This topic is too wide ranging to be covered
in its own section; instead, I will raise it as it occurs in relation to other elements,
whenever they occur.

The Zone
Many classic eighties movies feature a scene where somebody, often the hero, is
completely engrossed in a video game, typically played on a machine at an arcade.
Take for example that seminal piece of eighties filmmaking, The Last Star Fighter,5
in which the main character is so good at an arcade game that he is identified as
“The Last Starfighter,” an elite starship pilot destined to save the universe.
Or consider the opening scene in the film War Games, where Matthew
Broderick displays an unmatched mastery of arcade games, foreshadowing his
later conflict with, and understanding of, a pentagon computer on the verge of
unleashing thermonuclear war.
I am not including these examples of eighties cinema just out of some sense
of nostalgia (although I was one of those kids chained to the local arcade’s game
cabinets, in the town where I grew up). The reason I mention them is twofold;
1. Video game developers started to become really good at tying players to
their games,6 and gamers started to experience a state where, for prolonged
periods of time, they were completely immersed in the game they were
playing, often at the height of their ability.
2. Society in the eighties showed a genuine interest in this phenomenon
of video-game-induced deep focus, and this fascination was reflected
in countless eighties movies that have built scenes around this concept.
The two previous examples are typical in that they showcase the players
of games as people with extraordinary powers, echoed in their mastery of
video games. These players were able to stay “in the zone” for extended
periods of time.
This jargon of being “in the zone” became widespread7 and accepted, and
was subject to serious study and research. The zone is a concept where players are
kept in a completely immersed and entertained state of mind by the game they
are playing. The interaction between game and player is completely harmonious,
the game is working well, and the player is playing well. Often this leads to play-

5
Directed by Nick Castle, distributed by Universal/Lorimar, a joint-venture (original
release), released on July 13, 1984.
6
To make sure they kept pumping quarters into the game.
7
Previously limited mostly to the field of sports.
Immersion 145

ers performing at the peak of their abilities, which is often referred to as being
“in the zone.” The concept goes further than degrees of proficiency, however. It
may be more accurate to describe the zone as a place where the player is utterly
engaged with the game, or in other words, a place where the player is completely
immersed in the game. A place where as level designers we would like to lead
players, and keep them for as long as possible.

An Introduction to Flow Theory

Much that is known about the concept of being in “the zone” actually comes
from “flow theory,” a concept pioneered by Mihaly Csiksczentmihalyi, which
has many implications for level designers.

Optimal experiences
Flow theory was originally linked to the study of happiness. Through years of
academic research and interviews with thousands of people around the world,
Professor Csiksczentmihalyi has been trying to determine how people experi-
ence happiness and what causes this state of being. One of the terms he uses in
his books is optimal experience. He describes the happiness inherent in optimal
experiences as follows:
Getting control of life is never easy, and sometimes it can definitely be
painful. But in the long run optimal experiences add up to a sense of
mastery—or better, a sense of participation in determining the content
of life—that comes as close to what is usually meant by happiness as
anything we can conceivably imagine.8
These optimal experiences describe life experiences, and the professor does not dif-
ferentiate between activities that can cause this happiness to occur. The source can
be any experience, ranging from work to childrearing to survival sports. What is
more important is finding out what makes these (optimal) experiences enjoyable.

Elements of enjoyment
Throughout many interviews, and from further research, the professor found
that all respondents, regardless of culture, location, financial status, etc., shared
a number of “elements of enjoyment” that are fundamental9 to being happy. He
has formulated eight principle elements of enjoyment:

8
Mihaly Csiksczentmihalyi, Flow: the Psychology of Optimal Experience, Harper Peren-
nial, New York, 1991, p. 4.
9
Although not all of them need to be experienced simultaneously.
146 Emotional Feedback Systems

1. The activity is challenging and requires skill.

2. It requires full concentration (action and awareness merge).
3. It must provide clear goals.
4. The task undertaken must provide immediate feedback.
5. The person involved stays concentrated on the task at hand (it removes
worries of everyday life).
6. It provides a sense of control over our actions.
7. The activity involves loss of self consciousness (we lose the sense of being
separate from the world).
8. The activity involves the transformation of time (hours can pass in min-
utes, and vice versa).
These principles of enjoyment are fundamental to happiness as optimal (life) ex-
periences, but in practice they can easily be applied to level design, if interpreted
as optimal game experiences. It is remarkable just how closely they can correlate
with gameplay. These principles would not look out of place as a quality check-
list for most levels.
When enough of these factors are in effect, the professor describes the re-
sulting happiness as that of a person who is in a state of flow.
Csiksczentmihalyi has described being in a state of flow as follows:
… being completely involved in an activity for its own sake. The ego
falls away. Time flies. Every action, movement, and thought follows in-
evitably from the previous one, like playing jazz. Your whole being is
involved, and you’re using your skills to the utmost.10
And more specifically, there is
… a sense that one’s skills are adequate to cope with the challenges at hand,
in a goal directed, rule-bound action system that provides clear clues as to
how well one is performing. Concentration is so intense that there is no
attention left over to think about anything irrelevant, or to worry about
problems. Self-consciousness disappears, and the sense of time becomes
distorted. An activity that produces such experiences is so gratifying that
people are willing to do it for its own sake, with little concern for what
they will get out of it, even when it is difficult, or dangerous.11
I would be hard pressed to find a better description of what I would consider an
optimal immersive game experience, and therefore what we should aim for in

10
From an interview with Wired magazine, “Go with the Flow,” Wired, http://www.
wired.com/wired/archive/4.09/czik_pr.html, September, 1996.
11
Mihály Csíkszentmihályi, Flow: The Psychology of Optimal Experience, Harper
Perennial, New York, 1991, p. 71.
Immersion 147

Figure 8.1. Flow channel diagram.

our level designs. This becomes even more exciting if we look at the basic me-
chanics at play when we look at the circumstances under which flow occurs.

Flow channel
According to professor Csiksczentmihalyi, being in a state of flow can occur
when a person is participating in an activity that is balanced between the per-
son’s ability and the activity’s challenge level. This is clearly illustrated in the dia-
gram in Figure 8.1.
When the challenge is in balance with the person’s skill level, that person is
in the (flow) zone. But if the activity is too challenging, the subject will feel anxi-
ety. Conversely, if the person’s skill level is too high, boredom will ensue.

Flow Theory and Immersive Level Design

Just like the developers of those arcade games in the eighties, we should do our
best to keep our players in the zone (in a state of flow), not so much to keep them
pumping in quarters, but to keep them deeply immersed in the game. It is impor-
tant to remember that
A player who is in the zone is a player who is deeply immersed.
An immersed player is a happy player, and a happy player is much more recep-
tive to what the game has to offer. It should be clear that this makes immersion
a central goal to good level design. It is not a separate aspect of level production,
148 Emotional Feedback Systems

but a key aspect that permeates all other related areas. Flow theory can provide
us with valuable insights on how to keep the player immersed in our levels and
on keeping the levels as enjoyable as possible.

The Player Wants to Play

In most cases, we know that the player of the game wants to play. This is not as
flippant as it sounds; we encounter many games in life that we are unwilling to
play and that cause resentment if we are forced to do so. Examples include play-
ing mind games at work, avoiding the bully at school, and facing one-upmanship
in a socially competitive setting. With video games, however, we know that play-
ers want to play our levels, because they bought the game, borrowed it, or in one
way or another took a conscious step to try out the game. This willingness to
play needs to be awarded by the level designer and provides us with an excellent
opportunity to bind the player to the game.

What Players Want

It sometimes pays off to give players what they want in the first level they play.
There are of course many ways to do so, and there is no set way that always
works.12
To “give them what they want” early on does not necessarily mean pamper-
ing the player, being overly nice, providing an inordinate amount of challenge,
or that we can focus exclusively on any one thing. What it means is that we must
respect players’ choice in picking up our game, and understand why they have
done so. What is it they expect or hope to gain? A certain amount of psychology
is at play here; after all, we are trying to understand what is in another person’s
mind (not an exact science).
However, there are genre conventions that often help us determine certain
desires. It is likely that in the case of a first person shooter, the player is looking
forward to shooting things. A 3D adventurer may find joy in exploration, and a
platform game may have to provide good jumping action. These examples are
so generic that they are almost useless, but they nonetheless provide a basis of
understanding to work from.
This understanding of basic player desires can lead to much subtler uses, for
example in the case of a first person shooter, we may decide to withhold weap-
ons to shoot with, but to provide a large number of targets anyway. (This is an
12
Beware of designers who claim otherwise; they have developed tunnel vision, a dan-
gerous affliction for level designers.
Immersion 149

inversion of the principle, “do NOT give players what they want.”) Later in the
level, when the desire to shoot things has been amplified by this preponderance
of targets, the level design can deliver in a dramatically impactful way, providing
shooting action in spades. This is a dangerous method, since it can put off the
player if taken to extremes. But if used correctly, it can be an incredibly powerful
way of focusing the player on in-game events. As this example shows, from iden-
tifying player desire (the player wants to shoot things) a clear level design tactic
has evolved that is relatively nuanced and ultimately rewarding to the player.
Other player desires will yield other techniques, but we must first under-
stand what it is that the player wants to get out of the game. This can result in
charming the player by rewarding this desire immediately, or it may not. Some
games are best served by surprising the player, and others by giving players ex-
actly what they want. But an intelligent choice can only be made by level design-
ers if they know what the player’s needs are in the first place. Either way, this is
knowledge that through good application in level design is much more likely to
keep the player immersed.

Temporal and Historic Grounding

It is sometimes disconcerting to enter a newly built residential area, especially
if the architecture is uninspiring. This wrongness can be felt strongly in hast-
ily erected touristy areas, or cheaply built reconstruction areas. The experience
can feel slightly surreal or unsettling in its artificiality, and it is not always easy
to explain why. In my experience, much can be explained by a noting a lack of
historical grounding.
Normally our environment is strongly rooted in historic significance. There
are either strong memories associated from our own lives, or there is evidence
everywhere of past events. These include public art, architectural styles, wear
and tear, and all kinds of evidence of human habitation for prolonged periods. If
our environment is natural rather than man-made, there will be still more links to
the past, which ground the area, such as vegetation, animal life, erosion, etc. We
expect these things on both a conscious and unconscious level. When none of
these elements are present, we experience a disconnect. The environment doesn’t
feel grounded in time, which causes use to feel uneasy, or at least to notice that
something is lacking.
In level design, a great way to anchor players solidly into the experience is by
providing them with this temporal grounding, to make sure that the game world
feels real. After all, if the world shows evidence that it exists beyond our own
actions within it, it must be real, right? Well, maybe not, but it provides a good
150 Emotional Feedback Systems

argument for our subconscious mind that it is okay to suspend our disbelief. Any
good storyteller knows that adding key details, even if not directly related to the
story, makes a story come alive. A game’s level, for better or for worse, has the
capacity to convey a huge amount of historic information:
• Are the buildings old and abandoned, or shiny and new?
• Is the NPC aware of the player’s history?
• Has there been a previous battle is this location?
• Are there boat wrecks at the bottom of the sea?
• Has anybody been up this trail?
• Whose skeleton is that?
The list is endless.
The downside of is that the level designer needs to be aware of all this in-
formation, but the upside is that it allows a large amount of control over what
information is conveyed to the player. Because of this, it gives the level designer
many golden opportunities to really deepen the immersion that a player feels in
the level, something that should be embraced and used whenever possible. We
can take a sterile and uninspiring piece of suburbia and age it artificially by mak-
ing sure that the textures show age, and that the buildings have been adapted by
their owners and imbued with their personality. We can have NPCs reference
important past events, plant evidence of submerged subplots, add secret areas,
and many other things besides. If this is done correctly we should be able to
sketch out enough of an outline for the player’s unconscious mind to accept the
place as real. If this happens, we have achieved an important level design goal.

Non-Games Examples
This principle of historic grounding is known throughout art and entertainment
and there are many established methods and philosophies related to achieving
this goal. George Lucas, to take a famous example, has referred to it in the past
as wanting to create a “used future.”13 He talks about giving the future a past by
using props that are deliberately smudged or have been used in real life. Instead
of a shiny technocratic future, this creates a more believable one, where objects
look like they have always been there. This can be seen, for example, in his film
THX 1138, which features an industrial setting where workers operate machin-
ery that may be futuristic, but not implausibly so. The machines look like they
are made to be used (and have been used) instead of being incorporated in the
movie because they look sleek and stylish.
13
THX 1138: The George Lucas Director’s Cut, Special Edition, Warner Bros., 2004 (origi-
nally released by American Zoetrope in 1971).
Immersion 151

Photography is another field that provides a unique insight to the subject.

Ironically, in photography it is sometimes preferable to try to remove as much
historical grounding as possible, to create an image that is timeless and doesn’t
date badly. Although this is an inversion of the principle, it showcases how pow-
erful this concept can be. In these cases the photographer simply doesn’t trust
the viewer of the image to be able to interpret it without being affected by infor-
mation that places the image in a temporal context!
Another good example can be found in the field of Computer Graphic Imag-
ing (CGI), where it is important for the images to be able to blend in with real
photography. CGI work tends to stand out like a sore thumb unless it is made to
look used instead of shiny and new. With this example it is crucial that the imag-
ery be grounded in time somehow, or the audience will know that the imagery
isn’t real, even if they know it can never be real anyway.

Game Logic
An important aspect of suspension of disbelief is indeed believability. Sometimes
it is forgotten that suspension of disbelief does not mean that a level can incor-
porate any idea or arbitrary restriction that the level designer wants to use. This
isn’t because those ideas are too wild or too fantastical, but because they need to
conform to the game’s internal logic, or game logic.14
There are certain conceits that an audience is willing to put up with, even if
they are not realistic in the real world, as long as they link into the game’s reality
in a logical and consistent manner. For example, take Tolkien’s Lord of the Rings,
famed for its detailed immersive world. People are perfectly happy to suspend
their disbelief in magic, orcs, dwarves, and a plethora of other things that would
be utterly ridiculous in the context of our real world. This is the case because
these elements are part of a complete and logical set of internal rules. Imagine,
though, that at Helm’s deep, Gandalf suddenly would pull out a handgun and
start shooting orcs in the kneecap while spouting sarcastic one-liners? Both the
act and the behavior would be so inconsistent with the established internal rules
that readers would immediately lose their suspension of disbelief and most likely
abandon the book. (This behavior is perfectly acceptable, however in most Ar-
nold Schwarzenegger action movies.)
The same principle applies to level design. Just because it is possible in a
game to set up fantastical or science-fiction themed scenarios does not mean
that they aren’t subject to sustained and serious scrutiny by the player.
14
The concept of game logic is partly determined by game and genre conventions, a topic
discussed in detail in Chapter 7, “Towards a Shared Grammar for Level Design.”
152 Emotional Feedback Systems

Please note that this also means that the level designer needs to be diligent
in laying out the groundwork and must teach the player what the game’s internal
rules are, or in other words, must establish and maintain game logic.

The Game Crate

As an example, take archetypical game crate, used in hundreds of game levels.
Let’s assume that the level designer sets up a scenario where a number of crates
can be pushed together to create an artificial set of stairs to reach a previously
restricted area. The designer could end up with an engaging spatial puzzle. So
far, so good. What needs to be remembered, though, is that this also creates an
additional bit of game logic: crates can be pushed around to create new routes within
the level. This game logic needs to be maintained, or the designer risks destroying
immersion because the game world does not follow its own internal rules.
In our world of empirical evidence and scientific reason, it is reasonable to
expect that our laws of physics are consistent. In a game the same needs to be
true, or the player’s acquired skills are unpredictable and thus useless. A logically
consistent game world suggests a “real” game world, or at least one “real” enough
to suspend players’ disbelief and make them accept the presented fantasy as a
coherent whole.

Some Typical Mistakes

There are many mistakes to be made, and shockingly, some of the easiest ones to
avoid keep reappearing in game levels over and over again: to name a few,
• invisible barrier,
• cheating AI,
• impossible constructions,
• deus ex machina events.15
I am sure you can add many more yourself. Once the principle is recognized, it
becomes clear how widespread the problem really is. It must be clear, however,
that the problem does not lie with realism, but with those rules that apply to the
specific game or genre. That is why jumping on enemy heads is perfectly viable

15
“God from the machine.” In classical Greek plays, gods would sometimes resolve a
tricky situation through divine intervention. An actor playing a god (deus) would be
literally lowered onto the stage by a mechanical crane (machina). The term now refers
to an improbable intervention by the author to solve some problem, for example, the
appearance of a will in a murder mystery.
Immersion 153

in a cartoony platform game, but would make no sense whatsoever in a serious

World War II shooter. The most important thing to remember is that everything
must be contextual to the type of game being made.

Breaking the fourth wall

In theater and film, a well-known convention consists of not breaking the fourth
wall. This means that the filmmaker or stage director does not allow the produc-
tion to acknowledge the audience directly. No references can be made that allude
to the fact that the world portrayed is an artificial one. The film or play thus stays
in character. Imagine a serious film about ancient Rome where the extras can
be heard discussing the superiority of sneakers over sandals, or a detective in a
murder mystery discussing the fact that the audience also doesn’t know who dies
next. I can’t think of many other ways that are better at destroying suspension
of disbelief than actively making the audience aware that they are experiencing
something artificial. This needs to be avoided where possible.
It is surprising, then, to find out that there is a rich tradition of people break-
ing this golden rule. We can go as far back as Groucho Marx winking into the
camera, back hundreds of years in the past to find stage writers who employ the
use of asides, a technique that allows for a commentator to expand on the actions
onstage for the benefit of the audience.
Games themselves employ completely acceptable exceptions like tutorial
text, or opening credits. Nonetheless, most of these instances consist of estab-
lished game-specific conventions that we know we can ignore. However, there
are examples of more blatant disregard of this rule. In the game Metal Gear Solid,
the player is suddenly told by a psychic that he/she “saves a lot” and that else-
where in the game the solution of a specific puzzle is to be found by finding
a clue within the box art of the game! Although admittedly some players have
made clear that they enjoyed these unexpected moments, many have declared
their dissatisfaction at being pulled out of the game world in such a rude manner.
In most cases it is best to err on the side of caution, and in my experience this is
especially true in games where suspension of disbelief is important to the overall
experience, which is a very large category indeed.

Ambience and Atmospherics

The proper use of ambience and atmospherics can completely transform a game’s
levels from feeling boring and uninspired to being completely engrossing and
immersive play environments. With that in mind, one would expect much time
and resources to be allocated to this, but that is unfortunately not the case in
154 Emotional Feedback Systems

most games’ development processes. The reason is that in the very finite sched-
ule of commercial level design, the focus is on something that is good enough, not
on something that is very good. This often means that due to time considerations,
resources are focused on other matters, often as simple as getting levels to a ship-
pable state. A shippable state is often a far cry from a very good level, however. It
is essential for level designers to work to a sensible budget and specifications and
be able to create good immersive levels from that starting point. If appropriate to
the game, immersiveness needs to be part of its early spec and be incorporated
into the level design.
Often it will be argued that immersiveness is always trumped by direct game-
play implementation, but that is too simplistic an approach. Can you imagine a
game like Silent Hill16 working well without a serious amount of time assigned
to the level designers to implement levels with a good use of atmospherics and
ambience? Silent Hill 2 starts with an extended walk through the woods, without
actual gameplay, yet the scene is riveting and completely immersive. It is an ex-
cellent start to the game and sets up the whole experience of irrational fear and
dread very well.
Level designers should expect many arguments about an approach like this,
but they must be able to fight their corner. For this to be possible, they must have
a good grasp of what constitutes ambience and atmospherics. Vague as these
terms may be, there is much known about them, even if it is subjective to a de-
gree. Within the context of level design, I make a distinction between the two,
where ambience is to do with location aspects, while atmospherics have more to
do with general mood enhancement.

Ambience
Brian Eno, the inventor of ambient music, has the following to say on ambience:
An ambience is defined as an atmosphere, or a surrounding influence:
a tint. My intention is to produce original pieces ostensibly (but not
exclusively) for particular times and situations with a view to building
up a small but versatile catalogue of environmental music suited to a
wide variety of moods and atmospheres.17
Note that he talks about music that is part of the environment, an important dis-
tinction that sets the music apart from a normal soundtrack. (Although a sound-
track can be made up of ambient music, in which case the two can coincide.)
16
Famous survival horror game series, known especially for its disturbing atmosphere.
17
Liner notes from the initial American release of Brian Eno’s “Music for Airports / Am-
bient 1,” PVC 7908 (AMB 001), 1978.
Immersion 155

This direct link to the environmental spaces is crucial, and one that is greatly
helpful to level designers, since they are the designers of the game’s spaces, to-
gether with the art department. With that in mind, a number of immediate ex-
amples jump to the fore:
• ambient sounds,
• ambient lighting,
• particle effects18 (steam, rain, etc.),
• props,
• fauna and flora.
These are just a number of straightforward examples that are easily identifiable
but nonetheless need to be part of the overall spec.

Ambient sounds
We all know of a number of aural game clichés; on top of a mountain we hear wind,
in caves we hear drips, etc. However, these kinds of straightforward implementa-
tions are not just representations of sounds that need to be picked from a list to
complete a level’s outstanding tasks. Recorded sounds not only have documenta-
ry qualities, they also have psycho-acoustic qualities, which means they have the
ability to have an effect on the psychological interpretation of sound. Machine
hum can be made to sound calming, a purring engine, or it can be made to sound
unnerving, an infernal machine. This is a subtle but very powerful concept. Choos-
ing ambient sounds smartly gives the level designer a way to manipulate player
experience in a profound manner without things appearing overly directed.
(Much more on this in Chapter 11.)

Ambient lighting
Normally when we speak of ambient lighting we are talking about using the
lighting naturally present in the environment, which can be natural (sunlight) or
mechanical (room lights). In film and photography this is often not enough and
additional lighting is added to the scene by the photographer or camera person.
This artificial lighting is often recognizable as such, but there are tricks to make it
appear natural or ambient. This is very relevant to level design, since all lighting
in levels is artificial. The ultimate goal of lighting, beyond illuminating a scene, is
to support it. Sometimes that means making sure the gameplay works well, but
there is also a large part reserved for enhancing a level’s ambience. This is mostly
done by making sure that where possible, lighting is emitted by believable light

18
A computer “particle” rendering effect that can simulate diverse phenomena like fire,
rain, steam, smoke, sparks and others.
156 Emotional Feedback Systems

sources, from the sun to room lighting, glowing panels or monitors, or whatever
else can be used for this goal. It is surprising how effective it is to exaggerate weak
light sources in a room to create a natural feeling. Even if the light sources would
never produce this much light, the brain gets tricked into believing it.

Particle effects
An easy way to add to the natural ambience of an environment is by logical and
consistent use of particle effects. This can encompass anything from believable
rain, smoke and fire, or the old favorite; steam coming from pipes and vents. The
best use of this goes beyond the merely decorative. It can suggest much about
things that aren’t there (a working electricity net) or strengthen the impact of
things that are present; an animated model of a train will look much better in the
environment if it emits steam and smoke while sparks fly off its connection with
the energy grid.

Props
Levels need to be populated as well. Look round the environment you are in
right now, chances are that there are all kinds of objects scattered around. News-
papers, junk, furniture, crockery, you name it. Levels that don’t include this are
sterile and lacking in ambience. Even if the player doesn’t know what exactly the
problem is they will still notice that something is lacking.
It is even better if we can incorporate props into gameplay. We have seen
the use of things like crates so often now that it is almost humoristic, but surely
we can use other objects to achieve similar results. Props can be used for cover,
missiles, barriers, rams, artificial steps and many other things. Just a single useful
object can add to ambience, strategy, interactivity or symbolism… not bad for
a humble prop!

Fauna and flora

It almost goes without saying that a lived in world makes for a more believable,
and generally a more enthralling one. Curiously there is often more time allo-
cated to inanimate objects, or the consequences of habitation that actual habita-
tion itself. Incorporating elements of fauna (animal Life) and flora (plant life)
can help suggest a whole eco system, and a living breathing eco system suggests
a fully functioning world in which players can lose themselves.
As often in this book I want to note that there is no point trying to create an
actual ecosystem, or anything even approaching that kind of scope. There is no
need to do so as the player’s mind can be tricked into accepting small manifesta-
tions of one as sufficient evidence that the world is “real”. Showcasing elements
Immersion 157

of life that are independent of player action, suggest that the world only takes par-
tial notice of the player, which suggests that the world is bigger than the player’s
imagination. This often subconscious realization in the player’s mind is often
enough to generate a strong desire to suspend disbelief, as enough evidence is
presented that the game world is complex, coherent, and exists in its own right.
If approached in such a manner the level designer can consciously manipulate a
subconscious process that betters the player’s gaming experience. A number of
examples will be given in the practice section dealing with this topic, following
the current chapter

Atmospherics
While ambience is defined as part of the environment, atmospherics are en-
hancements or additions to the environment or to the experience in general.
Sometimes they overlap; an environment can be made to be atmospheric, often
however they are very different, at least in their approach. They share the same
goal however, immersing the player deeply into the gameplay experience and
thus aiding a suspension of disbelief. Atmospherics can be roughly divided into
atmospheric additions, and atmospheric enhancements.

Atmospheric additions
We speak of atmospheric additions when we add elements to the game world
that are not normally native to it, in order to add to the desired atmosphere for
the level. No definitive list of examples can be created for this as so many unique
additions can be envisioned depending on genre, platform, level, or game type. I
can however provide some typical examples.

Music
A very effective example of adding something that is not native to the game
world, yet which still deepens immersion, is the use of music. One only has to
look at film to see the potential impact music can have to a scene. This subject
will be covered in great detail in Chapter 11.

Voiceover
The use of voiceovers in games is not without precedent and can be very ef-
fective. A good example of this can be found in ID Software’s Quake 3 Arena,
where the gameplay has been turned into a tournament sport, complete with
voiceovers provided by a commentator who remarks on several in-game events.
158 Emotional Feedback Systems

Voiceover can be used in other game types, as well. A mystery game grounded
in classic hard-boiled detective fiction may benefit from a film-noir-inspired voi-
ceover incorporated in the levels, or a historic strategy game may include a sage-
like voiceover, providing extra weight or mystery to the level’s events.

Cutscenes and scripted events

One of the most often-used atmospheric additions is the cutscene. In video games
cutscenes are sequences, often cinematic, where the player has no control over
the onscreen events. In many ways, it is hard to imagine anything more intrusive
than being forced to go through a cutscene, although these day semi-interactive
cutscenes are being used more and more, which greatly improves things. It is es-
pecially in the latter where this can be used to deepen immersion in level design.
Together with scripted events, an in-game cutscene (as opposed to one that takes
control away from the player) can be extremely helpful, since it allows the level
designer full control over a level event. If this is done with some forethought, the
level designer can even use gameplay-related level cutscenes or scripted events
and incorporate elements into them that help deepen immersion.
Note that you should always try to avoid cutscenes where major decisions
are taken away from the player; this is often judged by the player as too much of
an intrusion.

Atmospheric Enhancements
We speak of atmospheric enhancements when we take already existing elements,
native to the game world, and enhance them to provide a more atmospheric
experience. This kind of enhancement is related to ambience, but a subtle dif-
ference occurs, insofar as that ambience wants to emphasize, or become part of,
the environment, while atmospherics want to enhance what is already there to
strengthen its impact. Sometimes these things overlap, and sometimes one fol-
lows the other. As with atmospheric additions, atmospheric enhancements are
too numerous to cover in exhaustive detail, but we can look at typical examples:

Lighting
Level design lighting performs two functions:
1. It determines what the player can see.
2. It colors the perception of what players see.
The first function has many practical secondary effects, including matters of di-
rection or misdirection. This will be covered in other chapters, notably in Chap-
ter 10, “Visual Experiences.”
Immersion 159

On the other hand, the second function has application in the area of im-
mersion. As is clear from the fields of film and photography, lighting can com-
pletely change the impact of a scene. Imagine a normally lit room containing a
character, a living area, and a bed in a typical apartment complex. This scene is
functional but nothing special. Now imagine the same room like this:
The bedroom is covered with dancing shadows cast by a slowly swinging
flickering light bulb above an unmade bed, while the living area is only slightly
illuminated by cold moonlight falling in through the only open window,
showing the backlit silhouette of an emaciated person. A soft droning sound
can be heard in the background.
Suddenly we have atmosphere in abundance, at very little cost to the level de-
signer. The use of lighting deserves its own chapter, but for now it suffices to say
that level design can really shine19 if it is applied well.
Consider this eloquent description of the power of lighting and the need to
reach a state where the audience suspends its disbelief:
Stage lighting was once the supreme manifestation of manipulative am-
bience: it paints the scene; it changes to adjust the mood for the audi-
ence while they just sit there. And that of course is its ultimate limita-
tion: in the darkened auditorium the audience remains detached and
unimmersed, never—in the famous phrase about yielding to staged
reality—fully “suspending disbelief ” about those onstage golden veran-
das, spangled dancing levels, bluish dramatic storms, and rosy dawns.20
Lighting can have a huge role in establishing the right environment for the player
to truly enjoy the game.

Game camera
Many games, but by no means all, allow a certain amount of control over the
game camera. This can include camera placement, lens type, field of vision
(FOV), direction (yaw, pitch, roll), camera transitions, and other aspects. It can-
not be emphasized enough how greatly bad camerawork can hurt a game, or in-
versely, how greatly good camerawork can enhance a game, including in the area
of immersion. Take survival horror as an example, including games like Resident
Evil or Silent Hill, and imagine them with simple unimaginative camerawork. Do
you think these series would still be as big as they are now? Furthermore, even
if this is not apparent, for example in the case of first person shooters, there may
19
No pun intended.
20
Nathan Silver, “The Suspension of Disbelief,” Metropolis Magazine, http://www.metro
polismag.com/html/content_0501/chn/index.html, May 2001.
160 Emotional Feedback Systems

still be possible uses. For example, what if the level designer can place triggers in
a level, such that when the player sets them off, this subtly changes the depth of
field of the player camera?
Note that I strongly urge any level designer working on a game where this kind
of control is theoretically possible to make a strong case from the beginning that this
control is indeed made available. In an ideal world this includes at least one dedi-
cated programmer and a good toolset to implement and try out camera work.
If a complex camera is needed for the game, but it is not worked on and
respected from early on in the project, you are in real trouble.

Dramatic scale
Much atmospheric mileage can be derived from good use of dramatic scale. This
goes beyond matters of just art direction, but much can be gained by building a
strong relationship with the art department in general and concept artists in par-
ticular, since it is their job to enhance the game through visual techniques. Don’t
be afraid to submit a request for concept work for the levels you are working on,
with specific requirements about atmosphere and immersion. For example, it
is perfectly fine to sit down with a concept artist and talk through the general
gameplay progression of any particular level. You could then add a request at the
end, which might look something like this:
All the way through the level, I want the player to be confronted with a contin-
ual sense of unease and dread, even if this is subtle and in the background.
Now that the artist knows the level progression and has a specific atmospheric
goal to aim for, he or she may come back to you with all kinds of ideas that you
would never have thought of. This coordination between level design and art is
especially important when it comes to matters of dramatic scale.
A good example of this is found in Sony Computer Entertainment Japan’s
Ico, in which players are constantly reminded of their own vulnerability and the
size of the task at hand by the massive and overbearing architecture. This gives
the game an oppressing weight and transforms the environment from what is
basically an excuse for platforming actions to something that almost amounts to
a character in its own right. It is hugely satisfying to be able to manipulate such a
characterful and impressive environment.

Theory Summary
In this book, I often argue that a level designer must be able to work to a spec
with limited resources at hand. The reality of commercial level design tends to
make this the norm rather than exception. It may seem strange, then, to focus
Immersion 161

this much attention on something that most game producers will not recognize
as a key area of interest. To some degree this is to be expected; the basic-level
framework needs to be guaranteed first. Often the question asked of the level de-
signer when he or she presents ideas intended to deepen immersion and broad-
en the suspension of disbelief is something like: “Can the game ship without it?”
The answer to this tends to be “I suppose so.” But is that really the right question
to ask? What happens if we change the question into a more general one: “What
if our levels aren’t immersive enough?” If somebody doesn’t see the danger in a
situation where there isn’t sufficient player immersion, then that person is likely
ignoring a very important question: “Who wants to play a game that does not
engage enough to be immersed in?”
From both a commercial point of view and an artistic one, it would be a
disaster for the game if it lacked this crucial element.
There are uncounted ways of making sure that a level provides an experi-
ence that is deep and immersive. An invaluable tool is provided by Mihaly Csik-
sczentmihalyi’s flow theory, especially when it is translated to optimal game ex-
periences. Equally so, we can learn from theatrical or filmic principles. Look at
intrinsic game logic and consider ambience and atmospherics. What stands out,
however, is that immersion is not separate from gameplay, it is part of it. An im-
mersed player is a happy player, and a happy player is much more receptive to
what the game has to offer. It is of the utmost importance that level designers
recognize this, incorporate immersive principles into the level’s design and im-
plementation, and fight for the level’s integrity in this regard when people start
to chip away at it.
The next section will describe a number of practical and useful examples that
level designers can adapt or adopt, but the whole of this book should provide in-
spiration, as well as other examples. This particular topic is intertwined with so
many other level design goals and areas of discussion that it is worth keeping in
mind all the way through the level design and implementation process.

Practice

Example 8.1: Immersion—Historic Grounding

Summary
Environments that are very believable in the way they have been realized and de-
tailed can be especially immersive to a player. The more convincing a gameplay
162 Emotional Feedback Systems

environment is in its specifics, the easier it is for the player to accept the environ-
ment as a whole, including in other areas. One of the areas that can help foster a
general sense of immersion is providing an environment with a past.

Game Genre
The technique is suitable for most games that use non-abstract environments.

Goals to Achieve
• Deepen immersion.
• Provide historic grounding.
• Assist a sense of agency in the game world.

Description
(Example type: Original)
The passage of time touches everything and everybody. In the real world this
happens automatically: things suffer wear and tear, people grow older, the weight
of historic events can be felt in the politics of the current day, and so on. In a vir-
tual setting like a game level, temporal impact has to be created. This means we
can have an influence over the perception of history and thereby influence the
immersive qualities of a level.
A place with a sense of history feels more real than one without. After all,
how can there be a history to something that isn’t real? If we suggest that past
events have had an impact on the environment, we suggest that current events
can also impact on the game world. This in turn aids the sense of agency a player
feels towards the environment.
This example will take a hypothetical environment and suggest some ways
in which past events can be suggested. The chosen level environment is a small
American town.
Mixing old and new. One of the more straightforward ways to suggest past
events is to make sure that there is a visible temporal layering of content in the
environment. What I mean by that is that we can include content that dates back
to different time periods in order to show the passage of time. In this case for
example we can use architecture to date the town. We can include modern build-
ings alongside older ones, a modern bank next to a classic 50s diner. A church
can display the date it was built, as can the town hall, on a foundation stone.
Historic props. We can go further by including things like a derelict building
that, on investigation, awards the player with newspapers stemming from the
Immersion 163

time the building was last occupied. A diary of a previous owner can be found
giving an account of events over a specific timeline.
There are less work intensive ways of creating a past as well. Street signage
may be used to date the town to a specific event or time period by consistently
incorporating references that the player knows to be from a specific period of
time.
Ripple effect. We can greatly improve a sense of history by showing the impact
of a past event throughout many aspects of current life in the town. This can
be seen as a ripple effect through the fabric of time and how it has changed the
town. Our town may have experienced a large meteor impact in the 1950s, af-
fecting almost every aspect of life from then on. The impact site itself may offer
great gameplay opportunities, for example by having been transformed into a
somewhat amateurish and spooky tourist attraction. Subtler little details can be
adopted as well. The diner for example may carry a meteor themed menu or be
named after the meteor.

Further Notes
There is not much that limits level designers when it comes to this kind of tem-
poral grounding other than time and budget. It makes sense, therefore, to incor-
porate these kinds of things into actual gameplay related scenarios, and not only
use the history of the place for atmospheric qualities.

Example 8.2: Game Camera and Immersion

Summary
One of the ways in which we judge our environment is by our visual percep-
tion of it. We form a mental image and make certain assumptions based on the
evidence presented to us by our eyes. In level design we potentially have control
over the game camera, which means we can influence the players’ perception of
their environment. This can be done in strong immersive ways.

Game Genre
The technique is suitable for games that employ a camera system that can be
influenced by the level designer.

Goals to Achieve
• Aid immersion through inventive game camera use.
164 Emotional Feedback Systems

Description
(Example type: Original/real)
Safe distance establishing shot. Sometimes it really pays off to give the play-
er a grand view of the level before they get to explore it. It can set a powerful
mood and allows the player to build a mental map of the environment right from
the beginning. Unfortunately, this is often made impossible due to technical and
practical limitations. The scene may contain too much visual data to render and
cause performance issues, or early access may ruin level design scenarios sched-
uled later. Whatever the reason, this can still be made to work by presenting the
player with a particular view of the environment, not necessarily the actual en-
vironment itself. As long as the player can see the environment, it will have the
desired effect. This can be done, for example, through a window or a fence or any
other believable barrier between the player and the view.
For extra effect we can include vantage points that overlook each other, so
when a player eventually reaches a view they had witnessed before, he or she
can be made to appreciate this more if it is possible to see the original vantage
point from the new location. It can be great to have a panoramic view of a mas-
sive tower, and then hours later be able to look back on where you came from
standing on top of the tower in question. Excellent examples of this can be found
throughout the game Ico.21
Special angles. At times it is possible to create a deep sense of immersion by
employing special camera angles. The game camera can be tilted to indicate the
player’s deteriorating mental state in a horror game, the camera can zoom out to
suddenly show an approaching danger, or the camera can move into a top down
position to reveal something about the environment, for example the aftermath
of a grueling conflict between the player and a group of enemies.
Many other possibilities exist.
Camera shake. Relatively easy but no less effective is the inclusion of camera-
shake at key moments. Imagine a level set in a series of underground tunnels
during a long bombing raid. A sporadic camera-shake accompanied by a low,
far-away rumble, and possibly some falling dust created with the game’s particle
system, can be incredibly effective in creating a deep level of immersion.

Further Notes
If at all possible, examine the use of camera lenses to provide further effects. It is
remarkable how much lens types, depth of field, color filter, field of vision (fov)
21
Published by Sony Entertainment Europe, developed by “Team Ico” in Japan, released
March 23, 2002.
Immersion 165

and other aspects can contribute to a level’s immersive qualities. Yet, very few
level designers have been able or are allowed to make use of those possibilities.

Example 8.3: Ambient Factors—Fauna and

Flora
Summary
If we can create levels that feel like they are stand alone, autonomous, living and
breathing environments, we provide the player with a much more immersive ex-
perience. This process can be helped by providing an ecosystem that the player
can observe and potentially interact with.

Game Genre
This is a useful technique for games that allow wildlife to be scripted.

Goals to Achieve
• Make the level environment more immersive.
• Encourage immersive deepening gameplay.
• Provide spontaneous or emergent gameplay in levels.

Description
(Example type: Original)
Catch the creature. If the game you are working on contains wildlife, or has
the scope to contain wildlife, it is a good idea to include a creature that the player
can have some fun with. Take, for example, a creature that needs to graze and will
wander over to specific patches of vegetation in order to do so. It may move from
patch to patch and otherwise mind its own business. Chances are the player will
eventually try to approach the creature. Make sure the creature has some rudi-
mentary AI that will make it run away when the player comes too close, running
too fast for the player to catch up. I can guarantee that a large group of players
from then on will try to find ways in which they can catch the creature. They may
try to ambush it or herd it into a pen or find some completely unexpected way
of catching it.
This kind of spontaneous fun directly linked to the level environment really
adds to the immersion of the level. Ideally, this is rewarded by allowing skilled
players to indeed capture the creature, in which case the player needs to be re-
166 Emotional Feedback Systems

warded—for example, with a tuft of fur that can be used elsewhere in the level,
or any other award that fits the game’s specific circumstances.

Further Notes
This kind of mechanic can work on several levels. Flytraps can contain useful
substances but the player needs to figure out how to approach the plant without
it closing its petals and denying access. Birds can circle above areas of interest.
Glow-worms can provide unexpected illumination at night, unveiling secret ar-
eas, and so on.
Negative Emotions
9
T here are times when conventional wisdom doesn’t apply. Normally we don’t
set out to frustrate an audience, or make them feel bad. We don’t want them
to feel negative emotions. In most cases this is true, we try to enlighten or enter-
tain. What happens, however, when we try to do that through the use of negative
emotions? When we use emotions like frustration or anger as a tool to achieve a
higher goal? That is the subject that this chapter will explore in more detail.

Concept
People like being scared.
On the surface, this is a slightly strange statement. Why on earth would anybody
enjoy feeling scared? Most people lead their life in a way that reduces risk and
fear as much as they can. But on the other hand, many people enjoy such things
as ghost stories, thrillers, and scary movies, so there must be something about
feeling scared that appeals to people.
Let’s therefore rephrase the previous statement slightly and say: People like
dealing with their fears in a safe environment.
This sounds a bit more acceptable. If there is no real risk involved, there is
nothing to worry about. This leads to a conundrum: if people like being scared
in situations without real risk, what logical reason do they have to actually be
scared in that context1?

Unreal Risk
The key to answering that question lies in the observation that there can be such
a thing as unreal risk, which on the surface seems a contradiction in terms. How-
1
There is after all no risk.
167
168 Emotional Feedback Systems

ever, this concept is sensible if one considers the human capacity for suspension
of disbelief, which as we know from the previous chapter is a massively important
ability that human beings possess. We have the ability to suspend our disbelief
and indulge in artificial scenarios that are completely fictitious. This allows us to
deal safely with concepts that, if they were physical and real, would be harmful
to us. In other words: in this case, suspension of disbelief allows us to experience
and deal with unreal risks. We can see now that this constitutes a clear example
of a positive use of a negative emotion.

Enjoying the Negative

We already know that sometimes human beings artificially create and experi-
ence scenarios that involve negative emotions, for example by listening to ghost
stories or by seeing a scary movie. We even talk about some of these experiences
in positive terms, like the “thrill” of being scared or the “lure” of the dangerous
wild. Some of this harks back to basic principles of play, insofar as some aspects
of play have to do with learning about and preparing for dangerous situations. It
seems fair to say that:
People have the capacity and the need to deal with negative emotions
and experiences, and they can do this through applying suspension of
disbelief to artificial situations where such emotions can occur in a con-
trolled and safe manner.
This principle occurs throughout human society, time after time. We have al-
ready encountered it in the discussion of reward systems in Chapter 7 where we
found it at work in the principle of catharsis. Throughout this current chapter, we
will examine a number of areas and examples that pertain to the subject matter
of this book.

The Example of Fear and Other Negative

Emotions
Fear is an emotion that has been studied by humankind for thousands of years,
so we actually know a little bit about the subject. Throughout history, it has dem-
onstrated a capacity to both paralyze and motivate people. At a basic level we can
argue that scary films and books are about facing fears. There are certainly many
stories about some kind of phobia or other deep-rooted fear, so we know that at
least sometimes this applies. Fear is just one example, however. There are many
other negatives in our lives that demand their own coping mechanisms. Other
Negative Emotions 169

examples that can be included are emotions dealing with panic, doubt, anger, or
frustration, to name just a few. Do they all have a place in this concept where we
absorb negative emotions into an artificial scenario of our own making and turn
them into a positive experience?
Let’s go back to the example of fear. Countless books have been written on
the subject, millions of songs are inspired by it, many classic films are based on
it, and countless paintings have been created because of it. It cannot be denied
that fear is a powerful emotion that has stood at the basis of all kinds of creative
expression. But it is certainly not the only one.

Art or Entertainment as a Coping

Mechanism
Works of art and entertainment often deal with negative subject matter. It is un-
deniable that a great diversity of artworks through the ages have been created
that would not have had their impact on society without delving into this ad-
mittedly negative, but rich, vein. Can we imagine a happy, carefree version of
Munch’s The Scream? Or of Dante’s Inferno?
I will not try to create any definition of what art is or what it should deal
with,2 but it is fair to say that at times creative expression incorporates negative
emotions and provides a relatively safe way for the audience to deal with them or
learn to understand its own personal instances of those negative feelings. Seen in
such a way, the use of negative emotions becomes a powerful tool in the hand of
the artist, a tool that allows the artist to create a profound emotional response in
the audience. Sometimes it even assists the audience in devising coping mecha-
nisms for dealing with such negatives.
This can be a powerful incentive for people to be exposed to negative con-
tent, things they would normally want to avoid. Artists have become quite adept
through the ages at devising methods to deliver these positives through negative
means. Let’s look at a number of known techniques to see how they are imple-
mented in practice.

A Deal with the Devil (Committing the

Audience)
There are multiple ways to keep an audience interested, generally through en-
tertainment and engagement, in the traditional positive sense. But something
2
I am not that foolish.
170 Emotional Feedback Systems

strange happens when it becomes clear that the audience is exposed to an ex-
perience more akin to an ordeal. If this is done with skill, the artist will not lose
the audience, but instead will enter into an agreement with it. This can partly be
explained by the natural fascination people have with certain dark areas of life.
Sometimes this is a healthy curiosity in areas that are genuinely interesting, for
example, a will to understand something about a painful subject. At other times
this is a more basic desire to be entertained by things like shock and revulsion.
Often it is not clear where one ends and the other begins. But what is clear is
that the audience is willing to go quite far with the artist because of the implicit
promise of a worthwhile experience. If the audience is ultimately rewarded well,
it will not resent the difficult route in getting there.
This is true in acceptable arenas like film, where a narrative about vio-
lent, reprehensible men can garner great critical and popular acclaim (Good-
fellas3) or gruesome horror can be elevated to a celebrated film art (Alien.4)
It is just as true in more obscure or specialist arenas like performance art,
where people like Marina Abramovic have been doing recognized work for
a long time, sometimes in ways that go beyond what is normally accepted by
an audience:
In Rhythm 10, she plunged a knife between the spread fingers of one
hand, stopping only after she had cut herself 20 times. Having made an
audio recording of the action, she then played back the sound while re-
peating the movements—this time trying to coordinate the new gashes
with the old. Using her dialogue with an audience as a source of energy,
Abramoviç created ritualistic performance pieces that were cathartic
and liberating.5
All of these examples show successful challenges to the audience where the artist
manages to make the audience commit to dealing with extraordinarily negative
subject matter.

Drugging the Audience (Adrenaline Rush)

I personally am incapable of enjoying a roller coaster ride. Even while I was
typing that sentence, I was getting an uncomfortable feeling in my stomach
just imagining the possibility.6 Having said that, I do understand the attrac-
3
Directed by Martin Scorsese, 1990.
4
Directed by Ridley Scott, 1979.
5
“Marina Abramovic: Rhythm 5. 1974,” Guggenheim Collection Online, http://
www.guggenheimcollection.org, 1994.
6
I suffer badly from vertigo.
Negative Emotions 171

tion in general. It is the same attraction that causes people to go wild wa-
ter rafting, bungee jumping, carting, paint-balling or participate in any of a
multitude of genuinely exciting activities. All of these acts are guaranteed to
flood the participant’s system with an intoxicating mix of endorphins and
adrenaline. People often literally describe this feeling as an adrenaline rush,
or even just as a rush. This rush can be so potent that people get addicted to
it, and they literally turn into adrenaline junkies. On closer examination, it is
very interesting to note that all of these activities carry some kind of potential
danger or penalty, which is the reason why they so easily cause a release of
adrenaline in the blood. Nothing like a bit of danger and fear to get the juices
flowing, right?
It is not a huge jump to take the kind of chemical award linked to these
dangerous activities and recognize the value this can have in creative expression.
Is it possible to challenge the audience with comparable negatives in such a way
that it causes the release of natural drugs like adrenaline? It is hardly surprising to
find that it’s not only possible; the use of this principle is in fact ubiquitous. This
can be seen clearly even in the language used to market these works of negative
entertainment. We have all seen movie posters and ads that scream things like:
“A truly shocking story! A terrifying film!” and so forth.
I had a teacher many years ago who explained to me why he disliked ten-
sion in novels. It was mainly to do with the fact that readers can get so drawn
in that they race through the book, ignoring the nuances of style and language.
This taught me a valuable lesson, completely counter to my teacher’s original
intention, about the role of prose versus the desired overall goal of the book,7
and one that I suspect is shared by authors of the exciting books he so dis-
liked.

Learning through Shock (Forced Focus)

Most artists want to make a point with their work. They want to teach the audi-
ence a certain lesson, or let it focus on a particular theme, or perhaps make it
think about a certain contradiction in society. Whatever the exact motivation,
they want to get something across in an impactful way. This can be done through
seduction with beautiful imagery, compelling music, beautiful prose, and many
other positive means that are known to work. It can also be done by shocking
the audience in such a way that people automatically give all their focus to the
creative work, and as a result of this sudden focus and the accompanying physi-
ological processes, the audience will be much more receptive to absorbing the

7
It is legitimate to make prose subordinate to story and impact.
172 Emotional Feedback Systems

presented content. It is well known that shocking or stressful events can lead to
enhanced memory function8.
Realizing that this side effect of shock can cause such a singular focus and
promote knowledge retention makes it easy to understand that there can be
many uses for it in artistic expression. If the artist wants to focus attention on a
new character, an effective way of doing so is to introduce him or her by showing
the villain committing a truly shocking act. This immediately will lead to a num-
ber of questions and doubts in the mind of the audience, which may be benefi-
cial to the work as a whole. Who is this person? What motivates this person to act
in such a way? Will this shocking behavior continue? Is this person redeemable,
or truly evil? If any of these questions are answered early on, just after the initial
shock, chances are that the audience will remember this for the rest of the work.
This principle applies to all information that the artist may wish to assign extra
significance to.

Dramatic Impact
In Chapter 7, “Reward Systems,” we have already gone into some detail on how
catharsis can be used as a formal reward system. I won’t repeat that content too
much, but I would like to reiterate that some of the earliest interpretations of the
concept come from Aristotle’s Poetics, where catharsis is mentioned as one of the
key elements of drama. I would like to note, however, that there are many more
ways to use darker subject matter to create dramatic interest. If we look even
further back in the book, a typical example presents itself in Freytag’s pyramid,
which was explained in Chapter 4. As noted there, in the climactic turning point
of a tragedy, a change occurs that is negative to the protagonist. This has all kinds
of potential dramatically beneficial consequences. The audience may feel pity for
the plight of the protagonist, and the change may give the protagonist emotion-
ally sound motivation and provide context for his or her future actions. Many
other advantages can be gained from one well executed shocking or disturbing
moment or development. It is easy to see why this is attractive to artists in almost
any creative discipline.

Concept Conclusion
Not only do we see that the use of negative emotions is viable as a creative tech-
nique, we find that it flourishes throughout diverse disciplines and can incorpo-
8
Much work on this topic has been done by James McGaugh. See http://darwin.bio.uci.
edu/neurobio/Faculty/McGaugh/mcgaugh.htm.
Negative Emotions 173

rate almost any type of negative emotion. What seems to be clear, though, is that
there is always a process of give and take involved, where audience members are
willing to be presented with these negatives, but only if they can get something
out of it themselves. Viewed in such a light, we can say that this technique of
creating negatives to produce positive emotions as a final result acts no differ-
ently than many of the other reward systems discussed earlier in the book. But
this does not mean that there are no reservations when it comes to adopting the
use of these kinds of things in creative expression. There are a number of dangers
associated with it, as we will see.

Inherent Dangers
As we all know, there is no such thing as a free lunch, and this definitely applies
to the above techniques. There are a number of real dangers involved in applying
these kinds of methods of playing with negative emotional impact. I will list a
number of the major ones.

Audience rejection
The first one is quite simple. The audience may not wish to be exposed to this
kind of manipulation, or to the particular incarnation of it that the artist chose
to use. As turnoffs go, this is an extremely strong one. It can do real damage to
the artist-audience relationship, even to the point of being irreparable. This is
of course something that must be avoided, unless you want to press the self-
destruct button.

Overshooting
Another danger comes from the possibility of overdoing it to such a degree that
the main purpose is not met. In this case, the audience will focus so much on the
particular technique employed that it is to the detriment of the desired result.
There is no point to shocking an audience if people are left numb, or completely
engrossed in the shocking event itself. This quite often happens to people who
mistake the technique for the goal and forget that there has to be something in it
or the audience as well.

Some ethical considerations

A further danger that presents itself is ethical miscalculations. Does the artist ac-
tually have the right to use this kind of negative emotional manipulation? There
are of course many very successful examples of celebrated artists who use these
techniques in their work. On the other hand, most people will find it hard to
174 Emotional Feedback Systems

have anything positive to say about the countless exploitative or even criminal
examples of art or entertainment that also exist in our environment. This in-
cludes films that revel in pointless depictions of extreme violence or build on
tired racial stereotypes, and art that shocks without rationale and has nothing
further to offer. I am sure you can think of your own examples to illustrate the
point. This is the flip side of the power of negative art and entertainment.
How does an artist know if he or she has gone too far, or lacks the skill
to pull things off without causing real harm to the audience? Does the audi-
ence have an inherent right not to be subjected to certain types of manipula-
tion?
These are real questions that are still asked daily and can create massive,
even international, controversy.9 Although I personally take a strong libertarian
stance in many of these matters, I still have to define my own perimeters on what
I personally see as acceptable use. I am not going to explore the topic of ethics
or freedom of speech in this context, but I do think it is important to highlight
that there is a certain amount of personal responsibility involved when dealing
with these issues. This is something that artists have been struggling with since
classic times, and it is important to try to be aware where one’s own line in the
sand is drawn. I highly recommend that you do the same, even if it forces you to
re-examine personal notions of artistic freedom of expression.

Theory
In the previous section I made the following observation:
People have the capacity and the need to deal with negative emotions
and experiences, and they can do this through applying suspension
of disbelief to artificial situations where such emotions can occur in a
controlled and safe manner.
If we try to view these concepts through the lens of video game development,
we once again find that these concepts apply, and that video games are perfectly
capable of dealing with serious issues and emotions. It isn’t too hard to find clear
examples: we fill games with scenarios that in real life would be undesirable but
somehow seem appropriate to the game. Often a player is confronted with levels
that feature dangerous action, war, disasters, fearsome creatures, and so on. The
list is very long indeed, and as diverse as the subject matter of other art forms. It
9
See the negative reaction to Salman Rushdie’s Satanic Verses, published in 1988.
Negative Emotions 175

is clear that we already force the player to deal with negative experiences. A video
game is by definition an artificial environment. Players have to suspend their dis-
belief in order to immerse themselves in the game world and accept its rules.
Because we allow the player a certain sense of control over the situation, and the
player deep down knows there is no real risk involved, the experience is ultimate-
ly a safe and controlled one.
So we find that the earlier observation contains many concepts that video
games, and consequently level designers, already incorporate. In fact, there is so
much overlap that it becomes clear that it is almost too snug a fit.
So how does all this help us? We know that on a conceptual level, negative
emotions can be and are used in video games just as much as they are in other
art forms. But, what kinds of techniques are used? Are there other more suitable
ones unique to games? How do we get the best use out of them, and are there
game-specific dangers we need to be aware of? These are the kind of questions
we will be trying to find answers for in this chapter.

Genre-Specific Examples
In many ways, video games are especially suited to the use of these techniques.
This isn’t just because through level design we have authorial control over the
content of a level. It is also because to a large degree, video games give con-
trol to players, often in order to test their skills against the challenges inher-
ent in many games. Players have a certain expectation that they will be tested
in a video game, and within certain genres they expect to be pushed very far
indeed. Although the techniques in question are applicable to most games,
there are a number of game genres that lend themselves especially well to them
because they are part of the core gameplay experience. Let’s look at some clear
examples.

Survival Horror
Within the survival horror genre, it is part of the expected framework that the
player is underpowered and up against dark and disturbing forces. This means
that the player will not resent it if confronted with unspeakable horrors or game-
play moments where it feels as if the player is at a huge disadvantage. Everything
in the game world can be used to scare or disorient the player. This includes cam-
era work, audio, architecture, props, cutscenes, and of course enemy encounters.
In other words, this game genre is more geared towards using negative emotions
than any other.
176 Emotional Feedback Systems

Shootemups (SHMUPS)
Classic shootemups allow things that almost no other genre can get away with.
They allow for a fundamentally different approach to difficulty and the learning
curve. Although not true for all shootemups, there is nonetheless a large group
of players who want to be severely tested and are happy to be frustrated and pre-
sented with what at first seem like unfair circumstances. This is accepted because
much of the attraction of this genre comes from mastering nearly inhuman skills
and reflexes and using them against a merciless onslaught of wave after wave
of enemies. It is interesting to watch skilled shootemup players. They somehow
manage to navigate a screen that is literally filled with bullets and enemies, ex-
cept for a tiny moving area that the player always seems to inhabit.

Retro Games
Retro games are a special category, insofar as that they are not necessarily played
for normal gameplay reasons. As the name says, they are played more to provide
a gamer with retrospective and can be quite a nostalgic affair. Remakes of classic
retro games are faced with the paradoxical situation that were they to improve
too much on the original’s gameplay, they could be criticized for not being au-
thentic enough!
Unfortunately, some classic games feature incredibly frustrating and diffi-
cult gameplay. Many rose-tinted glasses have been shattered by the realization of
how bad or unfair many classic games really were. This type of harsh and nega-
tive gameplay was often employed to make sure that the player couldn’t survive
too long on the coin used to start playing the game in the arcade. To this day,
there are still players who relish this unfair challenge. If one wants to court them,
it is inappropriate to stray too far from these harsh concepts. This type of game
has become a genre of its own, where negative emotions as a level design theme
are completely acceptable.

General Level Design Application

As the famous Dutch soccer player Johan Cruyff once said: “Every disadvantage
has an advantage.” In the case of the techniques we are discussing, it makes sense
to adopt the same philosophy and see how negatives can be changed into posi-
tives in level design.
Since to a large degree we are the authors of the player’s experience, it is
almost inevitable that at some point we will be tempted to frustrate the player
or confront the player with some kind of negative experience that goes beyond
Negative Emotions 177

the regular gameplay challenges in order to achieve a specific result that cannot oth-
erwise be obtained. Other than the inherent goals derived from the typical game-
play associated with some genres like survival horror or some retro games, as
discussed earlier, there are also many techniques based on general principles.
We know from the earlier conceptual examination of this topic that the right use
of negative emotions can greatly enhance a creative work. What happens if we
try to translate these general principles to level design principles and methods?
Let’s have a look at the previous general observations and see if they fit into a
level design framework.

A Deal with the Devil (Committing the

Audience)
Previously I stated that:
... the audience is willing to go quite far with the artist because of the
implicit promise of a worthwhile experience. If the audience is ultimately
rewarded well, it will not resent the difficult route in getting there.
In some ways, this seems unnecessary in video games. Hasn’t the audience al-
ready bought the game? In other ways, this seems entirely sensible. Gamers know
that they will be tested throughout the game and that reward systems are at play.
Where this really matters in level design is dependent on the goal and the skill
of the level designer. If the level designer wants to create motivation in the player
to endure an extended period of hard gameplay, this can be done by providing
an enemy that is so horrific that the player feels compelled to go through great
lengths to defeat the enemy. If players are helpless to avoid a situation that causes
great distress and suffering, they be extra motivated to avoid this a second time.
These are level design decisions. The level designer can create scenarios
where a negative occurrence creates the commitment and framework to engage
strongly with the presented gameplay. As before, this is especially viable if the
reward offered to the player is worthwhile. Unlike in real life, the level designer
has much to say about the play awards. So it is fair to say that this technique
translates extremely well to level design.

Drugging the Audience (Adrenaline Rush)

In the virtual environments of a game, especially if they aren’t abstract, there is
a great opportunity to explore the psyche of game players. This may sound a bit
pompous, but if we look at what happens in video game levels, it is quite easy to
reach this conclusion. Aren’t we putting real people in something like a virtual
178 Emotional Feedback Systems

experience chamber? We poke them here and there, change the maze around
them, bombard them with surreal experiences, and manipulate their emotions
and expectations.
Within our palette as level designers, we have a huge amount of color to play
with in shaping a play experience that is ultimately beneficial to the player. One
of these colors comes from playing with feelings of fear and danger and turn-
ing this into a measured release of adrenaline. This can come from exploration
of phobias like vertigo or arachnophobia, or from more scare tactics and unex-
pected shocks. In fact, there are countless ways to achieve this result and find
new and original ways of doing so. Within level design, we can control gameplay,
sound, camera, AI behavior, and so forth. This is all the control we need to create
the circumstances to reach the desired goals.

Learning through Shock (Forced Focus)

In a good game with good level design the player is often kept within a state of
heightened awareness. In some ways this is a side effect of the optimal experience
of being in the flow channel10, (being in the zone in other words) as created by the
level design. The player is fully immersed and open to the game world. Further-
more, whether explicitly or not, the player is by definition, through playing a game,
in a system designed to educate.11 These factors combined suggest that there isn’t
much need for shock-induced focus, especially if it is designed to aid the retention
of information. To a degree this is true; since the player is already actively engaged
in the creative work, games have an advantage over other art forms.
On the other hand, this doesn’t mean that there is no place for this tech-
nique at all; far from it. Some lessons are hard to accept, some artistic points are
hard to make, and some rewards only come at a price. Add to that the fact that
the player is often in partial control of the outcome of the gameplay scenarios,
and one can be forgiven for sometimes seeing a need to force the issue. If viewed
in this way, level design is actually not so different from other creative art forms.
It is up to the individual level designer to see if there is a need, and if so, to choose
how far to take the application of this technique.

Dramatic Impact
On this topic I can be short. One of the areas in which we can make great strides
forward in our level designs is by making sure that the player’s actions are re-

10
See Chapter 8.
11
See Chapter 2 on level design as a teaching mechanic.
Negative Emotions 179

warded by dramatic impact. This is of course already done throughout games,

but unfortunately, the application is frequently blunt and ham-fisted. This is in
some ways very strange, as drama has been studied for thousands of years and
all this knowledge is available to us. We have already gone back as far as Aristotle
to describe the use of fear and pity as key components to create drama, which
literally is a textbook example of using negative emotions to create a positive
experience. But there are key differences between stagecraft and video games,
especially when it comes to building and expressing narrative. A level designer
doesn’t always have the luxury of linearity or a passive audience that politely sits
through the whole performance. Due to the interactivity of the art form, we have
to deal with some unique challenges that are at times hard to solve.
It is good to remember, however, that if level design is the stage on which
gameplay occurs, we aren’t just the scriptwriters, but also the sound engineers,
lighting people, acting coaches, set designers, and so on. If we create a moment
of dramatic tension, perhaps the loss of life of the player character’s main rival,
we can raise the camera to look down on the scene, have an NPC utter a suitable
line of dialogue, play the right music, create a gust of wind, and more. Then on
top of that we can create gameplay scenarios that feed off or link to the dramatic
context just created. And that last aspect, where the drama is linked to the actual
gameplay, where the player acts out dramatically interesting scenarios, is where
our work can shine in unexpected ways.

Other Themes
Many of examples of these principles have been applied successfully in well-
known games. It is a much richer and universal concept than generally expected.

Panic and stress

In Tetris,12 the blocks stack higher and higher, and they will never cease or slow
down. Nobody “finishes” Tetris; it cannot be beaten and it has no happy ending.
This is on the surface a terribly harsh premise for a game. Yet who can deny that
Tetris is one of the biggest success stories in game development?

Betrayal
One may wonder how this can ever be a good basis for gameplay. But this is in-
stantly understood when one witnesses a bout of multiplayer Chu Chu Rocket.13
The mayhem and backstabbing that occur in this game, in which players can
12
Developed by Alexey Pajitnov, released June 1985.
13
SEGA Enterprises Ltd., 1999.
180 Emotional Feedback Systems

actively interfere and sabotage each other’s chances of success, is truly epic. It
is very funny to see how easily even lovely people turn to the most backhanded
techniques without any guilt.
These are all viable examples, and many others exist. Some of these will be
described in more practical detail in the following practice section.

Dangers and Disadvantages

The idea that negative emotions like fear or panic can be a rich resource to be
tapped for the purpose of art and entertainment is a very powerful one. Its use,
though, is often dangerous and subversive to other, sometimes more important
level design goals.
The biggest danger in all of the techniques described is that by attempting to
deepen players’ experience through the use of negative emotions, they may end
up feeling negative about the game itself. The gameplay becomes a negative, is an
outcome that needs to be avoided at all costs. It is essential that players maintain
a sense of fairness and be allowed to trust the game. The simplest way to sum it up
is that we need to make sure that we don’t put players off by treating them harsh-
ly in an arbitrary and pointless manner. Instead, we need to make sure that play-
ers can experience negative emotions that help them enjoy or appreciate the game14
(as opposed to making the gameplay itself negative). This sounds paradoxical,
but techniques that work towards this outcome are used all the time in other art
forms as well as in other game forms. It is a viable and valuable technique as long
as the principles of trust and fairness are not violated.
However, the problem is that most gameplay does not really lend itself to
this approach. As we have discussed earlier, level design is much concerned with
teaching the player’s skills and testing those skills through gameplay challenges.
Fun is therefore often associated with success, and if that experience of success
is frustrated too much, players will end up resenting the game itself. This occurs
either because it is unfair, or because players feel they cannot master the game to
a sufficient enough degree to enjoy it. This is an important conundrum that is be-
coming more and more understood now that so many games involve mature and
emotionally challenging themes and gameplay psychology is better understood.
Some of these concepts are now sporadically appearing in design literature, as
the following example shows:
Under the hood of every game is the same simple mechanism: we give
the player a set of skills to master, and then run them through the paces,
demanding effective performance of that skill-set. To know that they
14
Or at least make it a worthwhile experience.
Negative Emotions 181

did well, the player requires feedback. If the player thinks they screwed
up, but the game doesn’t send a clear message to that effect, they will un-
derstandably be frustrated and put off. Not to mention that the drama
will not be read as drama. This means that player’s actions – and their
results – are not good contexts for setbacks or reversals.15
This neatly sums up a key problem facing us if we want to use negative emotions as
a reward mechanism in gameplay. It is something we have to be very careful about.

Purpose and Gameplay

There are certain dangers that are easily avoided, because we have seen them
annoy players in games of the past and can therefore avoid them in the future.
A classic misinterpretation of the technique, for example, is seen when a game
hobbles the player’s abilities without good reason. This used to be rampant
among others in survival horror games that employ so called tank controls, which
make it very awkward to control the player character. It is often argued that this
aids the game by inducing a feeling of panic, which is appropriate to the genre.
This introduced level of difficulty is so arbitrary, however, that it is tantamount
to cheating on the designer’s part.
It is not that the decision to make the player-character physically less capable
that is suspect, as this indeed can add to the tension in an appropriate manner. It
is that it is done in a way that is unfair and arbitrary. Why not give the player in-
tuitive controls, but impose a logical restriction on movement, in which case the
player won’t feel cheated. This is where level design can step in. If the complex-
ity of controls is minimal, and the challenge lies in negotiating the environment,
then much frustration will be forgiven, since it is now a question of player skills
and weaknesses, as opposed to player character weaknesses. We should always re-
member that there is no dramatic impact to be derived from pointless irritation. If
the gameplay challenges make sense through the level design, if there is a purpose
to the difficulties experienced by the player, then there is at least some respect for
the player’s plight. Nobody wants to feel at the mercy of arbitrary circumstances.

Theory Summary
We have seen that in addition to traditional level design challenges, general prin-
ciples of negative gameplay for positive results are also applicable to level design
theory. If we want to tackle mature and emotionally deep themes, we need to be
15
Ben Schneider, “Losing For the Win: Defeat and Failure in Gaming,” Gamasutra,
http://www.gamasutra.com/features/20070215/schneider_01.shtml, 2007.
182 Emotional Feedback Systems

able to use any techniques available to us in a responsible manner. There are both
ethical and practical dangers and obstacles to overcome, but we have seen that
there are many ways in which unwanted side effects can be mitigated or avoided.
If we do so successfully and appropriately, we have the means to make our lev-
els much more meaningful and rewarding, which is something that is definitely
worth pursuing.
Some of the discussed concepts may feel at times counterintuitive or even
paradoxical, but the following section, with its practical examples, should clearly
illustrate their use.

Practice

Example 9.1: Dramatic Impact through

Emotional Loss
Summary
Sometimes the player can be extra motivated or involved in the game by being
exposed to something that has great dramatic impact on the gameplay narrative
of the level. In the context of this chapter, we are going to look at using a nega-
tive emotional event in order to create a positive play experience, specifically
through subjecting the player to an emotional loss.

Game Genre
This is especially useful in games with important AI companions.

Goals to Achieve
• Foster an emotional bond with an NPC.
• Create motivation through negative emotions.
• Frustrate the player in exchange for increased commitment.

Description
(Example type: Existing game)
Ico: A case study. In their game Ico,16 Sony Japan has done a remarkable job
in creating an AI companion for the player, who generates real emotional at-
16
SCEE, developed by Team Ico, released March 22, 2002.
Negative Emotions 183

tachment. In Ico the player controls a young boy who needs a girl named Yorda
to progress through the levels. She alone can perform certain tasks that make
progress possible, but she relies on the boy for protection. This creates real code-
pendency in gameplay terms, and through the skilled use of character enhanc-
ing speech and animations, the player comes to really care for Yorda. Time after
time throughout the levels, the player relies on her powers to unlock doors and
is called on many times to save her life and protect her from creatures that are
trying to take her. Most players of the game eventually develop a very strong at-
tachment to the girl.
It is interesting, therefore, to see that Team Ico decided that the way to make
this emotional attachment really shine through was by making the player experi-
ence the flipside of that emotion, namely emotional loss. Without much warn-
ing, the game forcibly cuts the connection between the player and the AI char-
acter, after carefully building it up through many hours of gameplay investment.
Yorda and the player are suddenly physically separated from each other, and a
strong sense of loss overcomes the player.
This is a remarkably brave level design choice, one that could have backfired
easily if not handled well. After making sure that the player has the capacity to
feel strong emotions towards the AI character, the player is then confronted with
a situation that makes the player examine how deep this attachment goes.
From that point on, the player is very motivated to progress through the
game and reunite with Yorda. This enhanced motivation and the final payoff
when they do reunite provide a powerful gameplay experience that could not
have been achieved without the initial use of negative emotions.

Further Notes
This technique is also employed in the game Shadow of the Colossus,17 which
takes place in the same game universe as Ico. In this game the player forms a
bond with their companionable steed named Agro. In fact, the game goes even
further than Ico into mining negative emotions by expecting the player character
to slay wondrous and titanic creatures that have done it no harm.

Example 9.2:Temporary Removal of Control

Summary
Players can become very motivated and involved in a level design scenario if they
are faced with a situation that is fundamentally upsetting. One such scenario can
17
Also made by Team ICO, published by SCEA, released October 18, 2005.
184 Emotional Feedback Systems

be the removal of a certain amount of control from the player, which creates a
real sense of urgency and discomfort. If used well this can fuel an eventual posi-
tive resolution.

Game Genre
This technique is suitable for any single-player or cooperative game in any genre
that allows the level designer to temporarily, physically trap the player inside an
area.

Goals to Achieve
• Instill tension by creating feelings of panic and trepidation.
• Eventually reward the player by allowing them to remove those feeling.
• Let the player be the agent of this removal of frustration.

Description
(Example type: Existing game)
This example’s inspiration comes from a level called “The Dismal Oubliette”
in Id Software’s Quake.18 It shows a very simple way to effectively use negative
emotions to engineer a positive gameplay experience. The principle applies to
many diverse types of games. In this example, we will show how by temporarily
restricting the player’s movement, we can simply and very effectively create a
tense gameplay moment. A minimum amount of scripting is necessary in order
to make the example possible.
In the level “The Dismal Oubliette,” the player enters a round room with a
central pillar. After the player has entered the room the door behind the player
locks, trapping the player in the room. The floor starts to descent and it becomes
clear that the room is a giant, slow-moving lift. A sense of panic and trepidation
comes into effect as the player is trapped and has no choice over where the eleva-
tor takes him or her.
While the elevator is descending, monsters start teleporting in and the play-
er has no choice but to dispatch them one by one. It is an unavoidable combat
situation over which the player has limited control for the duration of the lift’s
descent, and fear sets in as the player takes damage, and expends resources, with-
out knowing the full duration or the final destination of the elevator trip.
The environment and the available resources are such, however, that most
players should be in less danger than they think, and should be able to dispatch
of all the enemy creatures without perishing themselves.
18
Published by id Software, Inc., developed by id Software, Inc., released 1996.
Negative Emotions 185

When the elevator finally reaches the ground floor the player will feel re-
warded in a number of ways:
• relief because the creatures have stopped attacking,
• pride at their accomplishment,
• relief because they are free to move again.
To some degree, this kind of setup has become a bit of a cliché, but if imple-
mented in a novel or interesting way, it still provides a very effective setup.

Further Notes
The technique can easily be applied to almost any kind of game environment as
long as it allows the player to be limited in some way or another.

Example 9.3: Sound-Induced Shock and

Subsequent Tension
Summary
Extreme sound can completely overwhelm a person’s thinking. If loud and abra-
sive enough, it will disrupt normal cognitive processes and force the attention of
the player since it can’t be ignored. This kind of negative reaction to sound can
be used in positive ways in level design.

Game Genre
This technique is suitable for most games that are looking to scare or disturb the
player through ambient sound.

Goals to Achieve
• Create an unforgettable moment in a level.
• Engineer a lasting feeling of dread.
• Strengthen the sense of immersion.

Description
(Example type: Existing game)
Dead Space19 is a survival horror game that makes heavy use of sound to deepen
the dark and horrific atmosphere of the game. The game’s sound design is an in-
19
Published by Electronic Arts, Inc., developed by Electronic Arts Redwood Shores Stu-
dio, released October 24, 2008.
186 Emotional Feedback Systems

tegral part of the level design and is used in a multifaceted and very effective way
throughout each level. There is a standout moment, however, when the player
starts playing one of the earlier chapters. Right at the beginning an incredibly
abrasive, deeply unsettling and disturbing sound engulfs the player. It seems to
be part of the environment, although there is no clear source for it.
This is what the game’s Audio Director, Don Veca, has to say about it:
I have lived in the San Francisco Bay Area most of my life, and was in
high-school around the time that the “Bay Area Rapid Transit” (BART)
system was built, which in one section literally tunnels underneath the
San Francisco Bay. I’m not sure why, but for some reason the tracks really
scream down there… but in a very “scary” way—lots of high, screechy
over-tones, big bottom end, and very dynamic. In later years as a sound
designer at EA, I started thinking that this sound would be perfect for
a game, but I didn’t know what game or where. Fast-forward to Dead
Space—the perfect place for it. The sound we actually used in the game
was recorded in the tunnel while standing between two cars (where it says
“Do Not Stand Between Cars”). I didn’t really know where in the game to
put it, but in our early demo there was a very inconspicuous room right
after the first “zero gravity/zero air” moment that seemed to have no pur-
pose. Since the Horror genre is also known for its heavy use of contrast,
audio-wise, it seemed to be the perfect place to use this sound. When
you open the air-vac door from the virtually silent “zero G moment” into
this next room that has air (and therefore sound), you immediately get
this deafening, screechy, scraping, pseudo-mechanical ambient roar. The
art team jumped on this, and turned that room into a dark but strobing
visual environment to match the audio. It was pretty cool.20
When this sound eventually stops, the player is left completely rattled and un-
balanced, wondering what happened and feeling a real unease about an environ-
ment that can expose them to such an assault. The sound is never repeated but
its effect stays with the player for a very long time. It is a perfect introduction to
a new level for a horror game like Dead Space.

Further Notes
Although this may seem too “cheap” 21 a way of producing results, it is actually
quite hard to do this well. Get the sound itself wrong and the player may mis-

20
Jayson Napolitano interviewing Don Veca, the audio director for Dead Space, on the
Original Sound Version blog, http://www.originalsoundversion.com/?p=693, October
7, 2008.
21
As in “lacking in class” or “exploitative.”
Negative Emotions 187

interpret its occurrence. Misjudge the intensity and the player may get irritated.
Mess up the timing and it may occur at an inappropriate time.
If done correctly however the level quality as a whole will go up substan-
tially.
Game Environments III

R eality is that which, when you stop believing in it, doesn’t go away.

—Philip K. Dick
“We experience the world through our senses.” To most people this is hardly a
controversial concept. We define our reality through sight, sound, smell, taste,
“Labyrinth 1” from the Nordisk familjebok, Wikipedia, http://commons.wikimedia.
1

org/wiki/File:Labyrinth_1_(from_Nordisk_familjebok).svg, 2009.
189
190 Game Environments

and touch. When it comes to video games, however, several of our senses are tak-
en out of the equation. There is no smell associated with a game environment.2
Likewise, we cannot taste our game environments, which is something that may
well be for the best. We do have a very limited sense of touch, depending on
the platform and its input/output mechanism, rumble, force feedback steer-
ing wheel, analogue pressure-sensitive buttons, motion sensing, etc. Diverse as
these are, they still provide a fairly basic and limited experience in comparison
to the final two: sight and sound.3
This section of the book is about how people experience game environ-
ments, and how we can design levels that make full use of the sensory possibili-
ties this experience can offer. Touch is still too underdeveloped a dimension in
game development, or at least level in design, which after further discounting
smell and taste leaves us with only two chapters based on visuals and sound. What
rich and important subjects they are, however! Some of the most enjoyable con-
tributions to level design language are made within the context of these topics,
at least insofar as they are important in helping to realize game environments,
which is the main focus of this section of the book anyway.

2
At least not until games are developed using smellovision.
3
Or sound and vision, to quote David Bowie.
Visual Experiences
10

M ost established art forms have had a profound impact upon each other. A
healthy cross-pollination occurs, and the result is that art does not exist
in a vacuum, but keeps evolving through exposure to the rest of the world.
Since making video games is such a young art form, there is still much to be
learned from other art forms. This is true, for example, of some elements of sto-
rytelling or sound design (areas that are covered elsewhere in this book), but it is
especially true for visual design and visual direction. Although not exclusively so,
most level design is highly reliant on conveying visual information. Yet surpris-
ingly, not much is known in level design circles about the underlying processes
at work. There are a number of basic areas that need to be examined to help us
use visual techniques to their fullest potential. This is true for any creative me-
dium that is reliant on visual communication with the audience.

Concept

The Physiology of Sight

I have to admit that before I wrote this book I only had some vague notions on
how eyesight works. I already knew that it was important somehow to processes
that one would not immediately associate with eyesight, but how and why was
a mystery to me. I am happy to say that in my limited research I did indeed find
out some very interesting facts about how eyesight works, and crucially that it
has great bearing on the subject matter of this book. Let me share some of the
findings with you.
191
192 Game Environments

The Human Eye

One of the easiest ways to explain sight is to think of the human eye as a biologi-
cal camera. Light falls through the cornea and a hole called the pupil (diaphragm),
is focused by a lens, and is projected onto an area at the back of the eye called the
retina (film). The retina then takes the information it collects and sends it to the
optic nerve, which develops the image into electric signals sent to the brain. The
brain then constructs a mental image from those electric signals.

The Act of Seeing

Seeing is of course so much more than just sending electric signals to the brain.
There are some very important processes at work that have massive bearing on
how we experience the world visually. It is the interpretation of data that pro-
duces some special characteristics. For example, human eyesight (and that of
many other species) seems to be particularly good at processing visual data as
patterns and motion:

Pattern recognition (and recognizing pattern breaks)

Our brains have evolved in such a way that they have become very adept at pro-
cessing data in ways that are important in our daily lives. Of these skills, pattern
recognition stands out particularly well, since we use it so much. How else do
we differentiate between all the differing visual inputs we receive via our eyes?
Pattern recognition allows us to process visual data at a much-enhanced speed
because it lets us carve up the visual input into useful chunks to which we ascribe
further meaning and behavior. This process is actually called chunking1 and is key
to our intelligent understanding of our environment. Chunking occurs when we
encounter large amounts of visual data and summarize it into abstract patterns
or visual models we can deal with much quicker. For example, a master chess
player can mentally recreate the location of all the pieces on a chessboard by rec-
ognizing the underlying strategic patterns involved.2 When we lose something
in the grass, we don’t have to process each blade of grass visually in order to find
the lost object. Instead, we have a pattern (the grass) and we look to see where
the lost object breaks it.
Although we don’t often think about this aspect of our cognition and intel-
ligence, it is actually a key part of our ability to process information in the world.

1
I kid you not!
2
Tellingly, this skill falls away if the pieces are placed randomly.
Visual Experiences 193

Interestingly within the area of robotics and other related fields like artificial in-
telligence, it has become clear that it is incredibly difficult to artificially recreate
this ability. Some even see it as a necessary hurdle that has to be jumped on the
path to full artificial intelligence. An enormous amount of research is done in
this field as a consequence; so much that it has spawned dedicated journals3 and
academic gatherings.4 It is clearly a fascinating and important field and it leaves
no doubt that pattern recognition is fundamental to our ability to create abstract
recreations of the world around us. And that is something that marks us even
further as strangely evolved creatures.

Motion tracking
On a similar note, we have the ability to pick out, track, and process the move-
ment of countless objects in our vision. We know about the amazing skill of a
hawk to pick out the movement of a tiny mouse on the ground, or a cat focusing
on the movement of a wriggling insect. But in our own lives we can recognize
similar feats of motion perception just as easily. Take a moment to think of the
real complexity of a person crossing a busy street in a major city. That person has
to successfully track and judge the motion of dozens of moving bodies and make
individual assessments on how to react to them in many cases. It is not rare for
such a mundane act to require the person to process data dealing with several
other pedestrians, many cars moving at different speeds, the person’s own mo-
tion, sudden changes due to traffic light fluctuation, and so forth.

Op-art
One of the best-known art forms that deals with visual techniques is op art. Op
art deals with optical illusions elevated to the level of art. Often, through the use
of patterns, scale, or color, op art can create tension in impossible images. (See,
for example, Figure 10.1.) It examines the psychological reaction to physically
impossible or confusing images or other art works. Although not the most self
evident of techniques, when studied, it actually shows a wealth of understanding
of how human beings deal with visual input, and how easy it is to subvert these
reactions.

3
Pattern Recognition: The Journal of the Pattern Recognition Society, http://www.elsevier.
com/wps/find/journaldescription.cws_home/328/description#description.
4
International Conference on Artificial Intelligence and Pattern Recognition, http://
www.promoteresearch.org/2007/aipr/.
194 Game Environments

Figure 10.1. Penrose triangle.5

The Psychology of Sight

Beyond the physical, there is a further, even more interesting dimension. What
is it we actually do with all this visual information on a psychological and mental
level? How does sight affect how we feel?

Emotional Associations
Aside from the way human beings react to visuals on a mechanical level, there are
factors at play that are more nebulous.6 Certain visuals evoke an emotional reac-
tion within us, sometimes completely independent from physiological causes.
We are scared of the dark, clutter makes us nervous, towering architecture may
make us feel insignificant, and so forth. Why is it that we may label one visual
experience as beautiful and another as ugly or depressing? There isn’t one all-
encompassing answer to all these questions. There are many factors at play, and
some of them cannot easily be explained. Let’s look at a few of them.

Aesthetic sensibilities
Aesthetics is a rather large and complex discipline, sometime even a contradic-
tory one. It studies the senses’ reactions to things like art, but I won’t spend time
trying to explain the full scope, or even attempt to give my own understanding
of what it means. It suffices to say that most people when asked about aesthetics

5
The original drawing of this object appeared in Lionel Penrose and Roger Penrose, “Im-
possible Objects: A Special Type of Visual Illusion,” British Journal of Psychology, 49:1
(1958), 31–33.
6
But important nonetheless.
Visual Experiences 195

think of the study of that which makes things beautiful or enjoyable to the senses
(or ugly and repugnant).
We can learn many things from aesthetic principles in regard to making a
visual scene pleasing to look at, but the deeper lesson lies in the fact that aesthet-
ics gives us a measure of control, a tool, for influencing the audience’s mindset.
Instead of using aesthetic principles to create a pleasing scene, the knowledge
can be used to create a disturbing scene, for example by sabotaging an earlier
carefully established sense of aesthetic balance.

Primal reactions
Some emotional responses to visual scenes or stimuli seem to be hard-coded in
our brains (or taught at such a young age that they are involuntary). Who doesn’t
feel a slight tremor in their legs when standing on the edge of a precipice? A large
percentage of the population freaks out when confronted with insects crawling
on them, even if they are known to be harmless. We react well to smiles, not so well
to a stranger or strange animal running straight at us.

“Taught” reactions
At other times we are confronted to what amounts to long established visual
conventions. In nature an animal is taught by experience which colors to avoid
when gathering food. In human culture we have many conventions as well. In
a movie, a shaky point-of-view handheld camera may indicate a stalker. Red is
often the color of danger. Different creative fields have different sets of conven-
tions, but many visual ones are shared across disciplines. This is ripe picking for
anyone who wants to enlarge his or her vocabulary of visual techniques. Many
books have been written on the subject, so there is no excuse for ignorance in
that regard. Even a trip to the museum can do wonders.

Visual Direction
When we take all of these examples and the associated knowledge we gleam
from them, we suddenly find that we have at our disposal a huge toolbox for
artistic expression. The techniques available to us are often surprisingly effective
and versatile, and it should come as no surprise that they have many well known
and practical applications. When these principles are being applied to aid artistic
or creative expression7 we speak of visual direction.

7
They have other applications as well, but I don’t want to stray away too far from the core
of this book.
196 Game Environments

Few fields are as knowledgeable about the psychology (and physiology)

of visual direction as filmmaking. The field has been discovering and refining
techniques of visual direction for over a hundred years. There are thousands of
fantastic examples of directors of photography (DPs) or cameramen who have
been adding their own twist to the subject. A beautiful example can be found in
a scene in the movie Lawrence of Arabia that showcases one of the most famous
sequences in cinema history. It describes the first meeting between Sherif Ali,
played by Omar Sharif, and Lawrence of Arabia, played by Peter O’Toole).
Lawrence and his guide Tafa have just quenched their thirst from a well in
the middle of the desert. The camera shows Lawrence lazing about and the guide
getting more water from the well. The whole horizon can be seen for many sec-
onds. Finally it becomes clear that somebody is appearing on the horizon (Omar
Sharif ’s character) and approaching. It becomes clear that it is somebody riding
towards them on a camel. Suddenly Lawrence’s guide runs off to get a pistol,
grabs it, and takes aim. He is shot by the approaching rider. The following dia-
logue takes place:
Lawrence
Who is he? Tafas!
Ali
He’s dead.
Lawrence
Yes. Why?
Ali
This is my well.
Lawrence
I have drunk from it.
Ali
You are welcome.

It is a well-written scene with dialogue that is original and interesting. Had it not
been preceded by the approach on the horizon by the rider who appears as in a
mirage, it would not have been as famous as it is now. On location filmmaker Da-
vid Lean and the director of photography (DP) Freddie Young, realized they had
a particular problem with the focal point of the viewer. Omar Sharif was supposed
to ride into the scene appearing far away on the horizon, slowly becoming visible
to the audience. Unfortunately, with such a clear and wide horizon and the other-
wise empty desert, the viewer’s eye started to wander. Since the viewer wasn’t di-
rected where to focus he or she would start scanning the screen, missing the point
where Sherif Ali was to slowly materialize on the horizon. This meant that Omar
Visual Experiences 197

Sharif ’s character would not get the introduction that the scene demanded and
the whole sequence would fall completely short of the impact that was required.
Freddie Young’s solution was ingenious. He painted the desert. He literally
had his crew paint a line of desert sand reaching all the way to the spot on the
horizon where the Sherif Ali character was to appear. The line was slightly lighter
than the normal desert sand, just enough so the eye and brain would pick up on
this and follow it to the beginning, the intended focal point. It worked perfectly;
the viewer’s eye is now naturally drawn towards the trail in the sand, and is fo-
cused in the right area of the screen even before Sherif rides into view. Without
realizing it, the audience now is looking exactly where the DP wants it to look. It
makes the scene perfect, and has become a classic example of visual direction in
cinema. Most people who watch never realize that their eyes are being directed.

Visual Direction and Film Language

This description of the famous scene from Lawrence of Arabia is highly signifi-
cant, as it shows a practical and applicable example of visual direction in a cre-
ative art form and can teach us many things. It shows that it is possible to direct
the emotional impact of a scene, without being too overt, with simple visual di-
rection. This is extremely important if we want the player or movie viewer to be
immersed and willing to suspend disbelief. It is just one example, but it hints at a
whole toolset or language of visual direction that may be of use to level designers
as well. In filmmaking terms, this kind of visual direction is part of a larger body
of expressive and artistic means often referred to as film language or film art.
Film language constitutes an ever-evolving and vital body of knowledge8
that level designers would do well to study and incorporate in their own level
design language.

Related Disciplines
If we take a high concept approach to the topics we are discussing in this chapter
we start to see that there is much overlap with other fields of interest, some of
which are associated with a huge body of work.

Architecture
Architecture is often named as a logical subject of study for level designers, and
to a degree this makes sense. Architecture has much to do with designing spaces

8
Bordwell and Thompson’s work is a good starting point: http://www.davidbordwell.
net/blog/.
198 Game Environments

for human beings, and since architecture has been around for a few thousand
years, this gives us a wealth of information9 that could apply to level design. We
can look at aesthetics, construction and technology, visual direction, social ma-
nipulation (in a benign way) and so on.

Planology and urban planning

Some of these matters are also related to choices made on a societal level. How
do we design and plan public spaces to be effectively used by millions of people?
How do we reach a consensus that will appeal to the people forced to make use
of those spaces in the future? Answers to these questions can have much bearing
on multiplayer level design issues, especially in the areas of MMOGs and other
virtual worlds. Again, much important research has already been done and can
be of great help to us.

Environmental psychology
There is no shortage of interesting areas to study across the whole spectrum of
applied visual theories. In fact, the way we act in and interact with our environ-
ment within this context touches on too many disciplines to name. There is,
however, an area of study that tries to unify many of these loosely related topics
into a general area—Environmental Psychology. It is a field that is not clearly de-
lineated but has much to offer to level design, and in my personal opinion, level
design actually has much to offer to this field in its own right.

Concept Summary
All of the above examples, be they physiological or psychological in nature,
provide us with creative opportunities. The knowledge we have of how human
beings process visual information and how they react to visual stimuli on an
emotional level allows us to create situations where premeditated reactions can
occur.10 In other words, we can influence the reactions of people around us (only
with their consent of course) by manipulating what they are exposed to visu-
ally. Creative expression is filled with examples of artists doing this successfully.
Sometimes it has even been elevated to an art genre in its own right, as we can
see in the op art movement.
We can even go as far as saying that visual direction is a key part of the lan-
guage of many art forms, film being a notable example. Whatever the terminol-
9
A good starting point for study is Francis Ching’s book Architecture: Form, Space and
Order, Third Edition, Wiley, New York, 2007.
10
That sounds much more sinister than intended.
Visual Experiences 199

ogy used, however, these forms are only relevant if we take lessons from them
that we can use in actual level design theory and practice. This is something we
will explore in more detail in the next part of this chapter.

Theory
The previous part of this chapter goes into some detail explaining the core prin-
ciples of visual processes. It covers underlying physiology, general comments on
sight and psychological impact, and to a degree, artistic viability. The first and
the latter are mostly straightforward examinations, but the middle one dealing
with psychological principles deserves extra scrutiny within the context of level
design theory. In what cases can we use principles of visual direction established
in other disciplines? When do we need to make alterations to them? When do
we need to define our own principles and techniques? In other words: we need to
establish visual direction techniques and principles as part of level design theory.
We will look at this and other aspects of visual direction in the next section.

Camera
We cannot talk about as visual a medium as video games without spending some
time on the subject of the camera. The camera system of a game is the medium
through which we view the game directly. It provides our eyes with the visual
data that the game wants to player to receive. For that reason it is of the utmost
importance that the subject of game camera be taken seriously. There is a good
chance that on any given project the level designer will have some control over
the game camera. This makes it extremely important that the designer have at
least some knowledge of the capabilities of the game camera.
The introduction to Sidney Lumet’s chapter on cameras in his book Making
Movies reads as follows:
The Camera
Your best friend
First of all, the camera can’t talk back. It can’t ask stupid questions. It
can’t ask penetrating questions that make you realize you have been
wrong all along. Hey, it’s a camera!
But:
• It can make up for a deficient performance.
• It can make a good performance better.
200 Game Environments

• It can create mood.

• It can create ugliness.
• It can create beauty.
• It can provide excitement.
• It can capture the essence of the moment.
• It can stop time.
• It can change space.
• It can define a character.
• It can provide exposition.
• It can make a joke.
• It can make a miracle.
• It can tell a story.
If my movie has two stars in it, I always know it really has three. The
third star is the camera.11
This is about as ringing an endorsement of the value of good camera work as can
be found.
Without disagreeing with Mr. Lumet, it is nonetheless important to note
the difference between game camera and film camera. Although many princi-
ples overlap, the two aren’t the same. A major difference between film and video
games lies in the fact that while a film audience has no influence whatsoever on
what is displayed onscreen, games allow for a huge range of freedom in this re-
gard. Not only in respect to what scenes are shown, but also in what order they
occur, and in some cases games can even allow the user to change the actual
environment. This means that in most cases the traditional controlled version of
visual direction goes out the window. Other techniques have to be employed.
Let’s first look at some typical game camera situations so we can judge po-
tential limitations. These are of some importance because the choice of game
cameras determines how the player sees the game.

First Person
We speak of first person camera when the camera view is shown from the per-
spective of the player. Imagine the scene shown as if seen through the player’s
eyes. There are many variations possible, but the following are the most preva-
lent. (They are not mutually exclusive.)
On rails
In this variation the camera moves on a predetermined path and the player has
little or no control over the direction it looks at. (Although the player may have
11
Sidney Lumet, Making Movies, Random House, Inc., New York, 1996, pp. 75–76.
Visual Experiences 201

control over the direction it moves in. A good example can be found in Myst,12 a
puzzle-based adventure game, or in a similar vein. The Seventh Guest 13”is a game
with similar play mechanics.
By today’s standards, this type of camera control would seem overly limited
and even old fashioned. From a purely level-design-centric view, however, the
advantages are massive. The designer rather than the player controls the camera,
which gives him or her much more license to direct the gameplay.
A further advantage lies in the fact that designers and artists can build envi-
ronments based on predetermined camera positions, only having to build those
parts that the player can actually see. This is just as on a film set, where a street
only needs to consist of the fronts of buildings, since nobody will ever see the
inside or the back.

Player controlled
This is by far the most prevalent form of first person camera in games. Imagine a
camera glued between the eyes of the player, and you’ll get a pretty accurate pic-
ture. Camera direction and movement are controlled by the player, which means
that the designer will have to use all kinds of techniques of visual direction in or-
der for the player to follow his or her lead, receive specific in-game information,
get the right emotional messages, and so on.
The disadvantage of not having as much control over the player’s experience
can be seen as a major advantage, as well. Through the use of subtle visual direc-
tion, the designer can shape a play experience without leading the player by the
nose. This helps make the player responsible for his or her own actions, greatly
reducing resentment for being presented with lack of choice.
Much more detailed environments have to be built to accommodate all the
player choices, but on the other hand, this delivers increased chances for explo-
ration.

Third Person
An easy way to imagine third person camera views is to picture a floating camera,
disassociated from the player’s in-game presence. In most cases it follows the
player character, but this can be done in quite diverse other ways as well. The
most typical ones are discussed below.

12
Published by Brøderbund Software, Inc., developed by Cyan Worlds, Inc., released
September 24, 1993.
13
Published by Virgin Interactive Entertainment, Inc., developed by Trilobyte, Inc., re-
leased 1993.
202 Game Environments

Side-on
Classic two-dimensional platform games feature a camera that tracks the player
but is stuck in a side-on, 2D plane. Some minor control by the player may be
possible, for instance by “pushing” the camera slightly off center to reveal more
of the playing field, but most of the time the camera will be stuck in one position
relative to the player character and will not leave the 2D plane of movement.
These limitations produce their own unique problems and opportunities.
It is difficult to apply any traditional dramatic touches or to create a sense of
foreboding due to camera placement or something equally filmic. On the other
hand, it is an excellent camera system for showing the player things explicitly
and safely, away from player influence. A room with a monster might move into
view, forewarning the player of a situation that is yet to occur. Whole scripted
sequences can take place just outside of the player’s reach.

Third person free-cam and follow-cam

One of the most prevalent camera systems in modern video game is used in third
person games with a controllable player character. These include Tomb Raider,14
where the camera can move about in 3D space, but follows the player. This can
be further divided in games that allow player positioning of cameras (free cam),
and games that don’t.

Third person free-cam. This allows the player to rotate the camera, and sometimes
to adjust pitch and yaw. Since the player has almost full control over where the
camera is at any time and may have much control over where the player character
can go, much will be asked of the game environment. It needs to be much more
realized and must stand up to the scrutiny of being explored and looked at from
many angles and directions.

Third person follow-cam. This employs a camera that follows the player around, as
in the 2D side-on game example, but it is not limited to one plane of movement.
Instead, it is generally the level designer who determines where the camera can
go and how it gets there. (However, they may still have some control over pitch
and yaw.)
This allows for a much more controlled style of level design, where visual di-
rection can be planned with more accuracy or knowledge of what the player will
be able to see at any time. This has fairly obvious advantages, which are further
enhanced by the fact that the designer is also able to employ techniques dealing

14
Published by Eidos Interactive, Inc., developed by Core Design Ltd., released 1996.
Visual Experiences 203

with camera height and framing in general. In many ways, this camera system
presents the best compromise between player needs and directorial control.
Games like Zelda: Twilight Princess have employed this with much success.

Placed (static)
In some cases the camera can be placed in 3D space, but it doesn’t move around
relative to the player position. (However, it may well keep the player centered
onscreen, within the limits of the camera’s pivot distance.)
This is generally done to add drama and tension to an environment, as in
survival horror games, or to aid specific environment-specific gameplay, like
jumping sequences that may be hard to judge with a moving camera.
Whatever the reason for this system, whether temporary or used through-
out the game, it features some unique problems and opportunities.

Other Types
These examples all spring from game types that feature an in-game presentation
of the player. There are of course many game types where this is not the case;
chess, for example, is generally depicted through a camera that just shows the
whole playing field. In those cases many of the third-person circumstances still
apply; the camera is just not associated with a specific player character.

Camera as a Level Design Tool

Depending on the project and the people involved, chances are that the issue
of camera use will not be taken seriously enough. People may argue that they
are making a game and not a movie. They are missing the point. What matters is
that, just as in filmmaking, the camera has an enormous influence over the final
quality of the game. Ideally, if appropriate, this means that the level designer is
given as much control over the camera as possible. Therefore, it is almost al-
ways of the utmost importance that camera is given serious thought early on in
the project, and that there be sufficient code support and cooperation. I cannot
stress the point enough. Although this is a crucial component of level design
and game quality, it is rarely understood by other team members. The designer
nonetheless should argue this point as strongly as possible. The level design can
be strengthened immeasurably by giving the designer a set of tools that allow for
camera movement, lens changes, field of view changes, shot transitions, camera
types, etc.—whatever is deemed appropriate for the game.
This responsibility must also be recognized by designers themselves, who
must study and build a repertoire of video game camera techniques and limita-
204 Game Environments

tions. Books like this one can hopefully be of great use, but it is also extremely
useful to be up to date on real world camera techniques, especially motion pic-
ture camera principles. This includes subjects like framing a shot, set design,
camera moves, and lenses. A good book on the subject is The Camera Assistant.
A Complete Professional Handbook, by Douglas C. Hart, which is one of the key
reference works within the discipline.

Carrying Information
A fundamental aspect of camera work, and a key part of level design, lies in the
realization that every scene or shot contains information. This is pretty much
always true, even if the information isn’t always very interesting. Once we get
beyond the almost banal simplicity of this statement, we are nonetheless pre-
sented with a very important question: what information do we want to convey to
the player?
This is actually a very interesting question, with a limitless number of po-
tential answers. Every single visual decision made in the course of the design will
make an impact, and that includes “simple” decisions like item placement. It is
an old level-design cliché, for example that a room full of powerups signals an
imminent boss fight.
This principle can be extended to include other areas as well. Scripted se-
quences can carry information useful to the player. The original Half Life,15 for
example, would signal danger to the player by letting Very Bad Things happen
to NPCs in view of the player, often in scenarios that would also give the player
information on how to avoid a similar fate. In a game with a day/night cycle,
dusk may tell the player that it is time to find shelter from night crawlers. The way
somebody is dressed in MMORPGs may indicate rank and experience. The list
is endless. We need to be as aware as possible of the impact of the things that the
player can observe, and try to incorporate this knowledge in the visual direction
and the visual design of the levels.

Visual Design
There is no doubt that at times, as level designers, we need to be knowledgeable
of the visual arts, so we can create real impact on a visual level as well as on a
gameplay level. Furthermore, quite often the two should not be seen as separate
disciplines at all.

15
Developed by Valve Software.
Visual Experiences 205

Darkness and Light

Photography is often described as painting with darkness and light. Although col-
or can clearly be a part of the equation, the description still holds true in general
if we accept that color still comes from the consequence of lighting a scene. In
most cases it is the use of darkness and light that bring out all the distinctive or
interesting aspects within an image. This principle can be applied to games as
well, so we should at least look at some basic aspects of the use of darkness and
light in a scene, and at the interplay between the two.

Use of darkness
Traditionally, darkness is used in games to create fear and unease. Although eas-
ily turned into cliché, this is not a bad application per se. However, we must to be
careful not to create a lazy darkness = scary attitude. At least not without under-
standing why it can be scary. The key aspect of darkness is that it takes away sight.
This is an obvious observation, but it becomes more interesting if we look at
what that can mean in practical terms. For example, if the point of darkness in a
level is to create fear, taking away sight can work, because the player loses a sense
of control. If we can’t see dangers anymore, but they are still around, we rightly
become rather nervous. Dangers can also be hinted at. An imagined creature ac-
companied by sounds, for example, is scarier than a visible one.
However, darkness can be used in all kinds of other ways. If the gameplay
is about staying undetected, darkness can be a real friend. Darkness can provide
cover for exploration, or provide the play mechanic for a puzzle.

Use of light
In contrast, light is often used to create a sense of safety, a positive to counter the
negative of darkness. Again, this is not necessarily a bad application, but there
are many other ways that light can be used to create interesting gameplay and
therefore perform more functions than just lighting the scene. Good lighting is
as much the provenance of a good level designer as that of a game artist. A level
designer may decide to use the lighting in a level to create gameplay sequences
that simply cannot be done in another way. Take for example typical stealth
gameplay where the player needs to remain undetected, and therefore out of the
light. There is great fun to be had by giving the player the chance to control light-
ing that affects the game AI, by being able to switch lights on or off, or perhaps
by allowing the player to blind AI opponents with a searchlight.
There are plenty of possibilities where lighting can be used in an original
way that enhances gameplay. Some more will be highlighted in the practice sec-
tion of this subject.
206 Game Environments

Patterns and Motion Tracking

Since video gaming is such a visual medium, it is a good thing that so much of lev-
el design is about teaching mechanics, as we have seen in some of the earlier chap-
ters. It is a good thing because many powerful techniques make use of patterns
and pattern recognition. It is thankfully very easy to create level design scenarios
where these innate skills are used to their fullest and tested through gameplay.
Take for example classic platforming gameplay, for instance the ubiquitous sec-
tion where the player has to navigate underneath a set of pistons smashing down
on the player’s route. This is almost always only possible to do by observing the
pattern in which the pistons smash down and basing a timed run on that pattern.
Pattern recognition isn’t only applicable to movement patterns. Sometimes
it is just as clear in a static visual setting. In a shop with hundreds of multicolored
vases, one row of ten blue vases would stand out very clearly. Or in actual game
terms, the player can easily recognize the difference between environmental
props and (for instance) powerups. That makes them very suitable to help direct
the player visually or to suggest possible accessibility to areas in the level that
the player hadn’t considered. If there is a powerup, the player can surely reach
it— right?
Level design is absolutely full of potential uses for our innate ability to rec-
ognize patterns and motion, and we should make full use of this principle.

Scale
Scale can have a huge impact on the ultimate feel of a level’s design, and many
scale-related decisions have very specific results. Some situations are easily un-
derstood and can be implemented without too much fuss. A player may feel a
sense of elation after having conquered a long and difficult climb up a huge tow-
er. The scale of the tower when the player first looked up at its base, and the re-
versed situation when the player has reached the peak and is now looking down
at the tiny objects back at the ground, is bound to have an impact. This can be
further enhanced by giving the player some useful visual information related to
being at such a high vantage point, or awarding the player with some rare object
for reaching this area.
Negative feelings can be stirred as well. Imagine the horror of seeing the
scale of a growing mushroom cloud after a nuclear explosion. Populating a level
with big looming buildings that tower over the player can generate feelings of
oppression. A further sense of claustrophobia can occur by making the player
navigate small, cramped corridors. Whichever way you look, matters of scale be-
come relevant to level design issues in diverse ways. If the designer decides to
Visual Experiences 207

make a set piece out of scaling a very high mountain, space must be reserved in
the level to accommodate the scale of the mountain. Furthermore, is it actually
fun to traverse the real distance this would entail? Perhaps some shortcuts are in
order, possibly an action-filled cable car ride, and a further hitch on a lift built in
the mountain.
Clearly, seemingly small16 decisions on scale can have a huge impact on the
content and feel of a level.

Visual Style
Some of the art decisions that need to be made for the game are about choosing
an appropriate style. This is a decision that influences many things beyond the
actual look of the game. Real gameplay issues come into the picture and must
therefore be looked at before any art style is chosen.
We can initially make a rough decision between a style that is based on re-
alistic imagery (realistic in depiction, though not necessarily in content) and a
direction that is much more stylized, like the use of toon-shading or super-de-
formed proportions.

Realism
The advantage of a realistic style is that we can use the real world as an inspira-
tion and a resource. It is very handy that if we want to use a car in a realistic
game, we can just go outside and take the dimensions of a real car, or get some
blueprints from the manufacturer. Consistency and believability can be guar-
anteed by sticking with real world parameters, an advantage that shouldn’t be
underestimated.
Trying to achieve realism has its own problems, however. There are a num-
ber of dangers, some of them unexpected. First, there is the problem of detail.
In a realistic setting there is just more work to be done to ensure that realistic
representations are supported by sufficient levels of detail. There will be certain
expectations that must be met in order for the environment to be convincing.
A door on hinges will have to swing, a car will have to behave in a recognizable
way, and so on. To further complicate matters, the game’s AI needs to be realistic
enough, especially the AI of human characters. It is very easy to underestimate
the damage that can be done by not preparing for this problem. Level designers
can easily find themselves in a place known as the uncanny valley.
The theory of the uncanny valley was made popular by the Japanese
roboticist Masahiro Mori. In his work dealing with the creation of realistic
16
No pun intended.
208 Game Environments

uncanny valley
moving
still healthy
bunraku puppet person
humanoid robot

stuffed animal
familiarity

industrial robot

human likeness 50% 100%

corpse prosthetic hand

zombie
17
Figure 10.2. “The uncanny valley.”

human-looking robots and people’s emotional responses to them, Mori has

formed an interesting theory. He postulates that people react well to robots that
greatly resemble humans, but only to a certain point. When the robot’s look gets
very close to lifelike, the small differences between real humans and robots start
to take on an eerie quality. Imagine for instance a robot that looks lifelike except
for eyes that are too wide apart and blink too often, and that moves around in a
shambling gait.18 It doesn’t take much imagination to picture some very disturb-
ing, but close to human, behavior. Figure 10.2 shows the principle clearly.
It is very rare that level designers need to design the look of human AI char-
acters in the game, but what we do often need to do is to design the way they
need to behave. And this is where some real problems can occur. Scripting AI
behavior in a realistic game offers some well-known dangers. Lines of speech

17
“Uncanny Valley,” Wikipedia, http://en.wikipedia.org/wiki/Uncanny_valley, 2009.
18
I am creeping myself out here!
Visual Experiences 209

can be inappropriate, the AI can be too oblivious to its surroundings, its path
through the environment can be strange, idle animations can play at the wrong
times, and so forth. Many of these can be very detrimental to immersion, or even
worse, open to exploits. In other words: tread carefully!
Another problem within a realistic style is the lack of creative license if we
opt for realism based on our natural world. That is to say that the game is natural-
istic by only including elements that can exist now or that could have existed in
the past in the real world. Having to be realistic in this manner quite often means
that for an element to be believable, its use must be explained. In a cartoon, sci-fi,
or fantasy setting it is quite easy to create an impenetrable barrier that the player
can see through, but that will not allow anything else to pass. A magic fence or a
nice force field would do the trick, with few questions asked. Internal game logic
still needs to be established and followed, but more creative license can be used
to achieve these things.

Stylized expression
The extra creative license just described is not the only advantage in a stylized
setting. In some non-realistic styles, less detail is needed to create a clear im-
age. Cartoons can create very clear images by using very few brush strokes, and
similarly, a game that uses toon shading needs much less detail to paint a clear
picture for the player. To the level designer this can be a huge advantage. It makes
it much faster to create basic functioning game environments, and much easier
to change them around when needed. This in turn has a subsequent effect on
flexibility and iterative prototyping. These are all massive bonuses.
As always, though, there are downsides. The style may be so alien that no
real-world reference can be used. Alien technology and road signs may make it
hard to direct the player or to make sure the player cannot get lost. There may be
the need to portray fairly complex and detailed mechanisms within a style that
is too bulky to easily facilitate this. As always, we have to be very careful not to
fall into any of these traps. Luckily, there are many techniques available to us that
can help us navigate these issues.

Power of Suggestion
To make sense of the world, our brains are able to take limited information and
extrapolate a more detailed picture from it. They literally fill in the blanks. We
don’t need to see every individual brick in a wall to know that the whole wall is
there. If something comes running towards us, we can judge pretty quickly if it
is on a collision course with us, and roughly how long before it is time to get out
of the way.
210 Game Environments

We can do these things for a number of reasons. Most of them are neurologi-
cal in nature, reinforced by previous experiences. When taken together, these
factors give us a very decent predictive mental view of the world. However, we
may still be fooled, at times in surprisingly easy ways.

Let the Player Fill in the Blanks

I have already given the example of a movie set with streets, where only the
fronts of buildings need to be seen. As long as the camera doesn’t enter a build-
ing, there is no need to show, and therefore build, an interior. This principle ap-
plies just as strongly to level design. The example given is a very direct and literal
example. A gaming example would be a house with a locked door that the player
cannot enter. Put some curtains on the window and we can forget about the
interior. We can do better than this however; more interesting and versatile ex-
amples can be found.
The principle of visual suggestion doesn’t just apply to things that we can
observe. As we already touched upon in the chapter on immersion, sometimes
the presence or absence of something in the environment suggests deeper levels
of information.
• Train tracks suggest a train.
• Smoke suggests fire.
• Hot tea suggests a recently present person.
• Political graffiti suggests factions.
• A radar post suggests planes.

These are just a few.

Much better or more appropriate examples are easily found when we ex-
amine individual projects. What we wish to suggest depends completely on the
game and on the situation. It can be the presence of a certain kind of creature,
or perhaps shelter from environmental hazards. It doesn’t really matter. What
does matter is that we have a cheap and convincing method at our disposal that
can convey information about the game environment in many useful ways, of-
ten indirectly. A good understanding of this principle is a major addition to the
arsenal of visual direction. Sometimes this can take the form of misdirection as
well.
“Filling in the blanks” like this is sometimes referred to as closure. This
has much wider application potential than is suggested here, but to keep
matters organized, I want to revisit these concepts in more detail later in the
book.
Visual Experiences 211

Misdirection
Have you ever seen a magician at work? One of the main techniques of any magi-
cian is the use of misdirection. The famous magician Jean Hugard once said the
following about misdirection:
The principle of misdirection plays such an important role in magic that
one might say that Magic is misdirection and misdirection is Magic.
This is a quote that level designers need to take to heart. After all, we deal ex-
clusively in virtual, and therefore in many respects illusory, worlds. This is not
unlike how a magician deals with illusions to create magic. The principle behind
misdirection is simple enough: focus the audience’s mind and expectations in
one direction, while performing an act of some sort (undetected) that suggests
that something magical has happened. (For example, while a street magician is
focusing our attention on a complex maneuver with his right hand, he has al-
ready reshuffled the deck with his left.19) Of course it is unlikely that we will use
card tricks or anything like that within level design, but sometimes we do want
some actions in the game world to distract the player from others. Sometimes
this is done to hide immersion-breaking activities, or sometimes to set up a spe-
cific gameplay surprise or scripted sequence. You will be surprised at how effec-
tive this technique can be when used well.

Theme
What is it all about? I don’t mean what happens (that is plot), but why is it all
happening? What is the underlying reason or the intention? Generally, these
kinds of things can be attributed to theme. Theme is a subject that pops up a few
times in this book, and although it sounds imposing and serious, it isn’t some-
thing we should be scared of. In Stephen King’s book On Writing we find the
following thoughts20:
Writing and literature classes can be annoyingly preoccupied by (and
pretentious about) theme, approaching it as the most sacred of sacred
cows, but (don’t be shocked) it’s really no big deal. If you write a novel,
spend weeks and then months catching it word by word, you owe it
both to the book and then to yourself to lean back (or take a long walk)
when you’ve finished and ask yourself why you bothered–why you

19
I don’t know if that at ever happens, but the principle is clear.
20
Stepheb King, On Writing: A Memoir of the Craft, Pocket Books, New York, 2002, pp.
200–201.
212 Game Environments

spent all that time, why it seemed so important. In other words, what’s
it all about, Alfie?
He also notes that:
. . . Not every book has to be loaded with symbolism, irony or musical
language (they call it prose for a reason y‘know), but it seems to me that
every book—at least every one worth reading—is about something.
This applies as much to writing as it does to level design. And one way to re-
ally express this well in level design is by defining a well thought-through visual
theme, or more likely a set of themes. A good partnership with the game artists is
of great importance in this case. (More on that later.)
Once a clear idea emerges on theme(s), it is easy to start to apply it in several
aspects of the visual design. It is actually important to make sure that theme is re-
flected throughout several areas of the level, or it won’t have the impact it should
have. If a sense of death needs to permeate a ghost town, it is perfectly valid to
sprinkle the environment with graves.
Beyond these kinds of direct visual implementations, there are subtler ways
to incorporate theme in the level design. Much can be achieved via the use of
symbolism and metaphor, even without the player being aware of this.

Whose Job Is It Anyway?

Some people will argue that many of these aspects of visual design fall exclu-
sively within the domain of the art department of the project team. They are
partly right, yet at the same time completely wrong. They are partly right because
much of the relevant expertise will lie in the art department. A good artist will be
able to offer far more, and sometimes better, input on how to achieve these goals
of visual direction than many level designers. On many projects I would even go
as far as to say that the level designer has no chance whatsoever to do a great job
unless the art department is fully involved and on board.
Where they are wrong, however, is when they claim that visual design is the
exclusive domain of the art department. Nothing could be further from the truth.
Level design is applied game design, and the art of the game should support
this first and foremost. Function comes before form. Anyone who suggests that
this sabotages artistic expression is working in the wrong field. Imagine Martin
Scorsese having to direct a gritty scene that calls for harsh lighting but being
unable to do so because the cameraman or DP insists on shooting a tranquil
tableau in soft focus, independent of the dramatic needs of the shot. If this were
to happen, that cameraman or DP would be out of a job faster than you can say
“Print!”
Visual Experiences 213

The same danger exists in an inverted manner if level designers think they
can do the visual direction by themselves. In most cases this would be a laugh-
able conceit.

Theory Summary
The visual direction used in a level touches every other aspect of that level’s de-
sign. It simply cannot be left till last; it is the designer’s responsibility that the
consequences of related choices be well understood and implemented. This
means that the designer must act as a liaison to the art department in order to
coordinate and cooperate, and with programmers in order to receive the right
tools for the job. It also means that the designer must study the craft from all
angles. This includes studying many aspects of film and photography, including
the areas of cinematography and set design. When all these things are done, the
designer has in visual design a tool that can completely transform the experience
of a level. What at the beginning was a set of planned gameplay sequences can
now be full of meaning, beauty, drama, backup for gameplay, and more. There-
fore it deserves to be treated with a certain amount of seriousness.

Practice

Example 10.1: Use of Light—Visual

Direction and Mood Manipulation
Summary
Light can be used to focus attention as well as change the mood of an area. Com-
bining these two concepts can be very effective.

Game Genre
The technique is suitable for most games that require the player to navigate non-
abstract environments.

Goals to Achieve
• Guide the player towards a goal through visual direction.
• Equate light with a positive outcome.
214 Game Environments

• Alter the general mood of the environment.

• Build towards a rewarding scene.

Description
(Example type: Original)
“Climbing towards the light.” This example is going to take a hypothetical
scenario that can be adapted to fit with all kinds of setting and games. The basic
premise is that the player is to slowly work their way from an oppressive dark
area towards safety, represented by light.
The example is set against the background of a partly collapsed skyscraper.
The player is somewhere at the bottom level and can only go upwards. Electricity
is mostly compromised and the starting area is nearly completely dark. A num-
ber of discreet stages or elements can be incorporated.
• Starting off in near dark gives the level a very oppressive and claustro-
phobic feel. The game spaces should be very tight to emphasize this, while
also making sure that the player can’t get too lost in the dark.
• Use busy lights in areas that need to unsettle as well as attract. A slow pul-
sating light or a stroboscopic flickering give the player a goal to aim for, and
can provide local illumination. This kind of lighting unsettles by nature, so
it is a good way to frame dangers. These can be environmental—electrical
sparks, fire, lasers, and so on—or they can be AI based—an enemy guard
using a helmet with a flashlight, or weapons fire.
• Highlight exits clearly to indicate to the player where to navigate toward.
This also has the subconscious effect of teaching the player that well-lit ar-
eas are signposts towards safety. The exits can be diverse: they can include
actual doorways lit by an exit light, or a hole in the ceiling leading to the
next area lit by a nearby strip light.
• Start adding windows and additional light sources. As the player
progresses through the level and advances from area to area, it is useful to
slowly increase the intensity and the number of lit areas to give the player
a clear sense of progression towards an escape point. By adding windows
or other lookout points, the player can start to gain a better mental map of
the level, as well.
• Make the gameplay areas bigger as the player progresses towards the
level exit. To ease up on the sense of oppression and claustrophobia, the
environment should slowly become more spacious and bright. The player
is to be encouraged to keep following the path that leads up to brighter and
less negative environments.
• Have encounters start from farther away so the player can use the
light to their advantage. Enemy encounters or environmental dangers
Visual Experiences 215

should become recognizable from farther away as this will give the player
a psychological boost that they are starting to be in control of their prog-
ress. Note that the gameplay does not need to get easier at all, just less
panicky and unsettling. The encounters should become an obstacle be-
tween the player and his or her progress, rather than a consequence of the
darkness.
• Reward the player with a final exit that is both bright and expansive.
When he or she finally reaches the exit point of the level, the player should
receive a real reward for achieving the escape from the darkness—a fitting
finale for the journey they have been on. A good way to do this is to reward
the player with a grand and expansive view over a bright and warmly lit
vista. In this case it may be the roof of the building providing a sudden
unrestricted view over the whole city.

Further Notes
The example environment and setting above can easily be exchanged for one
that is more suitable to the game you are working on, like an underground min-
ing complex or a gothic wizard’s tower, for example. It should be easy to come up
with an original concept that allows the use of these techniques.

Example 10.2: Direction and

Misdirection—Camera Reveal
Summary
Every visual scene carries information. This is true in film, theater or level design.
The way this information is delivered to the player can be manipulated to add
impact and drama, especially if we use the game camera to underline the effect
we want to go for.

Game Genre
This technique is suitable for all games that have a camera system that can be
manipulated by the level designer. (This example uses a third-person game cam-
era.)
Goals to Achieve
• Frame important information in a level to grab the interest of the player.
• Reward the player for exploration.
• Deliver a revealing scene that adds drama and quality to the level.
216 Game Environments

Description
(Example type: Original)
We can use the principles of direction and misdirection to create a sequence of
events that add great value to a level. If we control what the player sees, through our
control of the game camera, then we can in effect create a little narrative sequence—
narrative because the images tell their own story, played out onscreen and accom-
panied by the player’s unfolding understanding of what it is he or she sees.
The gradual camera reveal. The gradual camera reveal is a technique that is
based on showing the viewer bits of scenery that, unbeknownst to the viewer,
actually comprises a much larger whole. As more and more of this whole is re-
vealed, a new understanding of the actual scene forms in the mind of the viewer,
until eventually a dramatic realization takes place when the final picture falls into
place. Difficult as it sounds this technique is relatively easy to implement on a
technical level.
Imagine the following sequence:
• Camera position 1: The (third person) game camera is positioned di-
agonally above the player character.
Scene: We can see the player character and some of his or her surround-
ings. The player seems to be on a dusty road, going through a desert.
• Camera position 2: The camera is positioned farther away, and tilted
more towards the horizon.
Scene: We can see the player character, and more of the road. The road
appears to be quite wide and long.
• Camera position 3: The camera is positioned even higher and farther
away. A further tilt at the horizon is established.
Scene: We can see the player character and a near panoramic view. The
road is littered by burned-out wreckage and rusted debris. The road is go-
ing towards an enormous structure on the horizon; we cannot yet make
out what it is.
• Camera position 4: The camera is lowered behind the player and finally
the full horizon can be seen.
Scene (final reveal): The horizon now shows the famous Cape Canav-
eral space shuttle launching pad. The entire structure including the space
shuttle is a smoking wreck. Pieces of debris are seen across the landscape,
along the road as well as elsewhere. Some enormous disaster has befallen
the installation.
All that is needed to frame such a scene is camera trigger boxes that tell the cam-
era what position to take, based on the player’s location. The result is a very dra-
Visual Experiences 217

matic and interesting setup. The player could subsequently find transportation
to the sight, on which the next level may take place.

Further Notes
The principle can be simplified if necessary. A top down camera view can be
zoomed out revealing that the player is surrounded by enemies, or a player can
reach the top of a hill revealing a new landscape that completely changes the
player’s perception of his or her environment.

Example 10.3:Visual Impact Techniques—

Scale
Summary
Use the dramatic scale of a thin bridge over a large and deep chasm for a strong
visual impact on the level.

Game Genre
This example is especially suitable for games that use a third or first person cam-
era mode and allow for action-adventure style gameplay.

Goals to Achieve
• Use dramatic scale in order to enhance visual impact.
• Successfully use the physical implication of scale and environment in a
level design scenario with multiple outcomes.
• Incorporate the physicality in a dramatically interesting sequence.
• Provide a strong set piece.

Description
(Example type: Original)
This practice example has the player engage in a set piece that has an outcome
that is crucial to gameplay and narrative progress. Whatever the outcome, it will
create a literal chasm between the player and the past. This can be done in a
binary fashion where success gives the player a sense of resolution, while failure
creates a temporary setback that can later turn into a delayed and therefore more
rewarding victory.
Setting and background. The player chases the game’s main villain to a rope
bridge spanning a chasm. The villain dispatches a group of henchmen to inter-
218 Game Environments

cept the player. The player has to engage the henchmen and defeat them in time
before the villain crosses the bridge and starts cutting the ropes.
Victorious outcome. The player defeats the henchmen in time to reach the
villain and capture him before he cuts the rope bridge. Now on the far side of the
chasm, players can cut the bridge ropes themselves, which will stop more enemy
creatures from following. (This act can be shown in a cutscene including the ar-
rival of hordes of furious but thwarted enemies on the other side of the chasm.)
Negative outcome. The player is delayed too long to reach the villain in time
before he or she can cut the bridge’s ropes. The bridge is cut while the player is
on it and the player falls down onto a ledge on the wrong side of the chasm. The
villain got away and the player is now forced to find a path down to the bottom
of the chasm, in the hope that a crossing can be found. There is now literally a
chasm between the player and the villain. Potentially a horde of further enemies
in pursuit can once again be seen to arrive, this time taking on the role of trying
to stop the player from finding a path across the chasm.

Further Notes
There is a level in Half Life 2 that successfully uses a bridge as a visually strong
backdrop to the action, as well as providing a vertiginous quality to the gameplay
that enhances mood and atmosphere to a large degree.
Audio Design
11
A s important as the visual dimension discussed in the previous chapter is,
there is a further important aspect to game environments. Although often
overlooked in creating our inner vision of the game world, the aural definition of
the gameplay space and our reactions to music and sound is vitally important.
This is heady stuff indeed, and something we should only attempt after taking a
look at the basic fundamentals of the way human beings experience audio.

Concept

The Physiology of Sound

The visual experiences chapter has shown us that studying the basic physiologi-
cal properties of sensory perception can teach us a great many things that are ap-
plicable to artistic expression. To build the same kind of foundation, we should
do the same thing when it comes to examining audio.

How Does the Human Ear Work?

Just as was the case in the chapters on visual direction, it is useful to get a general
idea of the physical principles involved in the subject we are exploring. By what
mechanisms are we able to hear? I am no expert on the matter, but in basic terms
that even I can understand, the following processes are important.
The cups on our head that we normally call ears are, as you know, just part of
the whole apparatus. They are called the pinna, and their main function is to cap-
ture sound waves in the air and direct them towards the ear canal. Incidentally,
the way that sound bounces around lets us judge it positional origin. The ear
canal ends at the eardrum, a thin membrane that can be caused to vibrate by even
219
220 Game Environments

the slightest pressure changes. When the eardrum vibrates, that signal is ampli-
fied in the inner ear, whose function is to transform the air pressure differences
picked up by the eardrum and turn them into fluid pressure fluctuations. This is
done by the cochlea. The fluid waves are picked up in the organ of corti, a part of
the ear lined by thousands of tiny hair cells, which translate the fluid waves into
electrical signals. These electrical impulses are taken by the cochlear nerve and
sent to the cerebral cortex, where they are finally interpreted.

Audio information
Audio waves carry information that can be vital to our well-being. As one of our
major senses, hearing is constantly used to dissect the world around us and give
us information and context in all kinds of circumstances. Our hearing range and
the stereo setup of our ears allow us to judge aural distances, and as importantly,
to judge the direction of that which produces the sound. (Is that tiger close? Is
it getting closer?)
If we think of the aural picture we have of our surroundings as a list of audio
information, we can construct a list of contributing factors that is surprisingly
long. Our environment provides us with echoes, the Doppler effect, muffled
sounds, clear sounds, soothing sounds, mechanical sounds, natural sounds, and
so on. All this information needs to be digested and interpreted.

Sound Interpretation
What we need to understand when dealing with the subject of sound is that even
though the physical aspects of hearing are quite well understood, human inter-
pretation of sound isn’t an exact business.
The previous dry description helps us understand the process to a degree,
but let‘s look at another way of describing it:
Imagine that you stretch a pillowcase tightly across the opening of a
bucket, and different people throw ping pong balls at it from different
distances. They can each throw as many balls as they like, and as often
as they like. Your job is to figure out, just by looking at how the pillow-
case moves up and down, how many people there are, who they are, and
whether they are walking towards you, away from you, or are standing
still. This is essentially the problem your auditory system has to contend
with when it uses the eardrum as the gateway to hearing.1
If seen in those terms, it is a miracle that we can hear at all. Indeed, we often do
get things wrong, sometimes in strange ways.
1
Daniel Levitin, “It’s Just an Illusion,” New Scientist, February 22, 2008, pp. 34–37.
Audio Design 221

A limit to sound perception

Human capability means that even being able to make this much sense out of
sound input is very impressive indeed, especially if we take into account the com-
plexities involved. It is nonetheless not limitless, and is subject to some strange
pitfalls. We may not always be aware of these things, but sometimes in a studio
environment they come to the fore with real clarity. The famous sound engineer
Walter Murch, who made films with people like George Lucas and Francis Ford
Coppola, talks about an interesting practical example that he encountered while
working on THX 1138. During the mixing phase of the film, he had trouble
making the footsteps of robots sound good. He discovered what the problem
was and offered this realization:
Somehow, it seems that our minds can keep track of one person’s foot-
steps, or even the footsteps of two people, but with three or more peo-
ple our minds just give up—there are too many steps happening too
quickly. As a result, each footstep is no longer evaluated individually,
but rather the group of footsteps is evaluated as a single entity, like a
musical chord. If the pace of the steps is roughly correct, and it seems as
if they are on the right surface, this is apparently enough. In effect, the
mind says “Yes, I see a group of people walking down a corridor and
what I hear sounds like a group of people walking down a corridor.
Sometime during the mid-19th century, one of Edouard Manet’s stu-
dents was painting a bunch of grapes, diligently outlining every single
one, and Manet suddenly knocked the brush out of her hand and shout-
ed: “Not like that! I don’t give a damn about Every Single Grape! I want
you to get the feel of the grapes, how they taste, their color, how the dust
shapes them and softens them at the same time. “
Similarly, if you have gotten Every Single Footstep in sync but failed to
capture the energy of the group, the space through which they are mov-
ing, the surface on which they are walking, and so on, you have made
the same kind of mistake that Manet’s student was making. You have
paid too much attention to something that the mind is incapable of as-
similating anyway, even if it wanted to.2
This indicates that as important as the ability to make sense of audio information
is, it is also subject to mistakes and confusion. How we process audio information
is an important aspect of the complete picture.

2
Walter Murch, “DENSE CLARITY – CLEAR DENSITY,” Volume Bed of Sound,
http://www.ps1.org/cut/volume/murch.html, no date.
222 Game Environments

As stated before, it is no easy task for the brain to get things right all the
time, and indeed it has to rely on extrapolation and inference. This leaves the
door wide open for manipulation of our audio perception, something that has
been raised to the level of high art in the field of music recording. Therein we find
that just as important as the physiology of sound is the notion of a psychology of
sound.

The Psychology of Sound

Beyond the physical there is another, even more interesting dimension.
What is it we actually do with all this audio information on a psychological level?
What feelings do we experience that are triggered by sound? Are there rules or
general principles we can look at? There are many questions like these that can
be asked about the way we experience sound mentally, so let’s look at some of
the basic areas of interest.

Describing sound
The mechanical explanation of sound is very useful, but it still doesn’t give us
enough of a vocabulary to describe sounds. For some people this language exists
in music theory, a comprehensive language that even has a written from in all
kinds of music notation. Although the world would undoubtedly be a more en-
tertaining place if everybody were taught this language from infancy, it still does
not provide a common vocabulary for our general experience of sound. Indeed,
it is interesting to see that most of us really struggle to discuss sound without
the aid of terms that are comparative with other senses. Sounds are described as
warm, or having a certain color, or being sad or melancholic.
Describing sound in this way shows that there is a deeper psychological di-
mension to our aural life.

Emotional Associations
Sound is a strongly associative medium. Almost everyone knows examples of
how sound is associated with specific emotions or events pertinent to their life.
Consider the pop hits of the time you were a teenager, or the TV shows you used
to watch as a child. Take any emotive memory, and chances are that a specific
inner soundtrack accompanies many of them.
This basic principle of sounds and emotions being associated with each
other is not limited to personal experience. Indeed, there are many emotional
associations that we can share with other people, often at a universal level. This
Audio Design 223

is where we can find a multitude of uses for level design. Find the general shared
associations that can work in a creative scenario, and you have found a remark-
ably powerful tool to help deepen the experience for the audience.

Aesthetic Sensibilities
The combination of cultural background and the properties of sound itself makes
us judge sound on an aesthetic level. This is especially true for music. There are
not many clear rules on what makes a sound pleasing or jarring; indeed, people
often completely disagree on this topic. Jazz can be irritating noise to one set of
ears and the ultimate musical expression to others. One person’s soothing wind
chimes are another’s insomniac nightmare. There are some historically and cul-
turally determined guidelines to be found in music theory; principles of har-
mony and dissonance come to mind, although these are often open to interpre-
tation. Nonetheless, we know that ultimately a judgment is made in many cases,
and often this judgment is predictable to a large degree.
This means that the aesthetic qualities of sound and music can be used in
a creative context. If one can predict that a sound will be pretty or calming, it
can be used as a contrast to something ugly or distressing, or it can be used to
strengthen the effect even more. The aesthetic sensibility of the audience is open
to all kinds of creative techniques; those involving sound and music are particu-
larly clear examples of this.

Primal Reactions
Some qualities of sound cause a reaction on a purely primal level and cause emo-
tional responses that cannot be ignored. This can be a matter of volume; nobody
can help but be startled by an unexpected loud sound, or this can be a matter of
pitch. The low rumble of an earthquake is bound to cause some serious unease
to the listener. Other qualities of sound may cause equally strong reactions—
human laughter or growling predators, to name a few. A mother’s reaction to her
crying baby is hard to describe as anything but primal.

Taught Reactions
On the opposite side of the equation, we find emotional reactions to sound
based on a cultural basis. That is to say, emotional reactions to sound that society
or our parents teach us. To a degree, these reactions can be seen as conventions
of sound. That is why rude workmen whistle after women or why we give a wide
berth to a truck that makes a repetitive beeping sound as it backs up. We know
224 Game Environments

what a disapproving tut tut means and can easily comprehend an overly sarcastic
comment due to the sarcastic tone with which the comment was made. Our daily
lives are filled with moments where we react to sounds that have meaning due to
conventions that we have learned to recognize.

Types of Sound
Within creative disciplines like film, television, theater, and so forth, there are
distinctions to be made between certain types of sound. There is a need to dif-
ferentiate and categorize for practical purposes, although there is no complete
consensus on how to do so.

Diegetic and Non-Diegetic Sound

One distinction that is often used is between diegetic and non-diegetic sound.
They are useful terms but require some explanation:

Diegetic
Sound is diegetic when it emanates from the actions visible onscreen, or when
the sound is explained by the implied sources coming from the film environ-
ment. This can include dialogue, objects in the set, the weather, and so on—
anything that has a natural source in the story world of the film.

Non-Diegetic
In direct opposition is non-diegetic sound. This is sound that comes from outside
the story world of the film or other medium and has no natural source linked to
the onscreen action. For example, mood music and voiceovers are in this cat-
egory.
This distinction between diegetic and non-diegetic can be very useful, but
it may also at times be a bit limiting, or too specific to film. I personally tend to
make a distinction between music and non-music first.

Music
When people speak of the music of a particular film or similar creative product,
they tend to refer to a number of diverse types of musical content. Some of these
types feature some unique properties that are worth looking at. For the sake of
brevity, the following examples are from the medium of film, but they often ap-
ply to other media as well.
Audio Design 225

Mood music
Mood music is music created to specifically create or enhance certain moments
within a film. This is generally not a complete song, but emotive music with
minimal structure, if any. It is not meant to stand alone as a musical piece, so it is
free from principles of song construction. The use of mood music is incredibly
ubiquitous in film, precisely because it is such a free and powerful emotional
tool. Some people dislike it for that exact reason; they see the emotional ma-
nipulation inherent in the use of mood music as inappropriate. The authors of
the rules of the DOGMA 95 school of film making even stated it as one of the
key techniques to avoid:
The sound must never be produced apart from the images or vice ver-
sa. (Music must not be used unless it occurs where the scene is being
shot).3
I am not advocating for or against mood music, but it is always interesting to see
how naked the content of a film can be if mood music is taken away. There is no
denying the impact of its presence or its absence.

Original score
The original score is just what the word suggests: the music originally and specifi-
cally written or recorded for the work. The music has typically been scored in
partnership with the film director and is frequently thematically consistent. Just
as a movie can feature some strong themes and recurring motifs, the same often
occurs in the original score. This is of course no coincidence, since the score is
meant to bring out the best of the work it accompanies. Some of the best musical
scores are highly rated because they are a particularly strong companion to the
underlying work, sometimes even on an equal footing.4

Soundtrack
Soundtrack and score are words that are often wrongly used as if they were in-
terchangeable. This is understandable, as there can be much overlap between
the two. But generally, the soundtrack is a product of music research and not of
composition. That is, the soundtrack generally consists of songs used in the film,
but not necessarily written for the film.

3
“The Vow of Chastity,” Dogma 95, http://www.martweiss.com/film/dogma95-
thevow.shtml, 1995.
4
The Kooyanisqatsi score by Phillip Glass is a good example.
226 Game Environments

Non-Music
Non-Music, sometimes called noise, is a term used in the film industry. Within
that discipline it doesn’t have the negative connotations that the term normally
has. In fact, it means no more than all the sound that doesn’t consist of music. That
is a very large group of sounds, so it is useful to identify some more specific
categories. Many of these categories are not as clear-cut as they appear to be,
and definitions vary from filmmaker to filmmaker. My personal understanding5
yields the groupings of sound categories discussed below. They are not strict
definitions, but are intended more as a rough guideline.

Sound effects and Foley

We speak of sound effects in cases where the sound is supposed to underscore
specific physical things or actions onscreen. Sound effects tend to be short but
can last a little while in some circumstances. In most cases they can be attributed
directly to things we can see onscreen. Good examples of this include footsteps
made by the onscreen characters, a gun fired by an onscreen character, a door
closing, and so on.
Another term for these kinds of effects is Foley. The best way to think of Fo-
ley is when specific sound effects are created to imitate actual actions. In the case
of footsteps, rather that recording the actual footsteps of an actor, a sound effect
is created that does the job in a more convincing manner. Strangely, in film it is
quite often more practical to create artificial sounds to represent actual sounds.
What the film ends up with is a sort of hyper-reality where sounds imply the
reality of the scene in a much stronger way than the actual sounds could achieve.
This also allows for more control over audio and enables the sound engineer to
try out all kinds of tricks and techniques to benefit the film.

Incidental sound
Incidental sound forms a different kind of sound effect, different enough to war-
rant its own category. Like mood music, incidental sounds are designed to give
power to certain psychological states of the viewer. For example, we all know
(and dread) the moment in a horror movie when something in the film pro-
duces a sudden and unexpected scare. In almost all cases, this moment is ac-
companied by some kind of loud sound effect that adds impact to the “incident.”
It is not a natural sound; it is created purely for the incidental purpose. These

5
I have studied some film at the university level and have and dabbled a bit as an amateur
filmmaker.
Audio Design 227

moments can also occur when the protagonist suddenly realizes something, or
when a dramatic moment is about to occur.

Ambient sounds
Ambient sounds, as already discussed in Chapter 9 on immersion, are linked to
the environment itself, as opposed to specific incidents or actions. They exist to
help give a general definition of a space or the environment that the audience
is experiencing. This creates a deeper sense of immersion, but it also conveys a
great amount of information. An ambient sound like machine-room hum tells us
that the machine room is active and purring along nicely. Hubbub in a meeting
tells us that everybody is talking among themselves, while fading out this ambi-
ent sound can signify that the group is focusing on a single speaker. Every envi-
ronment features ambient sound; it is almost impossible not to hear something.
Even when it is quiet, we tend to hear our own breathing or sounds like footsteps
or the rustling of clothes.
All of these sounds matter a great deal to the mood portrayed, and to the
information divulged to the listener. This can be enhanced even further by
manipulating the sound itself. Earlier we spoke of the extra information we
get from our environment by naturally occurring sound alterations like the
Doppler effect and echoes. We can add many others to that: reverb, cutting
out frequencies to create a muffled sound, resonance, and so on. It is not that
hard to manipulate sound in a studio, but the results can be incredibly power-
ful.

Concept Summary
We perceive audio through a diverse set of circumstances and experiences. There
are important physiological rules and restraints that allow us to hear in the first
place. These in turn are processed in our brains in very interesting ways, which
are not always correct. What is striking is how applicable many of these ideas
and principles are to creative expression. This is not just in a predictable manner
as musical accompaniment or performance, but along a whole range of uses of
audio, often in the subtlest of ways.
Luckily, there are numerous artistic conventions and techniques available
to us for study. What we should take from this is a spirit of open-minded inter-
pretation of audio principles and ideas and try to find applications within our
own craft.
As we will see in the next section of the book, there are many ways in which
we can do so.
228 Game Environments

Theory

The previous section shows how universal many auditory principles are. It also
shows how relevant knowledge of audio can be to creative application. Level
design offers a huge amount of scope and context in which we can place audio
design theory catered to our specific needs.

Level Music
Most game levels use music of some sort, sometimes to underscore in-game
events, sometimes to provide pleasant background music. This music isn’t al-
ways seen as something of great importance, and indeed in some games the mu-
sic used is inconsequential. But that is not the point. What we should do is look
at the role music can play, and when we adopt that view, we find that there are
many good reasons to incorporate meaningful music choices in our levels. As in
film, we can make a rough distinction between the score of the game and other
types of music, most prominently dynamic music.

Dynamic Music
We speak of dynamic music in video games in those cases when the music played
is dependent on dynamic in-game events. The use of dynamic music can greatly
enhance the emotional impact of a gameplay scenario. If used correctly, the mu-
sic can be triggered or cued to underscore moments that enhance the player’s
enjoyment of a game, for example by rewarding a victory over a tough enemy or
setting the mood at the start of a new level. The latter is an example where games
differ from films and require different choices. In a film the mood music can be
painstakingly composed or researched to fit every scene and situation at exactly
the right time. In video games on the other hand, there is very little predictable
linearity; most in-game actions are dependent on player choice and therefore
can occur at mostly non-specific times. This means that the use of mood music
has to be flexible to some degree and must fit numerous situations.
Dynamic music can also be used to give the player information related to
specific gameplay or in-game events. Just as background noise of a scene can
provide information that is vital to the experience, so can the music in a game tell
the player additional things that pertain to the onscreen action but may be hard
to convey in a different manner. If a player is close to losing control of a situation,
Audio Design 229

the right mood music can indicate this and enhance the tension inherent in the
situation. When a player needs to know that an environment is safe to explore,
specific music can indicate this. Whatever game you are working on will have its
own set of requirements, and it is generally up to the level designer to formulate
them.
This use of music to convey gameplay information has the added advantage
of giving direction to the player’s actions or understanding of game events. And
as level designers we should always be interested in tools that give us possibilities
to direct the gameplay experience.

Score and Soundtrack

Video games are much more dynamic than films and therefore offer less room
for the effective use of a musical score or soundtrack. That isn’t to say there is
no place for it. Indeed, some game types are perfectly comfortable with a set of
looping musical themes or songs, often in addition to moments where dynamic
music takes over. Specifically, classic arcade games and early console or com-
puter games often feature a traditional score accompanying the main gameplay.
Who can deny the genius of scores for games like Turican,6 Ghosts and Goblins,7
or Outrun8? The music research9 done for those games is of such quality that
many years later, these games are still associated with and remembered by those
scores.
Finding or composing the right score for a game is not a trivial process. Try
to imagine films like Pulp Fiction or The Godfather without their soundtrack or
score. Choosing the amount and implementation of this kind of music is com-
pletely dependent on the individual game’s requirements. In all cases, some
thought needs to be given to a few specific questions.

Noise
As already discussed, noise is what we call all audio that isn’t music. As a de-
scriptive category, that doesn’t give us enough information however. We need to
divide it into further subcategories. Or at least, into the ones that are typically
used in level design.

6
Developed by Lorriciels.
7
Developed by Capcom.
8
Developed by Sega.
9
The process by which the right music is found for the product in question.
230 Game Environments

Ambient Sound
In the chapter on immersion, I said the following about ambient sound:
We all know of a number of aural game clichés; on top of a mountain we
hear wind, in caves we hear drips, etc. However, these kinds of straight-
forward implementations are not just representations of sounds that
need to be picked from a list to complete a level’s outstanding tasks.
Recorded sounds not only have documentary qualities, they also have
psycho-acoustic qualities, which means they have the ability to have an
effect on the psychological interpretation of sound.
It is worth examining these types of sounds a bit closer. It seems that a further
distinction can be made between ambient sounds and their documentary quali-
ties and their psychological qualities.

Documentary
The documentary qualities of sounds are those that make it possible for them to
help define a game’s space, and to even help describe what happens within that
space. This may seem straightforward, but in reality is far from it. Here’s a simple
test: look around you and identify every single thing10 that can create sound,
no matter how loud or frequent. Chances are you won’t finish this exercise be-
cause the list is too long. This shows that we cannot go into a level design pro-
cess thinking that for documentary ambient sounds we can just assign a sample
to every potential sound-producing item. Not only will this completely destroy
our budget (both of money and memory); it will also create an aural mess that
muddles the definition of space, as opposed to helping define it. Remember that
we are experiencing these sounds through speakers or headphones; it is not like
we are actually there.
To create a clearer sound plan to describe the environment for most levels, it
makes sense to identify the key audio components and implement them wisely.
In essence, this is an impressionist approach: we pick those sounds that let us
paint a picture that is so descriptive that the player’s mind fills in the rest. Gener-
ally, this means we need to find those sounds that the player will associate with a
complete set of environmental expectations. A reverberating dripping sound can
be enough to define an environment consisting of watery caves.
Additionally, some sounds may have to support certain aspects of game-
play11 and therefore are even more important than others. Let’s go through some
typical examples:
10
Or creature.
11
More on this in a bit.
Audio Design 231

• Gameplay related:
■ footsteps (allow the player to judge others’ locations),
■ AI states (allow the player to react to behavior of creatures/NPCs),
■ engine frequency (allows the player to judge speed).
• General documentary audio:
■ crickets (define the desert at night),
■ hubbub (defines crowds of people),
■ rain (speaks for itself).

Psychological
Just as important as the purely descriptive nature of the sounds types discussed
so far are the psychological aspects involved: the psycho-acoustic qualities of
sound. Just as color theory gives us insight into a person’s reactions to certain
colors, and as scale can affect people’s emotional state in profound ways, sound
can have an emotional impact or cause a very specific reaction. Is there anybody
who doesn’t associate a countdown timer with haste or stress? Or to take a sub-
tler example: the sound of seagulls can exert a calming or even relaxing influence
on the player due to its association with the coast.
Documentary aspects in sounds help describe a physical space, while psy-
chological association helps describe a mental space or state. This is a very helpful
phenomenon for level designers, as quite often we try to exert influence over the
player’s state of mind.
Seagulls or countdowns are just two examples. In practice, each level
and each game will have a unique sound signature, and it really pays off to
spend some time thinking about which sounds best serve the overall experi-
ence.

Sound Cues
Many aspects of a level’s ambient sounds have to do with the fact that they
are information carriers. They tell their own story by way of the informa-
tion they carry within themselves. Sometimes this information is relevant to
gameplay decisions. When it is, we can classify them as sound cues, or, even
more likely, we can include sounds that were designed to be sound cues. We
have encountered some of these already when we were discussing documen-
tary aspects of ambient sounds, and those examples still stand. However, we
can add to this by realizing that sound cues don’t have to be a side effect of
environmental noise. They can (and often should be) specifically designed
to aid gameplay.
232 Game Environments

In other words, designing the speech of enemy AI in a level shouldn’t just

be a function of documentary sound design. It should be exploited in such a
way that it also fulfills specific gameplay goals. For example, if the AI has three
distinct behavioral states (such as passive, alert, and aggressive), it makes sense
to gear its speech to express those states in a clear way. That way, the player can
correctly read their behavior and the level designer can exploit this to create ap-
propriate gameplay scenarios.
This doesn’t just apply to character or creature sound cues. Anytime there
is a gameplay consequence or dimension to sound, it can be incorporated into
the level design. Electrical hum can be turned on or off in order to say whether
electricity is live. A dry click can indicate it is time to reload a gun, and so forth.
At times these decisions will be made within the context of the game’s overall
design. But just as often they are within the responsibility of the level designer,
or at least are fueled by level-design needs.

Sound Effects
At other times the level needs to communicate information through sounds that
don’t fit within the natural environment of the game. In those cases we generally
speak of sound effects. In Zelda games these may be sounds that are played when
the player solves a puzzle. In Unreal Tournament12 they can represent the re-
spawning of an item. There are thousands of examples for each individual game.
The nature of these kinds of sounds is almost always decided at a game design
level. However, in some cases, the use of these sounds is up to the level designer,
perhaps within the context of a scripted sequence or if the game’s development
tools provide extra leeway in these matters.
A surprising amount of knowledge is needed to design effective sound ef-
fects, and likewise, the level-design implementation of sound effects also re-
quires great care. If you are in a position where the level design is required to
deal with this matter, make sure that you don’t underestimate this task. Some
questions to ask include:
• Do the sound effects jell with the soundtrack or score?
• Can they obscure important sound cues?
• Do they clash with the general tone of the level?
• Can they be repeated often without causing irritation?
Questions like these at least need to be considered to make sure that no prob-
lems occur in the implementation of sound effects.

12
Developed by Epic Games.
Audio Design 233

Soundscapes and Level Design

In some games the use of sound is so fundamental to the experience that it re-
quires extra attention during the level design process.
For example, the independent game developer Nifflas13 has always made
clear that in his games, the music is as important as the gameplay. The way he
makes this musical aspect count is by providing environments that are strongly
underscored by the ambient music that he creates to accompany it. The two
cannot be separated; the joy of exploration that is central to his games is partly
caused by the music that fills the new environments or game areas that the player
encounters. Quite often these discoveries of new areas and music form the de
facto reward of the gameplay. In that important sense, we have to conclude that
the music becomes an integral part of the level design itself.

Music Games
A special case in this chapter has to be made for music-based games. In these cas-
es, the music often IS the level design, so I want to spend at least a few words on
the subject.

Rhythm Action Games

The classic form of music game is the rhythm action game. This is a form of game
not unlike an elaborate game of “Simon says,” where the player is asked to perform
specific actions to the rhythm of the game’s music. There are many examples, but
a typical one can be found in Parapa the Rapper,14” where following the onscreen
prompts to press joypad buttons in time causes the game’s protagonist (a rapping
dog, no less!) to win musical duels against a diverse set of opponents.
Games like this, and there are many, mostly test the player’s rhythmic skills.
The design of the difficulty levels often focuses on the speed of the task and the
ability of the player to pick out the right instrumentation in the music tracks and
the subordinate rhythms they perform. Examples include a string section in a
classical piece or a horn section in a funk piece.
The recent success of the Guitar Hero15 franchise shows how much mileage
this genre has. Watching a skilled Guitar Hero player shows somebody complete-
ly immersed in the flow channel of an optimal game experience.

13
Real name: Nicklas Nygren. See http://nifflas.ni2.se/.
14
NaNaOn-Sha.
15
Developed by Harmonix.
234 Game Environments

Music Creation/Emulation
There are games that take a less strict approach and allow a focus on actual music
creation. This is an area of gaming that is much less defined, so consequently the
level design can take many shapes. An interesting example of this is the musical
toy Audio Plankton, which asks the player to create compositions by manipulat-
ing the behavior of musical plankton.
A giant in the genre is Sony Europe’s smash hit Singstar, a game that requires
players to actually sing the part of various well known songs and be judged on
how well they keep to the correct pitch.
With games like this it is hard to talk about actual level design, although to
a degree there is significant overlap. For example, the difficulty level of the songs
and the songs themselves, or the expression possibilities of the electro plankton
all feature aspects of applied game design.

Music as Level Design

In some cases the level design is actually generated by the music itself,16 and a
few games have found a measure of success following this principle. Vib Ribbon
is a classic and clear example of this kind of gameplay. The game follows a simple
platform game formula, with the added twist that after the game has loaded, the
player can use any music CD in the console’s drive to generate a new platforming
landscape. Other games exist in this genre, but much more can be made of his
principle than has been attempted so far.

Theory Summary
Game audio is a vital component of the potential enjoyment of a game. Its im-
plementation covers many diverse areas, including player psychology and mood,
gameplay information, and at times specific gameplay design. In other words, we
cannot create a level design without at least spending some time on the impact
of these decisions.

Audio Plan
For that reason, it makes sense to create an audio design or audio plan for each
level you work on. The shape this takes depends on the type of game and on the
type of audio assets required, but just going through a process that takes into ac-

16
A worrying idea for those of us who are paid to do level design.
Audio Design 235

count the impact of the aural dimension on the level design can be invaluable to
the final quality of the level.

Practice
Example 11.1: Psychological Impact—
Tension and Sound
Summary
We have established previously that sound can impact on the mood of the player.
It can have properties that can have a major impact on how the player feels about
a situation on an emotional level. This impact is easily recognized when we see
efforts aimed at adding tension through sound.

Game Genre
This technique is suitable to most games.

Goals to Achieve
• Disturb and unsettle the player though sound.
• Enhance the impact and mood of the level.

Description
(Example type: Existing)
The sound design of Alien. Two good examples of this technique are found in
the film Alien,17 by Ridley Scott. They are based on the same principle, but each
adopts a different approach.
Heartbeat sound. A subtle but effective trick throughout the film is the use
of a heartbeat sample in the film’s audio track. Throughout the film, as the ten-
sion increases, the heartbeat sound increases in tempo. Most people are not even
aware of this, but the increasingly rapid heartbeat is disconcerting and creates a
17
Directed by Ridley Scott, distributed by 20th Century Fox, released May 25, 1979. The
film follows the ordeal of the crew of the space ship “Nostromo” as they are stalked by a
murderous alien.
236 Game Environments

palpable tension in the viewer. We can’t help being affected by it as we associate

a rapid heartbeat with danger or fear if experienced in the right context.
A similar effect can be achieved with breathing, turning calm and slow
breathing to rapid and panicked breathing.
Proximity sensor. While the heartbeat example is covert and subtle, we can
also find a more explicit example in the film when the endangered crew builds a
device that acts like a proximity sensor. It emits a low beep at regular intervals,
but when it picks up movement the intensity and the pitch increases if the move-
ment is getting closer to the emitter. The crew uses it to try to find or track the
alien that has been terrorizing them. It is a very simple conceit but one of the
most effective tension devices in the film. The audience (and the crew) cannot
see what the sensor picks up, and when the emitter starts to emit ever more fre-
quent and higher beeps, the tension mounts to very high levels indeed.

Example 11.2: Sound as Information

Carrier—Environmental Dangers
Summary
Sounds tell us a great deal about a space, and they can tell a great deal about the
gameplay or purpose of a space in level design. Sounds carry information to us
whether we are aware of it or not. Level designers can use this fact to greatly
enhance the level design, by providing crucial gameplay information through
the use of sound.

Game Genre
This technique is suited to nearly all games.

Goals to Achieve
• Create a level environment where sound conveys gameplay information.
• Translate this information into something perceivable to the player.
• Tie it into a gameplay scenario.

Description
(Example type: Original)
Once we start seeing sound as a potential information carrier, it becomes rela-
tively easy to tie sound directly into gameplay. All we have to do is provide a
Audio Design 237

situation where the sound carries information that is necessary for the player to
successfully deal with a level design scenario. This example shows how a number
of sounds can work together, in concert,18 to provide a comprehensive amount
of aural information of great use to the player.
Dangerous factory. The player finds him or herself in a fantastical factory en-
vironment and has to navigate a route through dangerous machines. The player
can get hurt by stomping pistons, jets of flames, manufacturing robots armed with
automated nail guns, and so on.
The machines cycle between dangerous and unsafe states, but these cycles
are not predictable through visual observation. They are not regularly spaced
out across a timeline. However, the machines all employ their own characteristic
sounds that precipitate a state change. In other words, the player gets an audible
warning when a machine is about to become dangerous or safe.
Careful listening and timing can then do the rest.

Further Notes
We have to be careful with these kinds of audio-dependent gameplay scenarios,
as they can completely ruin a game for the deaf and the hard of hearing. Alterna-
tive routes may be an appropriate way of dealing with this, or perhaps optional
visual indicators can do the trick.

Example 11.3: Sound as Information

Carrier—Multiplayer Sound Cues
Summary
As the previous example established, sound can carry information that is vital to
gameplay. This is true in single player games, but takes on an extra dimension in
multiplayer games, specifically in the area of sound cues.

Game Genre
This example is suited to multiplayer games where sound carries over distance.

Goals to Achieve
• Incorporate sound cues in a multiplayer context.
• Make it multifunctional.
• Tie it into interesting gameplay choices.
18
Half a pun, my apologies.
238 Game Environments

Description
(Example type: Original)
In multiplayer games it can be a matter of great importance to know where one’s
opponents are and what they are up to. Based on this information, players make
choices on what their next actions are and formulate strategies accordingly. This
is especially true in games where the same levels are played over and over again,
as this repetition means that the player can have a repertoire of strategies that are
derived from their knowledge of that specific level environment.
This should be acknowledged and utilized in the level design for such
games. There are, of course, all kinds of ways this can be done: the geometry of
a level can provide strategic viewpoints to facilitate this, the level can be made
of mirrored sections, balanced for all teams, so players can predict routes and
distances, and so on.
Less prevalent is the use of sound to give gameplay critical information.
There is much to be gained by employing this technique, however. In Figure 11.1
we can see a number of corridors that lead to an important pickup, let’s say heavy
armor (A). Several rival players (P1–P3) can potentially get to the armor. This is
not an atypical scenario, and there is nothing wrong with it per se.
We can make much more of this scenario, however, by introducing sound
cues to the mix. If we surround the pickup by puddles of water, and make sure
the audibility of running through the water encompasses all the potential player
positions (grey circle) then we suddenly have an entirely more interesting setup.
(See Figure 11.2.)

Figure 11.1. Multiplayer corridors with pickup and players.

Audio Design 239

Figure 11.2. Multiplayer corridors with pickup and players, and added sound
cues.

Player 1 can see the armor, and he or she could hear if somebody (Player 2
or 3) runs through the water and approaches the armor from the other side. If
Player 1 decides to run to the armor, Players 2 and 3 can tell by the sound from
which corridor Player 1 comes, as there is less water there than in the other cor-
ridors. (Players 2 and 3 can make that same distinction for each other, as well,
if one of them approaches the armor.) They can also hear in which direction
Player 1 goes after picking up the armor.
All this adds much depth and strategy to what previously was no more than
a few corridors and a pickup. Yet, it terms of assets or work, the technique is
very cheap to implement, and therefore a good addition to the level designer’s
toolbox.

Further Notes
It isn’t hard to add further sound cues to this scenario, but it is dangerous to add
too many. Beyond a certain number of sounds, the players will just get over-
whelmed and cannot easily react to all the aural input coming their way.
Game Stories IV

E arlier in the book we discussed the alleged disagreements between ludolo-

gists and narrativists and concluded that play and narrative aren’t mutually
exclusive. Play is covered to a degree in almost every chapter in this book, be-
cause it is natural to do so. However, concepts of narrative and story haven’t been

“Medieval Writing Desk,” Wikipedia, http://commons.wikimedia.org/wiki/File:Medie

1

val_writing_desk.jpg, 2009.
241
242 Game Stories

discussed much yet. This is partly because I made the decision that most of the
narrative-related topics are better handled in a separate and focused area, rather
than constantly addressing them in each subcategory. The result is the following
chapter, which explores general concepts and tries to apply them to level design
theory and practice.
Story and Narrative
12
A lmost all the arts and entertainment forms we are exposed to deal in some
way with narrative elements, storytelling, and creation of dramatic con-
tent and meaning. This is obviously true in most films that we see, books that
we read, and plays that we watch. We find it in comics and graphic novels, and
to some degree it often applies to less straightforward examples like painting,
music sculpture—you name it.
These expressions occur for many reasons, sometimes artistic, sometimes
informative or intended to make people laugh. There is a wealth of narrative
content that dates back thousands of years to cave paintings and probably even
farther. So, what lessons can we learn when we look at all of these types of nar-
rative in different disciplines? Are there unified concepts that transcend notions
of genre or art form, and if so, can we take this commonality and distill general
concepts from it?
To try to achieve something of that nature, this chapter will not try to write a
treatise on “the elements of style” or to find the ultimate definition of storytelling.
Instead, our approach will take us much further down the line and will simply
assume that we all agree that elements of narrative and story play an important
part in most art forms. We will also put forward that it follows that to be effective
in those instances where this is applicable, artists and entertainers through the
years have developed useful tools and techniques that help them bring dramatic
value to their work. This chapter is mostly interested in that specific aspect of the
craft and takes a look at a number of commonly used dramatic tools.

Concept
Dramatic Tools
There is a thing that most forms of art and entertainment have in common with
each other. They all want the audience to experience something, as intended by
243
244 Game Stories

the artist. They all want to get their ideas across in such a way that it makes some
kind of impact in the mind of the audience. In the cases of static art forms like
photography or sculpture, this impact comes from the audience contemplating
what it is observing. However, when it comes to non-static forms of art or enter-
tainment (and by non-static I mean work that progresses along a timeline), the
author has time to assert some control over the experience. Artists try to influ-
ence the audience as much as possible, so that it will perceive the work to have
dramatic (or artistic) value or impact. Almost every field has developed a large
array of dramatic tools that are employed to manipulate the audience in such a
way that the work is most powerful. Let’s look at a number of general examples.

Theme and Mood

Let’s assume that in most cases artists have something to say. It is fair to assume
that they generally want their art to have some meaning or at least be about some-
thing. There will be ideas that the artist is more eager to get across than some
other specific points. What we are talking about in this case are the major themes
of a work, and to a degree, the mood that the work tries to capture.
Themes can be major epic affairs that comment on society as a whole,
Tolstoy’s War and Peace or Dostoevsky’s Crime and Punishment come to mind.
Equally valid are works that deal with specific emotions or subjects that are
less sweeping, like the innocence of childhood in coming-of-age stories or the
moods evoked by the seasons.
One work can cover several themes and moods, such as a long-running TV
series. Others may focus specifically on one aspect, as is often seen in poetry.
The easiest way to think of theme is to ask the question: “What is this work
about?” More often than not, the answer will list the themes of the work.
Something that is related but slightly different is the mood of the piece,
which generally is found by asking how the work makes one feel.
Theme and mood rarely operate in isolation; generally, they touch on many
other aspects of the work. Other choices about the content are often made to
strengthen the impact of the main themes and overall mood. Good examples of
this are choices of setting, form and style.

Form and Style

Form and style define how a story is told.
Once it is known what themes and moods are to be explored in the work, a
choice has to be made on how to effectively shape it. It needs to have a form in
which it can be defined, and a style in which it can be approached.
Story and Narrative 245

Form
The form (or format) is no less than the chosen form of expression. This is some-
thing that is governed both by personal choice and by the theme and mood of
the work. “A lyrical expression of love” can be shaped as a poem or a western
pop-song, among other possibilities. Each choice will bring with it numerous
defining attributes that will have a huge impact on the final work.

Style
Style is the way in which the final content is approached. So to go back to our
earlier lyrical expression of love, if we chose poetry, we can make a further sty-
listic choice. Are we going for a traditional Japanese haiku or an experimental
tone poem?
If we chose the western pop-song format, we can also choose from many
styles, such as 80s new romantic, indie guitar music, reggae, and so forth.1
Any of these choices can prove to be effective, but they all heavily impact
how the work will be experienced.

Setting
Setting tells us where things happen.
In order for the major themes and mood to really resonate, a choice of set-
ting (if applicable) can have a huge impact on the dramatic effect of a work. Blade
Runner would be less respected as a work of noir science fiction if it had been set
in some bright future kindergarten.
The best settings let the artist underscore the chosen themes and mood in
a clear and effective way. This can be done through a direct connection between
setting and story. When commenting on what war does to humanity, it is effec-
tive to tell the story against the backdrop of a major war.
It can also be done through the juxtaposition of contrasting elements. What
I mean by that is that sometimes certain themes have more impact if contrasted
with a setting where they stand out. An explosion will have more impact in a
silent library than in a carnival setting.

Dramatic Progression: Plot and Pacing

Many creative works feature some kind of progression over a timeline. If there
is a discernable plot or narrative, it tends to follow a path through time, allowing
the narrative to unfold. In those instances, choices have to be made dealing with
1
Please don’t snort and just insert examples of current pop music, okay?
246 Game Stories

the right order, speed, and way in which to progress the story in such a way that
it has the desired dramatic impact.
We have already discussed this to a degree in Chapter 4, where we looked at
the traditional theatrical five-acter, as demonstrated through Freytag’s pyramid.
But other, non-literary, examples can also be found as well. Musical composi-
tion, for example, also tends to follow a time line, and in some cases this has led
to dramatic progression through chord progressions or similar musical forms.
Ask any professional musician about the importance of the 12 bar blues and you
will see what I mean.
Whatever the field in which dramatic progression occurs, there are choices
to be made dealing with the order in which to tell the story and its individual
components (plot) and the timing and speed applied to the use of these compo-
nents (pacing).

Plot
Plot is often confused with story. Quite often, when people try to explain the
story of a movie, they resort to telling you about what happens in the movie in
chronological order. But this does nothing to explain the story’s overall themes,
moods, and intentions. It merely tells us in what order “stuff happens.”
A well-plotted narrative employs plot choices with a great regard for the over-
all purpose of the work. It can include delayed revelations to heighten a sense of
suspense, or early background information to strengthen character motivation.

Absence of plot
Just as important as including events in a plot is excluding them from the whole.
A plot can skip many years in a character’s life so the audience can feel a sense
of mystery about how that character has progressed from an earlier encounter. It
can set a character on a specific path of dramatic progression and examine what
this has done to the character’s life twenty years later. In the process this can
strengthen the themes of the story as a whole.
The ultimate point of plotting in any of its incarnations is that it gives artists
and entertainers a tool that enables them to exert control over the way their work
is experienced. It allows them to direct the experience over a specific timeline,
determining which emotions are felt at what time.

Pacing
Pacing, on the other hand deals with the speed and rhythm in which plot is ex-
posed to the audience, generally in such a way that they are continually engaged
Story and Narrative 247

by it. This matters greatly, as the audience experience is subject to fatigue or

boredom or other distractions throughout.

Editing
To go back to the film example, imagine a film-editing suite where the editor and
director are looking at hundreds of reels of films with many takes of unedited
footage. When they start to splice it all together,2 they will constantly be faced
with very specific choices on how long to let scenes and shots run in the final
edit. The movie footage will be put in order (plotted) and then made subject to
durations; or in more practical terms, the film will be paced in a certain way.
These are not minor decisions. A deep, wistful ghost story may not be best
told at breakneck speed with furiously fast scenes racing the audience through it.
Time is needed for reflection and for a sense of foreboding and fear to gather.

Concept Summary
A helpful way of thinking about these choices regarding dramatic tools and ele-
ments is by placing them in a hierarchy. We have done this earlier in the book
by trying to formally place level design (and game design) elements in a specific
development hierarchy to determine an approach that lets us assign goal driven
values to the design process. We can do a similar thing on a smaller scale with
narrative elements. In this case the hierarchy looks as follows:
• Theme/mood:
■ form,
■ style,
■ setting.
• Dramatic progression:
■ plot,
■ pacing.

Theme/mood. Theme and mood tell us what the work is about and how it makes
us feel.

Form, style, and setting. After choosing the theme and mood, we can enhance
them by choosing an effective setting, the right form, and an appropriate style.
These choices now tell us where our story occurs and how it is told.

Dramatic progression: Plot and pacing. Finally, the story or narrative itself is
constructed and placed in some kind of progression over time. The story unfolds
2
Well, the contemporary digital equivalent of slicing film.
248 Game Stories

from a beginning to an end of sorts and is made up of individual (plot) elements.

Plot gives the story a cohesive narrative.
Then there are choices to be made on how much time to give individual plot
elements, which determines the pacing of the whole.
These dramatic elements don’t always apply. For example, I don’t know the
“plot” of Rodin’s The Thinker, but we do now have an overview of sorts of a num-
ber of useful dramatic tools from which we can choose when we need to. Let’s
see how they relate to level design and its dramatic elements.

Theory
In the previous section we discussed a number of ways in which creators and art-
ists use specific techniques to create or strengthen dramatic impact. They con-
struct their narratives in many diverse ways and all kinds of creative expressions,
but there are specific things they do that seem to overlap between disciplines.
What I want to do in this chapter is re-examine these kinds of tools and
techniques through the lens of video games and level design, see where they ap-
ply, and find our own ways of implementing them. Additionally, I want to exam-
ine where games are different, and where they require entirely new or different
ways of dealing with narrative and story aspects.
To do this effectively, we should look at those types of games where story
matters most, as games like that will have had to deal with these questions in the
past. Let’s look at a number of story-reliant game genres, including non-video
games, and then see how they can be approached with the previously discussed
techniques, as well as with potential new ones.

Story-Dependent Gaming
Before we try to define our own methods of in-game storytelling, we should find
out if others have already dealt with this subject matter before. I have highlighted
three types of gaming where storytelling takes centre stage, yet interactivity and
gameplay are still massively important, or even defining to the experience. They
are point-and-click adventures, artificial reality gaming, or ARGs, and traditional
pen-and-paper role playing games, or RPGs.
All of these genres have in common that the story or narrative elements are
crucial to the experience as a whole and therefore provide good examples for
study. They each have their own level design conundrums to deal with.
Story and Narrative 249

Adventure Games
Games like point-and-click adventures and text adventures belong to a genre of
games where the player takes on the role of a character, and performs tasks and
solves puzzles in a story-led adventure. These games can be purely text based, or
they can take place in a 2D side-scrolling environment, as seen in classical point-
and-click adventures. There are many famous examples, including the likes of
King’s Quest, Hitchhiker’s Guide to the Galaxy, and The Secret of Monkey Island.
There are even 3D examples of the genre, with Grim Fandango deserving a spe-
cial mention.
In pretty much all examples of the genre, the story matters and cannot be
removed from the gameplay. Slowly uncovering the story threads and seeing the
adventure unfold through the eye of the character(s) in the game is a core ele-
ment and a major attraction. Most of the actual gameplay will consist of puzzles
that have to be solved. At times they are fiendishly hard or infuriatingly illogical,
but the central experience is undeniably story led.

Adventure game level design implications

Because so much of the medium is based on taking the player through a pre-
defined storyline, these games tend to end up as fairly linear affairs. There are
ways to branch off story events and reconnect them later to the main story line,
but the end result is in most cases fairly rigid. This limitation is not necessar-
ily a weakness of the genre, though, as long as the story is compelling, and the
puzzle solving is rewarding. Interestingly, the player actions and story develop-
ment often overlap. Dialogue is presented interactively by player choice, gen-
erally through conversation with NPCs. This is often absent from more inter-
active games, where story exposition frequently takes place in non-interactive
cutscenes. This may well be because the story is so important to the experience
of the game that it is only natural to include the player’s actions in the overall
narrative when possible.

Alternate Reality Gaming

A relative newcomer to the (video)game family is the Alternate Reality Game,
normally abbreviated to ARG. ARGs are communal games that allow players to
take on the roles of participants in a story unfolding in an alternate reality (cre-
ated by the game makers3) and told via various real world media. The online
component is very important to these games, but other real world media are

3
Known as Puppet Masters.
250 Game Stories

used as well. A player may find clues to a puzzle on a website, have to ring a real
number, and then find an item in a real world location. Yet despite the overlap
between the real and the virtual, a key aspect of these kinds of games is that they
tell a specific story, and the players are characters within this story.
It is a fascinating genre, which at times is able to attract large numbers of
players, yet its successes are fairly unknown to most people. Famous examples
include The Beast, a seminal ARG created as a marketing tool for the Spielberg
move AI, and Perplex City, a game that included the additional uses of collect-
able cards and a cash award of 100,000 pounds on finding a real world object
called “the Receda Cube.”4
A particular quirk of ARGs lies in the fact that gameplay and story span sev-
eral forms of media, and as such, the genre comes with interesting and unique
challenges, and therefore with unique solutions. One such quirk is that it is a
general rule in ARG etiquette that the game and its players stay in character,
no matter how unlikely or fantastic the content. Although the puppet masters
formally (and secretly) design the gameplay and the story, they never allude to
this in any way or form. All communication and actions within an ARG universe
are in character and treat it as if it is not a game. This principle is often referred to
quite literally as TINAG, which stands for This Is Not A Game. The point is that
the interest of the game comes from people communally “living” and influenc-
ing the game narrative. The puppet masters may largely control the immediate
story events and the puzzles, but the memorable actions come from the choices
of the players.

ARG level design implications

ARG level design issues are complex and interesting. How does one tell a story
that is to be enjoyed by many thousands of players at the same time, but whose
progress is partly dictated by the pace of the players? How can somebody jump
into an ongoing storyline like this when most of the players have already experi-
enced it months ago? This is not a small problem, as the core gameplay is really
about participating in the story, which makes it hard to be engaged when many
of the story events are already handled by players who have come before you. It
is a balancing act, where events are created that need sufficient players to crack
the puzzles and obstacles, but the game as whole cannot always rely on having
large groups of people all the time.
Because human nature is unpredictable, many ARGs can revel in the fact
that their story lines unfold in a way that is partly dependent on player action. In
that sense, ARGs are very democratic games. Level designers (puppet masters)
4
Don’t go looking for it; it has been found, and the prize money is gone. GONE!
Story and Narrative 251

are very much facilitators, people who create rough outlines to be explored, ex-
perienced, and influenced, by the players. This is an exciting prospect, key to the
success of many ARGs.

RPGs (Role-Playing Games)

RPGs are role-playing games, which might sometimes be described as “playing a
dark-elf-smiting-warrior in pen-and-paper role-playing games,” as in Dungeons
and Dragons.5 For more information, see the section on “Level Design in a His-
toric Context” in Chapter 1, but it may be worthwhile to recap somewhat here.
So-called pen and paper role-playing games have been around for awhile, but
the most famous example of the genre is the fantasy role-playing game system
Dungeons and Dragons, developed in the early 1970s by David Arneson and Gary
Gygx. The genre typically consists of a Game Master who leads the gaming ses-
sion, presiding over the rule system and the adventure that is presented to the
players, who act out the actions of the characters they portray. The Game Master
interprets those player actions and makes sure they relate and interact properly
to the game’s official rules.
In addition, the Game Master narrates the consequences of players’ actions
and the independent events occurring in the prepared adventure. In many ways
the Game Master acts as the player’s eyes and ears in the virtual world and pro-
vides the simulation data that follows the game’s rule set. The players inform the
Game Master of the actions they wish to take and communicate in character
with the other players. The resulting game is one where players partake in virtual
adventure stories in a fantastic, interactive simulated gameplay session, bound
by a set of formal rules that provide the necessary level of simulation and logic to
bind the whole thing together.
This all may sound a bit complex or even contrived, but once a group is
used to the system, it is remarkably immersive and effective. RPGs can cover
many themes and moods and form a fairly diverse genre. Others include FA-
SA’s Shadow Run, which mixes cyberpunk with fantasy, Middle Earth Role Play-
ing, based on the Tolkien books, or Call of Chthulu, which has a Lovecraftian
setting.

RPG level design implications

To understand the concept of pen-and-paper role playing well, it is important to
acknowledge that the ROLE PLAYING elements are as much part of the genre

5
Nor does RPG stand for Rocket Propelled Grenades, or horrible simulations at the work-
place about dealing with a difficult client.
252 Game Stories

as the GAME elements. In strict terms, one could argue that the game elements
and the world simulation elements are no more than a backdrop for the chance
to play the role of a character and interact with the virtual world though its act-
ed-out actions. This leads to interesting results; if the role playing is interesting
enough, the player can have a rewarding experience trying to light a fire for hours
in some featureless tundra.
Once again, the level designer, this time called the Game Master, is an en-
abler of dramatic and interesting player actions. As before, the player’s actions
form the central story. Drama comes from the player’s responses to the envi-
ronment and its interactive contents, as presented by the Game Master. Even
though the Game Master will have prepared a detailed “module,6” the player’s
actions in reaction to its content are what defines the gameplay session.
Just as in ARGs, immersion is key, and role-playing games require players to
stay in character in order to make it all work. The Game Master performs the role of
level designer and is responsible for framing the story in such a way that it becomes
an enabler of dramatically interesting player choices and subsequent actions.

Similarities between Story-Reliant Game

Genres
The above examples show that even in those game genres that rely heavily on
story there still is plenty of scope for interactivity.
The main problem that arises from these discussions is one of definition.
“Story in games” can mean many things, from traditional linear cutscenes to the
actual gameplay experience itself. Video games are so diverse that story or narra-
tive can have an entirely different emphasis, depending on what it is used for.
An important strength of video games lies in their interactive nature, and
because interactivity and traditional narrative structures are hard to combine,
many people seem to think that this hampers the possibility that any real story
could be encompassed in a game. Many people even argue that story should be
eliminated from video games completely.
This is wrong because it is based on a misunderstanding of what story can
mean within the context of a game. Some people look at story in games through a
prism of linearity and other structures derived from film or literature, areas that
are suited to linear and passive kinds of story, instead of trying to understand the
meaning and purpose of “game-story.” But it does not have to be this way.

6
The RPG equivalent of a level.
Story and Narrative 253

Personally, I feel that the act of playing a game actively writes a story in the
mind of the player. For example, Elite has hardly has any story to speak of, except
for some background information. Through playing the game, though, a story al-
ways emerges, based on the player’s actions, immersion, and choices made while
playing the game. That said, something like Super Monkey Ball relies much less
on this, but the principle still stands. In other genres, “classic” story implementa-
tion is much more central to the act of playing the game. The Secret of Monkey
Island is a great example of this.
Throughout this book I have been arguing that we need to develop our own
voice, through the acceptance of a level design language. A major component
within this goal is to come to our own understanding of how to tell stories within
our medium, and level design is by far the most effective way of expressing this
knowledge.
We write the script, author the play, and set the stage of our game’s experience.

Interactivity and Storytelling

Through our exploration of ARGs and pen-and-paper role-playing games, we
have found that they have many things in common, and that those similarities
might exist within video games and their narrative aspects in general. If this is
the case, we are coming close to defining a number of techniques that can work
within our field. Although video games occupy a unique position within the
overall family of games, they are not the only interactive game types.
And since we have already found out that the whole principle of interactiv-
ity—of gameplay—seems to many people to form the defining obstacle to in-
game storytelling ,it may prove useful to look at interactivity and storytelling in
more detail.
David Freeman gives the following summary of the conundrum as most
people experience it:
At first it seems obvious that the player should be in charge of what oc-
curs in a game, or at least the agent of much of what takes place. But in
games that involve stories, or sequential missions, creating the feeling
that the player is in charge can become a challenge. For isn’t the player
following a pat laid down by the game designers?7
This is a prime example of the term “story” being used as a static and preor-
dained concept, as in books and in films. However, story in games is not neces-
sarily much about written story/literary connotations.

7
David Freeman, Creating Emotions in Games, New Riders, Berkeley, CA, 2003, p. 328.
254 Game Stories

In a game the story’s plot is the sum total of the player’s actions throughout
the game. Many of the themes and the subtext8 are the emotions experienced by
the player, and how they are dealt with through the filter of play.
Where many people go wrong is by trying to force these things into a lin-
early controlled and developer-constructed “story.” This misses the point about
how games work. Instead, we should craft levels in such a way that gameplay
moments form dramatic markers and narrative focal points that can be interacted
with by the player in all kinds of different ways (gameplay). If this is done well,
the actions of the player throughout the game then write their own story—one
that is incredibly compelling and worthwhile. Game tropes like challenge and
interactivity do not impede this kind of story at all, games make use of them
by making them dramatically interesting to the unfolding story of the player’s
interaction with the game.
Nonetheless, traditional snippets of story can still be used. We can have mo-
ments of exposition or talking heads in cutscenes, and so on (when appropri-
ate). However, these are only a small part of the narrative as a whole.

Agency and Story

David Freeman acknowledges the role of the player in the story and offers the
following:
By feeling your actions have consequences, you have an impact on the
game. To the degree that impact is either real or feels real, it creates the
sense that you are playing, to some degree, a self created story.9
What David is talking about is often referred to as giving the player agency in
games.

Agency
Agency is an important aspect for level designers, since it ties in directly with what
they are trying to do. We speak of it when describing the ability of an “agent”
to act in (and make an impact on) the world. This agent can be human, which
opens up the discussion to matters of choice, as well as to matters of morality
and ethics. (With choice comes responsibility.) When we look at this concept
in the context of this chapter, we can describe agency as something given to the
game player in the game. The player has the ability to act in the game world, and

8
The underlying thoughts and motivations of the character, not explicitly expressed.
9
David Freeman, Creating Emotions in Games, New Riders, Berkeley, CA, 2003, p. 328.
Story and Narrative 255

ideally his or her actions have an impact on that world and on the progression of
the game. This kind of agency is much more likely to happen to players of video
games than, for example, to readers of a book or to a cinema audience.
So what this all seems to boil down to is the notion that the player is a partic-
ipant in the unfolding game story, and that it is often desirable to give the player
the opportunity to take an active part in the content of the narrative. People
call this kind of narrative or story different things; we have seen David Freeman
refer to it as self-created story. That is a good term, but to me it avoids the fact that
level designers still have some influence over what stories can be created by the
player. I often refer to the process of telling and experiencing a game story as the
Gameplay Narrative.

“Gameplay Narrative”
It is becoming clear that in video games there are two types of narrative. There
is the type of narrative we normally associate with books, plays, and scripts,
and the type that is built as the sum of the player’s actions. This latter one is in
many ways the most important one for us to grasp, as it applies to pretty much
all games. Let’s call it Gameplay Narrative for the purpose of this book. In a
gameplay narrative it is the gamer who tells the story, through his or her own
actions.
Seen in this light, the narrative occurs in most cases during or after the
gameplay event. It is a natural result of gameplay. Because players have the ability
to act within the game world,10 they write their own narrative.
For all practical purposes at any moment, it is the player who experiences
the emotions and the emotional commitment of the character. The levels them-
selves can push and pull and steer and provide an outline and a setting for the
player to experience the game in, but in most cases it is the player to whom it
all happens, not a pre-written character in a static story arc. In fact, it can clash
terribly if the game tries to make the player the same as the character, insofar as
it tells the player how to feel. This is the opposite of using techniques to solicit
appropriate emotions.
The goal has to be to provide a chance for the player to create experiences
that form an interesting narrative.

Agency and interactivity

If the most important narrative actions come from agency, that is to say, from
the player’s actions .This gives us an excellent opportunity to create specific
10
Yep, “agency ” again.
256 Game Stories

story points to occur throughout the game’s levels. If we provide moments of

interactivity that are geared towards letting the player experience a compelling
action-narrative, we contribute greatly towards the gameplay story of the game
as a whole.
In this case, story and narrative are not necessarily classic, literary story
points. We don’t have to have interactive sequences that are only about story
exposition or similar literary devices. We are talking about meaningful actions
within the gameplay narrative.
Example: many old FPS games feature a dramatic moment when the player
finds a powerful new weapon. This is a specific and meaningful narrative mo-
ment within the sequence of actions that form a game’s narrative experience.
It is important to allow the player to make the decisions, or in other words,
to exercise player agency, in those circumstances where the resulting actions
have dramatic impact.

Temporal aspects of gameplay narrative

In all this it is crucial to realize that a game’s narrative plays out alongside the ac-
tions of the player. It does not exist until the player makes the choices that define
the plot and story that the player experiences. This is different from linear stories
like books or films, where the audience is confronted with an accomplished story.
In game levels, the player needs to be an active participant within the story,
or at least must have an influence over the telling11 of the story. In practice this
means that story events (gameplay scenarios) can occur on a fluid timeline. If
we accept this as a fact and embrace the possibilities this gives us, we are ahead
of the pack. Rather than lamenting the loss of control for designers, we can cel-
ebrate the increased dialogue and interaction with our audience.

Level design outline

To do this, the level designer is best off to create an outline of a dramatic progres-
sion, and not a strict linear plot line. Think about the kinds of experiences the
player is likely to encounter on a per-level basis, and use this to create tools for
players to paint their own picture.
If these dramatic focal points can be encountered in a non linear fashion, it
will add even more to the immersion of the player in the game world. His or her
sense of place and impact on the world will be much strengthened if encounters
happen on the player’s schedule, rather than on the designer’s.
In the player, we have an active and complicit partner in the story. This par-
ticipation is subconscious when the player reacts to what we put in his or her
11
As often occurs in classical text adventures.
Story and Narrative 257

path. It is more explicit when the player knows that certain actions are expected
or that a suspension of disbelief is required to enjoy the experience.

Emergent Stories
To a degree this makes gameplay narrative somewhat emergent. The level design-
er only provides seeds that grow in ways that are not entirely predictable. Even
better, some of these emerging stories can interact with other gameplay-provid-
ed experiences. Things that have happened to the player previously can directly
fuel a player’s current experience12 in the game. For example, a tense survival
subplot may open up because the player has been wasteful with resources in an
earlier part of the game. As level designers we can make sure there is space for
this kind of fluidity to the way the game is experienced, which gives us a more
natural approach to story in games. Rather than having a formal storyteller ap-
proach, we are better off using an approach where we sketch the boundaries of
the world, add some dramatic focal points, and give players the tools to deal with
them as they see fit. This gives the added advantage of having levels that allow
for a whole spectrum of experiences and are never exactly the same between
different players.

Dramatic Tools for Level Design

Now that we have examined the types of story and story elements that games
deal with, and the way these manifest themselves within the medium, we should
look at how we can use them as level designers.
We can take the findings from our examination of story-reliant game genres and
the nature of gameplay narrative and see how they relate to the basic dramatic ele-
ments explored in the previous section. Here they are again, as a quick reminder:
• Theme/mood:
■ form,
■ style,
■ setting.
• Dramatic progression:
■ plot,
■ pacing.
Each of these applies directly to level design, if in particular ways that may dif-
fer in execution from most other forms of entertainment and art. Nonetheless,

12
Gameplay narrative.
258 Game Stories

many of the goals are still fairly similar: we want to create levels that have some
kind of emotional impact, whether that is pure gameplay enjoyment as found in
a game like Tetris, or orchestrating a sense of deep fear in a survival horror game
like Silent Hill.
We address these topics one at a time, but I want to stress one point. All
these elements are in a codependent relationship with each other, where one
choice heavily impacts the next choice.

Theme and Mood

When we ask questions about the themes and moods of the levels we are mak-
ing, we are really asking about purpose. This is where our earlier examination
of a hierarchical approach to level design becomes really helpful. What are the
level’s external goals? What are its internal and intrinsic goals? Where does it fit
in the overall picture of the full game?
If we have followed up on the advice in earlier chapters,13 we should be able
to answer all those questions and determine what that means with regard to set-
ting, form and style, and how we can make those elements work for us to create a
dramatically interesting experience to emerge for players of the game.

Form and Style

Now that we know the “what” of the levels, we should look at the “how.” In most
cases, game levels are fairly complex. They are multidisciplinary affairs that com-
bine all kinds of media elements into a coherent whole. Nonetheless, there are
still identifiable elements of form and style that we can look at, and in some cases
there are even well-defined formats to choose from.

Form
Form elements are often determined by the genre conventions of the game in
question, as well as by the game design. Strategic team-play games may suggest
“capture the flag” levels, while racing games will be best served by certain types
of looping racing circuits. The choices of these kinds of forms are as important as
any others in level design and should be considered carefully.

Style
Within the forms there are certain types of gameplay that keep surfacing in levels
to the extent that they can be described as styles. This is reflected in some typical
13
See Chapter 3, “Level Design Goals and Hierarchies.”
Story and Narrative 259

style elements that can be used in a level. There are obviously too many to con-
sider here; the use of styles entirely depends on the kind of game you are making
levels for. But we can name a few: timed sequences, stealth sections, and run and
gun can all be incorporated in levels, either as the sole basis for gameplay or as
one of many other gameplay styles.
We can have levels that deal with mirrored player bases (to maintain bal-
ance between the teams), capture and hold scenarios, and time limited scenarios
where a team has to fulfill a task within a certain timeframe. In a survival horror
game we can have levels based on fleeing (escape an area before it is too late), or
defend to survive (being barricaded into a room while an enemy tries to get in).
There are many advantages to using well-known gameplay styles. The level
designer will have lots of information on how to implement them, the player will
have a very recognizable gameplay environment to enjoy, and the programming
teams will know what to look out for. The downside, of course, is the lack of
originality that can come with these kinds of pre-made styles. Having said that,
innovative play environments can be considered a style in their own right.

Setting
Finally, we need to address the where. The setting of a level can be one of the
defining features that determine much of the content the level will carry. It is po-
tentially one of the most important decisions that can be made, as it will lead to
certain restrictions and boundaries to the gameplay action, which in effect makes
it a practical framework for the level content. If this framework is chosen well,
the level can be very appropriate to the gameplay that the level designer wants
to implement. If it is chosen poorly, it can become a huge obstacle to the desired
gameplay. So let’s look at some of the things to consider in choosing a setting.
When choosing a setting, we should check against a number of things that
should be on our wish list for levels:
Depth and versatility—Can the setting sustain the level design goals it needs
to support for a long enough period without becoming boring or too samey?
Does it allow for flexible and varying application of gameplay mechanics?
Take for example a level of a real-time strategy game, one where terrain influ-
ences the troop movements a lot. It may be more interesting to place the action
in an old European city built on two flanking hills than on a large homogenous
and perfectly gridded city, as seen in many American states.
AI—Does the setting allow for the AI to shine? Does it offer the right en-
vironment for path planning, showing off their best decision making skills, al-
low for good interaction with the player, and whatever else can be thought of to
make them as effective as possible.
260 Game Stories

Player Character Abilities—Similarly, does the environment jell with the

player character’s ability to navigate and interact with it? Is it a fiddly environ-
ment with many obstacles? Is it easy for the player to get lost or get stuck looking
for exits? Is it easy to signpost in the environment?
Appropriateness—Does the setting fit the desired mood and themes you
want to establish? Does it jell with the plot and the background story? Will it
show off the rendering capability of the engine and the art style the game is sup-
posed to convey?
There are hundreds of additional factors that we can add to this; it is all
about finding the right setting for the level design elements you have in mind.
For example, many levels take place in underground bases for very understand-
able reasons: they easily support corridor/room/corridor settings, are suitable
to having lock-and-door gameplay, can be very easily provided with navigation
direction in the environment, and so on. The trick is of course to tick off many
positives in settings that feel fresh and original.

Dramatic Progression
If we do our work well, levels should provide some form of dramatic progres-
sion. We can affect this by a number of methods, including the use of plotting
and pacing and their level design equivalents.

Plot
In level design the plot is the sum total of the player’s actions. This means that we
don’t know the entire plot when we design a new level, since the player’s actions
have not yet occurred. An immediate natural reaction of many level designers
is to start orchestrating the content of a level in such a way that we have total
control over the player’s actions and therefore know exactly what the plot of the
game will be like. This is a grave mistake. While throughout the book I advocate
ways of exerting authorial control over the levels and the experiences therein, I
do not advocate spelling out the exact steps to be taken by the player in order to
have the “right experience.”

Agency and gameplay narrative

To make sure that players feel involved in the plot of the gameplay they are expe-
riencing, we should avoid explicitly dictating all the experiences the player has
to undergo. Instead, we design our levels in such a way that we give players real
agency in the game, and we let their own choices become the focus for most of
Story and Narrative 261

the plot progression. What we want to do is give players the feeling that they are
the ones that matter in the world, that their actions determine the outcome of
the narrative.
A player’s self-determination is to be encouraged though creating environ-
ments that react favorably to the player’s actions. It is to be encouraged to let
players try out different abilities, award them when they try different things, and
not punish them for exercising those skills that they have been taught.

Dramatic markers and set pieces

Even within such a free level design environment, we can still shape the outlines
of important scenes and story elements. We will want to orchestrate memorable
moments and at times influence the pacing of the gameplay narrative. To achieve
these things without giving up on the sense of player agency or limiting the ex-
pression possibilities of the player we can make use of dramatic markers and set
pieces.
An important thing to remember is that we can create these kinds of focused
and important gameplay narratives without necessarily prescribing the exact
time or method in which the player experiences them. If, for example, we wish
the player to be exposed to a very important encounter with an NPC, we could
make this encounter occur based on other conditional level elements. That is
to say that the NPC encounter can be triggered by a number of different condi-
tions, especially those that are subject to the player’s actions. This guarantees
that the player cannot predict when the event occurs, which makes for a dramati-
cally more interesting world.

Pacing
As we have seen in the preceding chapter, pacing is an authorial tool of great pow-
er in most non-static art forms and creative endeavors. As level designers we find
ourselves in exactly the same position as composers or film editors, inasmuch as
we want to exert control over the timing and intensity of the gameplay experi-
ence we are creating. We may use a different vocabulary, but the overall aim is the
same. However, there are elements to pacing that are specific to level design and
need to be further examined.

Pacing and flow theory

One of the most important elements of pacing in level design is that of keep-
ing players in the zone, that perfect state where immersion and enjoyment and
other positive experiences keep players in a state of gaming bliss. As discussed in
262 Game Stories

Chapter 8, “Immersion”), keeping players in the zone has wide-ranging effects,

not the least of which is that it has an impact on pacing.
This is also borne out if we look at our flow diagram again. We can see that
Csiksczentmihalyi’s diagram14 progresses along a timeline, which brings us into
the realm of deliberate pacing. As levels progress, the designer needs to make
sure the right balance between challenge and ability is maintained along a learn-
ing curve that develops over time. Since we want levels that don’t shut out or
bore players, we have to take great care. It is easy to see why this is important to
level designers, since they are the ones who control the pacing of levels.

Learning curve
It is then fair to say that closely related to a level designer’s task of keeping a
player in the zone is the need to maintain a sensible learning curve. As the flow
diagram shows, it is vital not to over- or under-challenge the player. Under-chal-
lenging is bad because players may become bored, but at least in that situation
players may still try to entertain themselves. Over-challenging the player is even
worse, as this will make the player so frustrated that he or she may stop playing
altogether.
So it is vital that the game’s learning curve15 be a function of a progression
that keeps the player within the flow channel, or in the zone of an optimal game
experience. But how do we do that? Players’ individual skill sets can be wildly
divergent, making it very hard to cater to individual abilities.
One way of achieving this goal is by allowing aspects of the game to behave
dynamically in concert with the player’s actions.

Dynamic progression
Keeping a balance between challenge and skill is easier said than done. Since we
are dealing with human beings, we are also dealing with wildly different abilities
and skill levels. One of the biggest dangers in level design is to create an experi-
ence that is suitable to one player, but that creates an insurmountable obstacle to
another. If the latter occurs, it is not unlikely that the player will get so annoyed
with the game that he or she will stop playing it forever. Even worse, that person
tells friends not to buy it. This is pretty much the worst thing that can happen
and needs to be avoided if at all possible.
14
See “An Introduction to Flow Theory ” in Chapter 8.
15
Learning curve is actually a misnomer, since the learning aspect needs to stay at an
enjoyable constant. A better term would be something like challenge curve or progression
curve.
Story and Narrative 263

To counter these massive dangers, designers and game developers can cre-
ate levels and games that employ dynamic difficulty adjustment. The game experi-
ence is tailored to the varying skill levels and play styles of its players. This is a
technique that has been known for a few years, but that wasn’t studied in detail
until Jenova Chen did the game development community a great favor by writ-
ing his thesis on the subject.16 Chapter 16, “Challenge,” goes into great detail on
this matter.

Typical Pacing Methods

Because pacing has such a strong impact on how games are perceived, I want to
spend some extra time on it. Looking at the pacing elements of past video games,
we will discover a massive amount of data available to us about pacing methods.
We can easily distill a number of typical techniques and concepts that have be-
come part of the level design toolbox:

Connectivity and Layout

The most obvious way of influencing pace is simply by creating gameplay spaces
with navigational limitations and opportunities. This allows the level designer
tight control over the progression of the player, and over how and when specific
gameplay situations are experienced. This makes it an often-used tool by level
designers in all kinds of situations. Let’s look at some useful examples.

Bounce
We speak of bounce when the player encounters an impediment to progression
that forces him or her to backtrack and perform an action that removes the ob-
stacle when revisited. In the early days of level design, and indeed to this day, this
was often realized by the locked-door scenario. The player would literally find a
locked door and would have to go back and hunt or a key.
These days, we often find more elegant approaches to the bounce scenario,
at times implemented so well that players don’t even recognize it as such.

Linear progression
Sometimes a straight path from A to B is all we need from a level. Linearity has
some clear advantages, and if not overused can be a very effective method of
controlling the pacing in a level. It gives the player a great sense of clarity as prog-
16
Jenova Chen, Flow in Games, http://www.jenovachen.com/flowingames/index.html,
2008.
264 Game Stories

ress is along the one path. This kind of direct challenge can be very enjoyable,
especially if it is clear to players that in order to succeed or progress they must
finish an “obstacle course” of gameplay scenarios. Pacing is controlled by the
gameplay along the way. It is very easy for the designer to create and test such a
formal environment, which makes it an attractive proposition from a production
point of view.

Loops
A loop is a very similar method of controlling pacing , with the difference being
that players end up at the starting point. All the other linear advantages apply,
but we have the additional bonus of being able to show players their goal from
the get-go. This can greatly enhance the sense of context in which players per-
form actions.

Sandbox
A sandbox approach allows players to come up with their own gameplay solu-
tions, or at least must create the illusion that they can. The central point to sand-
box games is that players must feel that the world doesn’t feel too prescribed.
Instead, players must be able to bend it to their own will. The sandbox provides
players the raw building material (like sand) to create their own mark on the
world. It isn’t always easy to influence pacing in such an environment, although
much can be done by placing certain conditional events in the world, and by
creating natural-looking bottlenecks. Perhaps a player character needs to be of
certain strength to pass a particular area, or a specific type of enemy creature is
only released into the environment after the player obtains a particular weapon

Stickiness
A good way of subtly influencing pacing is to slow the player down by creating
areas that are sticky. This stickiness is achieved by providing positive reasons for
staying in an area, for example by providing a safe area in which the player can
indulge in some fun activity or by creating a tempting discreet challenge that
may yield a great reward. If these instances of stickiness are optional, players will
never feel resentful about this kind of manipulation, since they are engaging with
it by choice.
Many good examples of sticky areas can be found in Half Life 2, where all
throughout the game completely optional areas can be found, filled with op-
portunities for almost carefree play. Some of these are almost banal in imple-
mentation. I have once stopped my progression on the main path after spotting a
Story and Narrative 265

large number of crates and decided it worth my time to try to stack them as high
as possible in order to gain an enjoyable view of the world. (The game allows
a playful approach to physics.) From a level-design point of view, all that was
needed to give the player this experience was a deep understanding of the fun of
the game’s physics engine and the placement of some crates in an eye-catching
area.

Push and Pull

Push and pull mechanics are born from elements in the environment that seek
to influence the player’s progression, either forcibly or by necessity. Take for ex-
ample the scenario where in order for players to progress, they need to go down
a lift, but after they have reached the desired location, the lift breaks down and
makes backtracking impossible. This is a clear example of a push. Now imagine
an alternative scenario where the player is nearly out of health and being chased
by a strong enemy creature. Somewhere in front of the player, a health-pack can
be spotted. Chances are the player is drawn to the health-pack. This is a clear
example of pull.
If we take these principles further, we realize that a game level is full of op-
portunities for pacing adjustments by pushing here and pulling there. If we do
this with skill, it can come across as perfectly natural. The level can become a
gameplay engine, and items placed among many other things can be used as
fuel.

Alternative Methods and Ideas

There are many other ways in which to influence the pacing in a level. Some are
yet to be invented, and others are fairly obvious or well known. It is pointless to
try to capture them all in this chapter, but it may be useful to at least make some
notes on possible areas of interest:
• fear (scare players into adjusting their pace of progression),
• hurry (create a necessity to rush),
• time (affect the progression of time to create pacing changes),
• physics (make the properties of the environment determine progress),
• AI (Let the player depend on the progression of a NPC).
Many others can be added, and in fact it may be wise to create a list of useful
ones for the game you are working on. It is always nice to have a large repertoire
of techniques. Please check the upcoming practical examples for a whole raft of
them.
266 Game Stories

Puzzles
A special mention needs to be reserved for the use of puzzles as pacing devises.
Puzzles are among the best and most versatile tools that level designers have at their
proposal. Among their many uses is their ability to influence the pacing in a level.
Because puzzles can be designed to fit a certain solution, they are extremely
useful in exerting control over a gameplay/puzzle situation.17 If we can deter-
mine the likely duration of a puzzle sequence, we are in effect given control over
the pacing of a level, or at least we have it to a significant degree.

Space
Davis understood that the space between the notes was sometimes just
as important as the notes themselves. But he also understood that too
much space between the notes, and people would think that the concert
was over and go home....18
In level design, pacing is as much determined by what you leave out of a level as
by what you put in. Almost every professional level designer will at one point
have a serious disagreement with somebody about level areas being “too empty.”
In almost every case when I have had this argument in my own professional life, I
was right to fight my corner. The spaces between the gameplay events are incred-
ibly important, and in fact, are part of the gameplay experience as a whole. Mark
Hollis has been reported19 to have said that:
Before you play two notes, learn how to play one note—and don’t play
one note unless you’ve got a reason to play it.
Not only are these wise words, they can be expanded on to fit video game develop-
ment even further. Sometimes leaving empty spaces is a level design event in its own right.
There is need in level design for spaces where nothing happens. Don’t listen
to producers or other meddling colleagues or bosses who want to have some-
thing happen every two seconds without rhyme or reason. They don’t neces-
sarily understand that the full picture has to incorporate room for the player to
breathe. Allow for reflection and space within the player’s head to do things like
think, observe, contemplate, and strategize.

17
For all kinds of additional ways to include puzzles in level design, please look at Chap-
ter 14 on Puzzles.
18
From the Genius Guide to Jazz, July 2001, cited at http://www.allaboutjazz.com/php/
jazzquotes.php.
19
Quotation from the unofficial Mark Hollis and Talk Talk website: http://users.cyber-
city.dk/~bcc11425/.
Story and Narrative 267

These things are as vital to the action in the level as the action-packed mo-
ments are. And last but not least: they provide a powerful tool to affect pacing.

Theory Summary: Tying It All Together

This chapter shows that there are several ways to think about story and narrative
processes in level design.
Despite what many people claim, there is much storytelling involved in level
(game) design, but it is focused on providing the player with the right setting and
tools to experience the game in a dramatically interesting way. The level designer’s
job is to create a level in such a way that it forms a compelling narrative of the
player’s interactions within it. It is a self-created story or a gameplay narrative.
This type of story is different from stories in books and from other writ-
ten stories, which are different from film stories, which are different from other
stories yet again. Gameplay narratives are the sum of the player’s actions and
experiences and are subject to their own rules and techniques with regard to
dramatic progression.
It is also a misconception that typical game elements like interactivity or
challenge are unique to video games. Other game genres like ARGs (Artificial
Reality Games) and role-playing games have to deal with the same obstacles and
are well worth studying in that respect.
With that focus in mind, we still have similar goals and tools to other me-
dia that employ story. We deal with overarching themes and use tools to enable
dramatic progression. It is the translation of those goals into level-design-specific
tools and methods that can make the difference between levels that provide a
compelling gameplay story and levels that are hampered by traditional or non-
applicable notions of what “story” is supposed to be. The next chapter will show
a number of practical examples that showcase how this can be done.

Practice
Example 12.1: Themes and Moods—
Practical Application
Summary
To enhance the game’s themes and moods, we create levels that support the-
matic gameplay. This can be done through the use of symbols: content that rep-
268 Game Stories

resents more than its literal interpretation. What does this mean in a practical
sense, however? This practice example shows straightforward ways to tie themes,
moods and symbolism into everyday level design situations.

Game Genre
The technique is suitable for most games.

Goals to Achieve
• Identify main themes and moods.
• Assign thematic and symbolic value to gameplay scenarios.
• Use this to underline the themes and moods the level needs to convey.

Description
(Example type: Original)
We generally try to make our levels about something, even if that something is
solely the gameplay mood that pervades a level. In most cases it makes sense to
emphasize certain themes and moods that the level needs to convey, and use this
as a framework for the gameplay that occurs within the level.
We can do this by taking general gameplay that would occur regardless of
thematic goals, and making it specific so that it ties into the required themes or
moods we want to portray.
If we take a hypothetical set of themes and moods, we can show how this
could work.
Powering the town. Our hypothetical game is a third person action adventure,
where the player can traverse and explore the environment and perform simple
context-sensitive actions: throw switches, push objects, and so on.
In this example the main themes we want to put forward are based on the
universal struggle for freedom, and the mood we want to portray in the level is
one of hope and positivity.
We can achieve our goal by providing an environment that is restrictive in
a number of ways at the outset, including the player’s ability to act within that
environment, and slowly transforming it through the actions of the player to be-
come one of wide-ranging freedom and promise.
Take, for example, an abandoned town filled with ancient machinery, doors,
lights, vehicles, and so on. All of these used to be powered by electricity derived
from wind and solar energy. The original inhabitants of the town have abandoned
it after an earthquake damaged all the sources of power, like the windmills and the
solar panels. They now live a hand-to-mouth existence in the local countryside.
Story and Narrative 269

To fit with the theme and mood of the level, we will allow the player to
perform a number of context-sensitive actions that slowly transform the town,
enable things in it, increase the player’s ability to move around and interact with
it, and lure back the original inhabitants.
• To begin with, the player can rotate a windmill back into the path of the wind.
This will allow a system of aqueducts and pipes to start to power the town’s
running water system.
• This then enables the player to go around town and clean the dirty solar
panels with the water pipes now running again.
• The new electricity from the solar panels allows the player to repair the
broken generators with electric tools that can now run again.
• With the generators running, the player can charge the battery of an aircraft,
which is able to reach previously unavailable solar mirrors.
• The player then repositions the solar mirrors to make them shine light on the
town’s massive central solar panels.
• This finally allows the player to kick-start the whole electricity net, enabling
the town’s streetcars, residential equipment, and so on.
All the way through this process, a number of things occur that are in keeping
with the level’s thematic goals:
• The player can navigate more and more of the town.
• The town becomes bright with all the new sunlight and electrical illumination.
• Life starts to return to the town: villagers start to occupy the houses again.
• Music is heard throughout the town.

Further Notes
An approach like this one allows the level designer to create interesting and ap-
propriate gameplay sequences all through a level. It is more a matter of choosing
the action that supports the general moods and themes than of having to pains-
takingly come up with a large number of individually interesting but incoherent
gameplay actions.

Example 12.2: Non-Linear Narrative—

Dramatic Components
Summary
There are times in which we like to avoid linearity in level design and still make
use of a progressing narrative. This example shows a way of approaching the
problem that allows for this to happen.
270 Game Stories

Game Genre
This technique can be used in any game where a dramatic progression does not
have to follow a set order.

Goals to Achieve
• Allow for a non-linear but still progressive narrative in a level.
• Let the gameplay narrative be assembled by the player.

Description
(Example type: Original)
The dramatic components technique assumes that the player is able to form a
narrative out of individual components that are presented in an arbitrary order,
or in an order determined by player choice. If enough components have been
assembled, the narrative structure is complete.
One way to do this is by treating the narrative as a meal that is made up of
specific ingredients. The level designer provides the meal’s ingredients, and the
player is the cook who is putting it all together through gameplay actions and
choices. It is key to accept that the player takes individual narrative ingredients
and slowly constructs a whole in their mind regardless of the order in which the
ingredients are discovered.
Another way of looking at this is by imaging a criminal case that can be solved
if enough evidence is found. The level designer spreads the evidence throughout
the level, and it really doesn’t matter in which order the player uncovers it. The
final picture of the narrative is the same regardless.
Whatever analogy we employ, the idea is to break up the narrative—be it fac-
tual content, dramatic sequences, set pieces, or any other components—and break
it down into discreet individual instances that can be placed throughout a level.
We can envision dramatic components like the following arsonist scenario:
• The player rescues a child from a burning building.
• The player discovers that several buildings have burned down in the past.
• A letter reaches the player, telling him to deliver $100,000 or another child
will die.
• The Player finds a diary of a psychiatrist, mentioning the case of a deluded
arsonist.
Each of these components can be experienced by the player in any order, with-
out harming the narrative at all. So, what we end up with is a coherent narrative
whole, but it is presented in a way that respects the non-linear interactivity that
Story and Narrative 271

can be so valuable in our levels. It is the player who pieces the individual parts
together, and it is the player who will feel as if he or she plays a vital part in this
narrative.

Further Notes
If necessary the level designer could tie it all up by allowing the next sequence in
the level design to unfold only after these four events have happened. This would
turn it into a semi-linear progression, based on conditional events. In that sce-
nario, some control is applied, while still allowing for the dramatic component
technique to be used.

Example 12.3: Pacing Techniques—Bounce

Summary
As defined earlier in the book, we speak of bounce when the player encounters an
impediment to progression that forces him or her to backtrack and perform an
action that removes the obstacle when revisited.

Game Genre
This technique is suitable for most games that feature physical progression
through a level.

Goals to Achieve
• Control pacing by introducing a bounce scenario.
• Make this stand out from the preceding action.
• Present this in a way that is appropriate to the setting.

Description
(Example type: Original)
In this example, the player is traveling by car along a path through the level. The
player is progressing along this main path when the tire blows, and he or she is
subsequently forced to stop. This is the event that triggers the bounce situation
(E in Figure 12.1).
The remainder of the main path is of such a nature that progression on foot
is not viable, so the player has to find a solution to the problem. The only path
accessible on foot is the bounce path that takes the player past a number of chal-
272 Game Stories

Figure 12.1. Bounce scenario.

lenges (C). These challenges are up to the level designer, but should be of a na-
ture that significantly changes the pacing of the gameplay.
Eventually the player reaches the solution (S), which in this case is a spare
tire found in a room off of the bounce path. The player now bounces back to the
position where the vehicle was abandoned, and is confronted with more chal-
lenges along the way.
Finally the player returns to the car and is able to progress along the main
path.

Further Notes
In practice this is no more that a lock-and-key situation: the lock is the broken
car, the key is the spare tire. The situation can be embellished without difficul-
ties, however: the path can be more complex, the bounce event can consist of a
number of sub-events, and so on.
Designing Gameplay V

I n a book covering a field as encompassing as level design, it is very hard to or-

ganize data in a way that is not too contrived or constraining. There is no ideal

Charles F. Foley and Nell W. Rabens, “Apparatus for Playing a Game wherein the Player
1

Constitute the Game Pieces,” US Patent 3,454,279, issued July 8, 1969. Available at “The
Twister History,” http://www.mathematik.uni-bielefeld.de/~sillke/Twister/history/,
2000.
273
274 Designing Gameplay

way of doing it, and this book is no exception. I have done my best to organize
chapters in a sensible manner, but some explanation is needed.
The following set of chapters is organized under the heading of Designing
Gameplay. This is because the chapters cover areas that are most directly con-
nected to gameplay mechanics and their implementation. They are the bread
and butter of level design and often work as part of larger gameplay systems. Ul-
timately, this will be true to a certain degree of most topics covered in this book,
but it is most true for the chapters explored in this section.
World Building
13
A n interesting aspect of what we do in level design is that we create imagi-
nary worlds for players to enjoy. This is an aspect that isn’t unique to video
games of course; the same thing occurs in books and theater and in all kinds of
creative disciplines. However, we do have a heightened emphasis in our imagi-
nary worlds on player input and participation, perhaps more so than in other
artistic endeavors.
Either way, no matter which way you look at it, the act of level design is
partly an act of world building. We create a headspace for players to spend a lot of
time in. Since these worlds are virtual rather than of this world, normal rules don’t
apply. We need to take a good look at how to approach world building through
that filter.

Concept

Terminology
I want to spend a small moment discussing terminology, just to make sure I don’t
create the wrong expectation for these chapters. In a subject as wide ranging as
this one, it is very easy to lose focus and get tangled up in definition problems, so
I want to avoid that trap completely. The term world building can be used in many
more ways than I am going to discuss in this book, and although they may all be
interesting to look at, I am limiting myself to those aspects that I find relevant to
the subject matter we are dealing with directly.
The way world building is used in this book is generally not a metaphorical
one. I think it is completely valid in other instances, like the worlds that can be
conjured through the notes played in an evocative piece of music or other such
275
276 Designing Gameplay

examples, but it is not really what we deal with in level design. I am mostly fo-
cusing on the actual creation or recreation or simulation of actual or possible
worlds. Sometimes this is as complete and grand scale an effort as can be seen as
in Lord of the Rings. At other times, it is a small aspect of a world, as in a theatrical
play that takes place in one set. These smaller-scale examples may themselves act
as metaphors for other things, but they still employ the art of world building in a
direct and recognizable manner.

Some Examples of World Building

It often pays to look at a subject like this through the various filters provided
by the approaches taken by other disciplines. Various points of view are likely
to yield more diverse insights, which is very helpful, while overlap between
diverse art forms suggests commonalities that can be forged into general tech-
niques.
So let’s look at some typical examples of the art of world building.

Science Fiction and Fantasy Literature

It is hard to think of a clearer example of world building than the many efforts
found in science fiction and fantasy literature, and in related genres. This in-
cludes utopian or dystopian books like Brave New World1 or any other book set
in a world that isn’t our own. Many of these books are meant to invoke a sense
of wonder in the audience. This term specifically describes a feeling of wondrous
realization that is carefully built up through world building techniques.
Good science fiction or fantasy does not engage in world building simply
to create a background setting. The point of the exercise is to create a world
that supports the main themes of the work in an integral way. That is to say that
the world-building aspects are not incidental or separate from the core of the
work.
We can say without a doubt that the detail in the world described in George
Orwell’s 1984 matters greatly to the impact of the whole. The posters stating “Big
Brother is watching you” are so apt and appropriate to the themes of the book
that they have touched a real nerve in people who have read the book. So much
so that the iconic image of Big Brother has been adopted across the cultural land-
scape. Concepts like these can closely mirror the realities of our own society, but
by presenting them through the lens of science fiction or fantasy, they can be
viewed in a way that gives us a new perspective.
1
Aldous Huxley.
World Building 277

Equally interesting are the efforts to evoke an imaginary world that stands
up to scrutiny and has the power to immerse and delight, simply by being con-
ceptualized in the text. Or to put it in less flowery terms: it is really nice to read
stories set in imaginary or fantastical places, if these places are realized with real
skill and thought. Many readers have lost themselves in the nooks and crannies
of Mervyn Peake’s Gormenghast or the haunting places of Ray Bradbury’s Mar-
tian Chronicles.
It is very educational to see what kinds of choices are made in this field when
it comes to describing what these worlds actually contain. What is put into the
text that makes it so successful?

Detail and symbols

The examples given above (the iconic concept of 1984’s Big Brother or the fine
descriptive detail in Gormenghast) show two sides of the same coin.
Regarding detailed meticulous descriptions, we have this:
Over their irregular roofs would fall throughout the seasons, the shad-
ows of time-eaten buttresses, of broken and lofty turrets, and, most
enormous of all, the shadow of the Tower of Flints. This tower, patched
unevenly with black ivy, arose like a mutilated finger from among the
fists of knuckled masonry and pointed blasphemously at heaven. At
night the owls made of it an echoing throat; by day it stood voiceless
and cast its long shadow.2

And regarding the creation of iconic symbols, we have something like Fig-
ure 13.1.
Things like these two examples all function to achieve a similar end: using
world building to express the work’s themes and ideas.
This high level of detail is often needed because the reader is dealing with
unfamiliar territory. Sci-fi and fantasy deal with unknown places, and the re-
quired sense of wonder is hard to achieve without creating a very precise picture
in the mind of the reader.
The use of symbols and icons, though not specific to sci-fi and fantasy, is
especially useful in these genres, since it becomes a shorthand for some concepts
and ideas described earlier. The author cannot keep expressing every bit of detail
in this constructed world. Some of the world’s content needs to be able to func-
tion on the iconic or symbolic level.

2
Description of The Tower of Flints from Titus Groan, available at http://www.goodreads.
com/author/quotes/22018.Mervyn_Peake.
278 Designing Gameplay

Figure 13.1. Book cover for 1984.3

Simulations
While the previous example of world building deals with creating new and
imaginary worlds, we also have world-building examples that deal with recreat-
ing worlds or environments as simulations. In this case there is no direct artistic
goal, but an educational or scientific one. In simulated environments, certain
characteristics and processes are examined within the context of a specific for-
mal system. This can be an ecosystem or a weather prediction model or a geo-
logical process simulation. The principle of these kinds of simulations can scale
up all the way to simulations of the universe itself, and can therefore be seen as a
subclass of world building.
Again, because it is impossible to incorporate all aspects of the world into
a created simulation, the makers of these simulations have to make choices on
what to include and what to leave out.

3
Newly commissioned book cover for Penguin Books by artist, Shepard Fairey. George
Orwell, 1984, Penguin Books, London, 2008.
World Building 279

In simulations, what matters most are those factors that determine the accu-
racy and predictability of the data they produce. So for example, when studying
cloud formation above oceans through a simulation, we do not have to include
data on the mating rituals of the fish beneath the waves. The difficulty lies in
deciding what data is pertinent and what data is superfluous, which isn’t always
clear.4. Difficult or not, however, in this kind of world building relevance is a key
concept.

Cinematography
In cinematography we deal with an art form that has experimented extensively
with ways of suggesting and showing a vast array of possible worlds. Few art
forms have become so adept at mixing techniques and methods to portray
imagined worlds, showing detail when needed, yet bamboozling the audience
and suggesting all kinds of implied (but nonexistent) content with smoke and
mirrors.

Midgets and magic

One of my favorite examples of this is the final shot in the classic movie Casa-
blanca. If I were to tell you that the famous final scene is filled with midgets, you
wouldn’t believe me, right?
Here is what Snopes5 has to say on the subject:
That final scene, in which Rick Blaine (Humphrey Bogart) and Ilsa
Lund (Ingrid Bergman) say their final goodbyes as the plane to Lisbon
(the one which will carry Ilsa out of Rick’s life forever) warms up in the
background is one of the most famous scenes in cinematic history, and
this is the scene Disney recreated in their Great Movie Ride, for which
they needed a real Lockheed Electra 12A.
However, when Warner Bros. filmed the scene, they found their sound-
stage was too small to accommodate a real airplane, so the studio’s prop
men constructed half- and quarter-size models of a Lockheed Electra
12A out of plywood and balsa. Midgets garbed in jumpsuits were hired to
move in and around the replica planes to camouflage their smaller-than-
life scale, and the whole scene was swathed in machine-generated fog to
further conceal the artificiality of the setting. (Fog certainly wasn’t add-

4
Especially when dealing with chaotic systems in which small factors can have large
consequences. But let’s not go into chaos theory right now.
5
Snopes: the famous myth-busting website. See www.snopes.com.
280 Designing Gameplay

ed for any touch of realism, as foggy nights are a rarity in Casablanca.)6

(Emphasis mine.)

This is a clever use of forced perspective, which is also used to great effect in the
Lord of the Rings movies7 to show the hobbits as being smaller than humans. In
some ways this can be described as a special effect, and to a degree that is all it
is. But on another level, it is way more than that, especially because the audience
doesn’t interact with the scene on that level. Viewers experience the available
information, onscreen visuals and audio, and construct the world in their minds,
following the blueprint of the director and cinematographer. When it comes to
world building in film, this kind of direction occurs all the time. It happens in the
audio cues given to the audience, the cutting of consecutive shots, the framing of
the scene, the content of the dialogue, and so forth.

Theater
Theater is an interesting art form when it comes to world building, since it is
so limited in its available space and time. All action takes place on one stage
(in most cases), and since performances are by definition live, there is only
limited scope for set changes. This is very different from film, for example,
where many locations can be used and meticulously prepared for the best pos-
sible shots.
In most theater productions it is necessary to work with broad brushstrokes.
Is the setting a temple in ancient Greece? One or two foam Greek pillars will suf-
fice. Does the action occur underwater? A blue theater light with some waving
cloth will do the trick. French front in World War II? Use an evocative sound-
track of battle noises and French and German soldiers shouting at each other.
What is key is that theater set designers are able to evoke a whole world with
very limited means, not only because there is only so much that fits on stage or
because set changes need to be quick and simple, but also because it is just more
efficient. Why build a whole temple if the one pillar will do?
Theater set-designers are very good at finding iconic visuals and sound to
transport the audience to where they need to be. The dialogue and acting are
of course of the utmost importance, but intelligent use of props, sounds, and
lighting really places everything in a fully realized world. They serve as symbols
that represent far bigger concepts and ideas, places, and emotions in a way that is

6
“The Plane Truth,” Snopes.com: Casablanca Plane in Great Movie Ride, http://www.
snopes.com/disney/parks/casablanca.asp, 2007.
7
New Line Cinema.
World Building 281

elegant and powerful. The ability to quickly sketch out the core of what matters
is a valuable technique that benefits all kinds of artistic projects.

Comics and Sequential Art

Finally, and often mistakenly seen as least, there is world building as it occurs in
sequential art. Sequential art is a term that almost nobody knows and even fewer
people use, but it is in fact a very useful collective term for things like comics,
strips, graphic novels, and so on.
Scott McCloud’s book, Making Comics, features a chapter on world building
that opens with the following words:
The world is a BIG PLACE.
Capturing all of its sights and challenges in little rectangles can be a
daunting CHALLENGE.
But with knowledge –
Effort –
And a willingness to go beyond the merely “adequate” –
– your worlds can seem as POWERFUL and VIVID as any others, real
or imagined.
Sweating such details can make the difference between drawing a page
in SIX hours and drawing it in TWENTY—
– but for your readers it can make the difference –
– between KNOWING where your story takes place –
– and BEING THERE.8

Even without seeing the images that he accompanies these words with, these
words very effectively describe many of the feelings I have towards level
design.
The almost unthinking dismissal of sequential art as a non-serious art form,
or even less than an art form, is tragic and wrong. Tragic because it denies people
an incredibly rich and dynamic source of worthwhile experiences, and wrong
because the form is capable of so much that it does not deserve to be seen in this
negative light.9
One example of the artistry and power of sequential art comes from its ap-
proach to world building, and especially its expert use of a technique or concept
known as closure.

8
Scott McCloud, Making Comics, Harper Perennial, New York, 2006, pp. 158–159.
9
Just like video games, really.
282 Designing Gameplay

Filling in the Blanks or “Closure”

In Scott McCloud’s seminal book Understanding Comics: The Invisible Art, we
can find the following passage:
Here in the limbo of the gutter, human imagination takes two sepa-
rate images and transforms them into a single idea. “Nothing is seen
between the two panels, but experience tells you something must be
there.” Comics panels fracture both time and space, offering a jagged,
staccato rhythm of unconnected moments. But closure allows us to
connect these moments and mentally construct a continuous, unified
reality. “If visual iconography is the vocabulary of comics, closure is the
grammar. And since our definition of comics hinges on the arrange-
ment of elements—then, in a very real sense, comics is closure. (Em-
phasis removed.)10
This should really be read in context with the images and preceding chapters of
the book, but I wanted to highlight it nonetheless, as it points at an important
concept that we use throughout this book in one way or another.
What McCloud is talking about is what happens in between the panels of
a comic book. Something almost never thought about, but almost magical in
nature. The human brain has an enormous capacity to see patterns, recognize re-
lationships between concepts, and extrapolate detail from those relationships.
Imagine three image descriptions presented as the panels of a comic strip11:
Image 1:
A rude young man throws a banana peel on the sidewalk. A finger-wag-
ging old man looks on disapprovingly.
Image 2:
The youth is walking elsewhere now.
Image 3:
A loud sound startles him and he runs off.
Image 4:
In his rush, he fails to see a banana peel, slips on it and falls on his be-
hind. He is laughed at by the old man we saw earlier, now holding a horn
and a peeled banana.
This is a very banal example of a sequence of events that could occur in a news-
paper comic strip. Yet if we examine it a bit closer, something extraordinary has
10
Scott McCloud, Understanding Comics, Harper Perennial, New York, 1994, pp. 66–
67.
11
Be warned, it is a terribly bad one I just made up.
World Building 283

occurred. Somehow the reader has constructed a coherent world view in which
four drawings placed side by side tell a story and give detailed information on
the world in which it occurs.
For example, we know that a little humorous revenge story has unfolded.
We know that the old man got his revenge by scaring the youth with the horn
and placing a banana peel in his path to enact some poetic justice.
All these realizations and connections occur in between the panels, which is
why McCloud has dubbed comics the “invisible art.” Our brain takes informa-
tion from the world (in this case the panels from a comic strip) and reconciles
them by creating cross-connections between the sequential images, thereby
adding meaning and relationships to the events depicted.
The process of resolving these images into a narrative or conceptual struc-
ture that we feel we can understand is called closure. It is one of the most im-
portant processes that an artist needs to understand. If employed well, closure
allows us to do much more with much less, and we can let all the hard work of
building a world in the mind of an audience be done by the audience itself.

Connecting the Dots

The mind needs VERY little to construct a view of the world that it can under-
stand. We discussed this process previously, in the discussion of the physiology
of sight in Chapter 10, but it is important to return to the concept for a moment.
Let’s recap what I said earlier:
Our brains have evolved in such a way that they have become very ad-
ept at processing data in ways that are important in our daily lives. Of
these skills, pattern recognition stands out particularly well, since we use
it so much. How else do we differentiate between all the differing visual
input we receive via our eyes? Pattern recognition allows us to process
visual data at a much-enhanced speed because it lets us carve up the vi-
sual input into useful chunks to which we ascribe further meaning and
behavior. This process is actually called chunking12 and is key to our intel-
ligent understanding of our environment. (Emphasis added.)
Previously I was speaking of pattern recognition and the ability to differentiate
between all the data presented to us specifically, but this also applies to the con-
cept of closure. Because we cannot help trying to make sense of the data we are
presented with, we more often than not try to organize what we see into a world
view. This can be done with very little data indeed, and it provides us with enor-
mous artistic firepower. Our brains are always engaged in this process, whether
12
I kid you not!
284 Designing Gameplay

we want them to be not; this means that we can tap into this process at any time
we need to.

Commonalities and Lessons Learned

There are a number of parallels and commonalities in the previous examples, as
well as some genre-specific insights. They offer an insight into the kind of general
techniques one can exploit, or specific solutions found in various world building
exercises.
We have seen that very specific use of fine detail can be made when needed,
especially in fields like Sci-fi and fantasy, where the artist has to paint a picture of
many unknown things. But there is also a general need for iconic and symbolic
content to convey important themes and content efficiently. At times there is an
emphasis on relevance, as demanded in simulations, and we can use the concept
of closure to our advantage in almost all world building scenarios.

Theory
To a certain degree much of what we do as level designers constitutes world
building. What we build is a virtual world, a gameplay world, which players
hopefully will spend much of their time in. The least we can do is do our best
to make their stay a worthwhile one. But what does “worthwhile” mean to the
player? And how do we make best use of the things that are or aren’t available to
us in the virtual world we are creating? How do we combine all the gameplay and
story necessities of a game and incorporate them into coherent and rewarding
game levels?
These are all important questions. This chapter will try to answer at least
some of them.

The Detail of the World

To answer questions about world building, we have to look at what our resources
and building blocks are. We need to be able to understand what the details of the
world that we are trying to build are. To recap McCloud’s comments on world
building:
Sweating such details can make the difference between drawing a page
in SIX hours and drawing it in TWENTY—
World Building 285

— but for your readers it can make the difference —

— between KNOWING where your story takes place —
— and BEING THERE.”
Now the type of detail that we need to focus on is different for each art form, and
level design is no exception in that regard. So let’s take a look at some areas that
are typical to making game levels.
To help categorize things, I have made a distinction between two kinds of
detail: things that define how the world works (in other words, the rules of the
world), and things that exist within the world, or differently put, the content of
the world.

The Rules of the World

Games and game levels are artificial constructs that come from the minds of hu-
man beings. Due to their artificial (and virtual) nature, they require a lot of de-
sign in order to function, let alone to function to a high degree of artistic merit.
To make them do so, we have to create rules on the basis of which the world
and its content can exist. Those rules are derived, or apply to, a number of areas,
including the following

The game design

Unlike in the real world, in game levels we have the game design as the defini-
tion of what can and can’t happen.13 The game design formulates the rules of the
world, which are expressed chiefly though gameplay mechanics in the context of
the game’s physics and overarching rules of play. In some ways this is the most
fundamental set of rules, since all subsequent ones are based on these. Level
design being the application of game design makes this the fundamental start-
ing point to find the rules of the world that fit into the levels as world building
vehicles.

Gameplay mechanics
In this case, when I speak of gameplay mechanics, I mean all aspects of gameplay
mechanics, including character controls, environmental rules, and the physics
against which mechanics are used. Ultimately these are the rules that govern
the player’s ability to act, and are therefore particularly interesting from a world
building standpoint.

13
And a technical design that describes how to engineer the game’s creation and per-
formance.
286 Designing Gameplay

The technical framework

The technical and physical framework used to run and experience the game is also
very important. It encompasses things like platform (portable device, console,
personal computer, mobile phone), display device (television, computer screen)
and input device (mouse, keyboard, joystick, peripheral, motion controller).
All these elements provide an interface between our actual senses and the
virtual world in which we wish to spend some time. Their effect is therefore not
to be underestimated. It has real impact on the effectiveness of world building in
level design. Imagine the difference between creating a level for a portable mo-
bile device that will likely only be used for 10-to-30-minute bouts of gameplay
on a tiny screen with only a small keypad for input, and a modern PC with a
good monitor, keyboard, and mouse.

The Content of the World

The content of the world is best understood by imagining those things that exist
within the game’s rule set. When we create levels, we fill this world up with all kinds
of items, creatures, and environments that together form the content of the world.

Item Placement
So other than applying the rules of the world and designing gameplay scenarios
or circumstances, what do we place in the levels? What are our building blocks
for level design, and how do they contribute to world building? Since I cannot
describe all the possible content of all the possible levels in all possible games, I
think it makes more sense to point at the consequences of incorporating some of
the more typical types of level content.

Pickups and powerups

Classic level design content is to be found in pickups and powerups. From health
packs in Doom to stars in Mario 64 and countless other games and items in be-
tween, pickups and powerups are as integral to the history of video games as props
are to film and theater. Strangely, many of these items appear as if they are not of
the game world, which one would expect in some way takes something away
from the kind of illusion we are trying to foster. In a typical Quake 3 level, for
example we have fairly realistic14 and cohesive environments filled with bright
glowing and levitating (bobbing up and down, no less) pickups. Such incongru-

14
Well, kind of. In a dark gothic sci-fi/fantasy way.
World Building 287

ous pickups incidentally show that immersion can be quite robust. These items
perform a very important role in a Quake level; they fuel gameplay and help de-
fine the flow through the levels in ways that the player can clearly see. They are
important demarcations of gameplay choices, and their bright and noticeable,
even idiosyncratic, appearance actually helps the play mechanics in a level. Play-
ers don’t tend to mind this too much, probably because they are an accepted
part of the illusion and the players will happily suspend their disbelief in order
to enjoy their benefits.
In other circumstances, pickups can be completely integrated in the envi-
ronment, and can even contribute some functionality of their own. Keys can
open doors; notes can give gameplay hints or background information. What-
ever piece of the puzzle is required to progress in a level, chances are it can be
done through pickups.

Rewards and collectibles

Similar rules apply to abstract rewards and collectibles, like the stars in a Mario
game or the statues in Vice City or rings in a Sonic the Hedgehog game. The player
tends to accept these items as part of the deal when playing a game, and level
designers can use them to great effect. They are very powerful in directing the
player along the paths the level designer has in mind, or they can make players
stop in their tracks just to investigate or collect an item.

Props
Props have their own indirect ways of assisting world building. Generally, a prop
is considered no more and no less than a game object constructed to sit in an
environment. This is generally just for immersive purposes.
However, cleverly used props can do more than window-dress an environ-
ment. They can be made to evoke surprisingly strong emotions and reactions
in a player. Silent Hill 3 and Silent Hill 415 employ a giant pink rabbit that does
nothing in and of itself. Yet its placement and appearance are so disconcerting
that it massively underscores the themes of dread and horror that the game is so
dependent on.
Props can be made to resemble other objects that carry a certain meaning
in the game, in order to attract the attention of the player, or they can tie in to
story events that only take on meaning after the player has learned about certain
events. These may be items left behind by a hastily fleeing group of NPCs, or
items affected by some event in the past or the passage of time, like a corroded
metal toy or an overgrown children’s playground.
15
Konami.
288 Designing Gameplay

If used well, props can have a very powerful impact on the player’s percep-
tion of the world.

Scripted Sequences
Scripted sequences can be seen as the deus ex machina16 of level design. They are
the way in which we can directly influence events in ways that are not possible by
the regular means with which we design our levels. They are therefore very pow-
erful, but it needs to be understood that they constitute a double-edged sword.
Because scripted sequences essentially meddle in the affairs of the player the
level designer has to make sure that they are implemented with great care. If they
are used too blatantly and without care, they can be recognized for what they
are: essentially, a forced experience. This is not always bad, but there are times
when this is inappropriate, or when it takes away from the suspension of disbe-
lief fostered by the level designer. If a rock is scripted to fall down the mountain
every time the player goes past a trigger area, and the player has to avoid being
crushed by it, a problem occurs when the player is killed by that rock. The next
time the same thing will occur, and chances are there will be a sigh and an ac-
ceptance that the world is just smoke and mirrors.
On the other hand, scripted sequences can be extremely helpful in adding
spice to what is basically an automated experience. A level and its players will
run their course according to the basic design that has been implemented. Smart
scripting can bring everything to life by providing detail and character in quirky
or original moments that would otherwise not occur. Scripted sequences can be
devised that are flexible in appearance and occurrence and always feel like fresh
events, rather than seeming premeditated and predictable.

AI and “Actor” Placement

AI placement is another key area of interest. Populating the world with characters
and actors, by which I mean any creatures in the game, is fundamental to level
design. AI placement can be seen as adding detail to the world, like the use of
props, and it can be seen as something that directly adds to or influences game-
play.

16
“God from the machine.” In classical Greek plays, gods would sometimes resolve a
tricky situation through divine intervention. An actor playing a god (deus) would be
literally lowered onto the stage by a mechanical crane (machina). The term now refers
to an improbable intervention by the author to solve some problem, for example the ap-
pearance of a will in a murder mystery.
World Building 289

Level Design Symbols and Icons

Symbols and Icons are objects that represent more than themselves and their
own existence. They also stand for larger concepts and ideas and can be used
to great effect in level design and its world-building aspects. This is useful in a
metaphorical sense, but just as true in a purely practical manner.

Well-Known Symbols and Icons

Symbols and icons come in many shapes and sizes, some more obscure than oth-
ers. I have highlighted some instances of their usage they may be useful.

Hyper reality
Sometimes we can use symbols and icons, or other representational depictions,
to enhance the player’s perception of the game’s reality to such a degree that we
speak of hyperrealism. Under these circumstances, whether they are symbolic
or representational, they are made to be “more real than real.” Not only do they
resemble real subjects; they also enhance them in subtle ways to make them rep-
resentative of the core meaning of the subject (falsely or not). This may seem like
a strange concept on the surface, but it is in widespread use in art and in advertis-
ing, to name a few examples.

Time to crate
Behold the crate: that most humble servant of level designers. Is there anybody
who plays video games who doesn’t try to interact with a crate on finding one?
Crate usage in level design has become so recognizable that it has been im-
mortalized in satire. The creators of the Something Awful website have decided
that a game’s value can be measured by how long it takes a player to reach the first
crate. This has been turned into a unit of measurement known as Time To Crate
(T.T.C.) The lower a game’s T.T.C., the better it must be.
No book on level design can be considered complete without the inclusion
of this important measuring tool.

Internal Logic
We already know from our chapter on immersion that at times it can be essential
to use a game’s internal logic or game logic to maintain a suspension of disbelief.
Even if something would not be believable in the real world, such as omnipres-
ent magic crates filled with powerups, if explained well and used consistently
290 Designing Gameplay

and appropriately, they can become part of the game’s identity in a way that isn’t
questioned by the player. The advantage of effective world building in this sense
is that it allows the designer to create a much stronger internal logic than would
otherwise be possible. This in turn allows for creating a sort of shorthand with
which the level designer can convey messages to players about the world they are
in or about actions they are supposed to perform.

Making Your World Memorable

World building heavily relies on the ability to create something that is very vivid
in the mind of the audience. The newly created world must feel real and worth
spending time in. For that reason, it helps if we make aspects of it stand out.
There are a number of useful techniques with which this can be achieved,
some of which will be explored next.

Set Pieces
Remember that shower scene in Psycho? If you have seen the movie, you will. Or
the scene in Jaws where the big fin first appears in the water? Or how about the
beach-landing scene in Saving Private Ryan?
Of course you do. These scenes are amongst the most memorable ones ever
created for the cinema.
Almost all great movies feature one or more scenes that are more memo-
rable than those in lesser movies. Alfred Hitchcock went so far as to suggest that
this was fundamental to the way he made films. He was known to say that most
of what he did amounts to creating one unforgettable scene, and constructing
a movie around it. Of course there was much more to the great man’s art, but it
is true that nearly all his masterpieces feature one or two pivotal scenes that are
indeed unforgettable.
In earlier chapters we have looked in depth at level design goals. We have
found that it is almost always wise to define a set of requirements before starting
work on a particular level design component. In the case of designing a set piece,
we can often assume that one of the requirements will include the criterion that
Hitchcock has described. It has to be memorable. The way this requirement is
met depends entirely on the game and its mechanics, but nonetheless it would
be nice if the game featured moments that the player will always remember,
hopefully fondly.
Some of these moments will occur without being explicitly defined before-
hand. They can come out of emergent gameplay sequences or stem from the
player’s own imagination. Sometimes, however, we have to set up and design a
World Building 291

memorable set piece ourselves. There are innumerable ways of doing this, and
it is beyond the scope of this chapter to give an exhaustive list. A number of
examples can be mentioned, however. Scenarios like the following are not un-
common:
• against all odds (seemingly insurmountable challenges made possible),
• boss encounters (classic video game convention),
• ambush (sudden intrusion on the player by hostile forces),
• dramatic markers (encounters that carry great meaning in the gameplay
narrative that the player has embarked on).
The level designer needs to read the possibilities of the game to accurately deter-
mine what kinds of gameplay scenarios are most effective as set pieces. This will
be different for each game, and will need to fulfill completely different require-
ments from set piece to set piece.

Relevance
A word of advice on set pieces. Just because you have a chance to go to town on a
specific gameplay scenario and can design everything in it doesn’t mean that you
should make the player’s skills and abilities irrelevant to the action.

Set pieces and suspension of disbelief

We have already covered this subject in great detail in Chapter 9, but I wanted
to make sure this is not forgotten. If at all possible, make sure that the player’s
suspension of disbelief is guarded even in non-typical scenarios, as these often
occur in set piece design.

Adding Character
Sometimes it pays to add content to the levels that doesn’t necessarily fulfill
an immediate or direct requirement. It is worth considering the inclusion of
content, gameplay or otherwise, that helps support the general effort put into
the level design as a world building exercise. These are not just inclusions that
heighten a suspension of disbelief, but elements that add life and vitality to the
world.
For example, take an exploration platformer. In this genre it is essential that
the player find it rewarding to go off the beaten track and enjoy exploring the
game world. We can assume that this is already covered in a formal way, through
designed gameplay scenarios that make up the gist of the level designs for the
game. However, we can also underline this theme by providing other kinds of
292 Designing Gameplay

rewards and content that add another dimension to the game by enhancing its
character and identity.
For example, a level could feature a number of complex abandoned struc-
tures that contain texts and items that explain a certain amount of backstory to
the player. This can be accompanied by formal awards to be given to the player
as payment for the exploration effort. But it also contains content that greatly
enhances the sense of history and the mood of the environment. It grounds the
game world in a past and tells the player that there are things to learn by explor-
ing the environment. One could even include mysteries that may or may not be
solved,17 but that effectively breathe more life into the game’s character.
Efforts like these are not a luxury or a waste of resources. They are instead
an integral part of making level designs that stand out and are full of character
and identity.

Interactivity and World Building

We can’t talk about breathing life into the game world without spending some
serious moments investigating the concept of interactivity.
The word interactivity has the ring of something new in our current com-
puterized environment. People speak of interactive movies, interactive online
communities, interacting online with other people’s avatars, interactive storytell-
ing, and so on. It is clearly a wide-ranging concept that applies to our daily lives
in all kinds of ways. Interactivity in some form or another matters profoundly
to people. It has done so at a fundamental level long before we could articulate
these intellectual questions.
There are many definitions of interactive: some apply to computer science, while
others are of a more general nature. But they boil down to reacting or responding
to actions or input. This has deep underlying philosophical importance.18

I Interact Therefore I Am
One of the things philosophy looks at is the question of self-definition:
• What am I?
• Where am I?
• Is this place real?
• Are those around me as real as me?
• Does a tree that falls in the wood….?
17
Be careful not to frustrate the player’s expectation, though.
18
Please bear with me. It may sound like I am going on a self-indulgent tangent, but trust
me, it will reach a point that applies to level design.
World Building 293

These can provide endless fun if one is inclined to ponder these matters. One of
the answers to these questions lies in interactivity. We know many things about
the world because we directly or indirectly interact with it on a daily basis. Our
interaction with the world is part of how we define our own existence. Our (inter)
actions show that we do not exist in a vacuum. In other words, there are conse-
quences to our actions:
• Touching fire hurts.
• We can have an impact on other people.
• Planting a seed can lead to a new tree.
We can verify all these things through our interaction with the world. We can with-
draw our hand from the fire, convince a friend that life is worth living, and eat the
apples from the tree we planted a number of years earlier. Importantly, this means
that to some extent we have control over some of our actions, and over their pre-
dictable consequences. This control gives our actions meaning and context.

Meaningful Interactions
Ultimately, our existence and its impact in the world are determined by our sense
of interactivity. In many ways, our existence is a huge feedback system. But this
would still not add meaning to our actions if we did not perceive them to have
the power to change things around us, even if only at a minor level. Actions with-
out consequences exist in a vacuum. They lack meaning, since they have no im-
pact on the world. Most people would agree, however, that we do have an impact
on the world around us, and that our actions are therefore meaningful. This is
observable on a physical level; see the impact humans have had on nature, as well
as on a personal level: for example, in our interaction with family and loved ones.
Like it or not, the way we interact with the world around us is meaningful.

Choice
Logically, if our interaction with our surroundings has an impact on them, this
automatically leads to the concept of choice. If we believe that the outcome of a
situation, or the status of our surroundings, can be altered by our actions, it is
logical to believe that we have a choice in what actions we take to achieve the
outcome we desire.19
19
There are people who claim that nothing we do matters at all. This idea is often derived
from some extreme form of determinism, but I can only say that I disagree with them
on a fundamental level. Which incidentally is a road that leads directly to ethics, but that
goes beyond the scope of this book.
294 Designing Gameplay

This is something that for many people, or even for whole societies, lies at
the core of what it means to be human, and of what our place in the world can
be. We all have to determine what choices to make in life. It is by making those
choices according to whatever personal value or belief system we have adopted
that we give our actions meaning.

Meaningful Choices
Let’s sum up some of the philosophical material we have covered.
We experience interactivity through feedback from our environment to our
senses, which allows us to define ourselves within the context of our surround-
ings. Our actions within our environment have the potential to change things,
and since we have a choice of which actions to perform, those actions can have
meaning.
So what it all boils down to is that interaction is all about meaningful choic-
es20. A conclusion which applies directly to level design, something that will be
explored next.

Level Design Application

Many designers claim that what makes our work different, even unique from
other art forms, is that it is interactive. Because of this, they argue that we must
use and emphasize this interactivity as much as we can. This is particularly
stressed in level design, where people argue against linear levels, or levels that
are on rails, or that are too much directed by the level designer. Some even argue
that the more interactive a video game is, the better. There are even alternative
terms for videogaming that incorporate this idea. We see this, for example, in the
term interactive entertainment.
Identifying interactivity as the core of videogaming feels right on the sur-
face. Obviously, interactivity is important, but when placed under closer scru-
tiny, this does not stand up.

Mega Tetris
Tetris is one of the best and most successful games ever made. It is hard to find
any designer who would dispute this. If we try to use interactivity as a yardstick

20
As we have been told by Sid Meier already.
World Building 295

for quality, though, we find that the level of interactivity and choice in the game
is surprisingly limited.
In Tetris, the player can only shift, rotate and accelerate falling blocks on-
screen. The feedback the player is given from these actions is limited to seeing
completed lines disappear, or seeing blocks stay where they have been dropped,
a score counter counting up, and the game finishing when the blocks stack up
too high.
There are some further elements of interaction that are less important, but
this is pretty much it. If more interactivity = a better game, surely we can improve
on this simplistic design? Let’s give it a go. Our bigger and better Tetris shall be
called Meta Tetris.
MEGA TETRIS
Perhaps we can add some enemy AI to the game in the form of robots
that randomly shift blocks around in ways that are bad for the player.
These robots can be defeated by crushing them with new special anti-
robot blocks, added to the regular shapes in the game.
Furthermore, it makes more sense if the blocks come from a space ship
that is under the player’s control, possibly a freighter. The blocks are
lowered from the cargo hold, and since the freighter has a mining laser,
the blocks can be shot. Oh, and the robots can be shot too!
The laser uses up energy, however, so this needs to be replenished.
Perhaps this can be done by using the ship’s grappling hook to capture
the energy beings that are released when the robots are destroyed.
The energy beings must be destroyed as well, or they will reach large
enough numbers to merge and form a huge energy cannon that will try
to shoot down the player’s spaceship.
Okay I’ll stop there. Mega Tetris is clearly an abomination of a game. (For those
of you not familiar with the game industry, some suggestions to make improve-
ments to your game will be just as sensible as those made in Mega Tetris. These
will often come from publishers, or even from other colleagues.)
Although this admittedly is not a completely honest attempt on my part to
improve on the design of Tetris, I think it is telling, nonetheless. Mega Tetris fea-
tures a much higher level of interactivity than the original, yet it clearly is a worse
game. What, then, has gone wrong?

Game Specific Interactivity

If we look again at our earlier summary of interactivity, we may find some an-
swers:
296 Designing Gameplay

We experience interactivity through feedback from our environment to

our senses, which allows us to define ourselves within the context of our
surroundings. Our actions within our environment have the potential to
change things, and since we have a choice of which actions to perform,
those actions can have meaning.
All of these elements can apply to video games:
• Game environments give a player feedback on his or her actions.
• The interplay between this feedback and the player’s actions define the
player’s capabilities within the game world.
• The player’s actions have an impact on how the game progresses, and the
player’s choices within that context can be meaningful.
It is that last bit, “can be meaningful,” that we need to focus on.
Sid Meier famously defined21 games as a series of interesting choices22. In the
context of games, there is much overlap between interesting and meaningful
choices.
This points at the crux of the matter: interactivity only makes sense if it is
interesting or meaningful. Otherwise it is completely redundant, or, as in the
case of Mega Tetris, no more than an irritant. (The trick is of course to determine
what choices are meaningful or interesting to the player.)

Meaningful Interactions and

Choices in Levels
As we have seen, it is not the amount of interactivity that is important, but the
quality of the interactivity. Is it appropriate? Does it support the game? Is it
meaningful? To a degree these concepts will always be subjective, as they often
require an individual judgment call to be made. In a way, we are talking about
the quality of experience, which is a philosophical minefield.23 I’d rather focus
on level design that makes gameplay non-arbitrary and employs interactivity as
a tool to attain this.
For example, if a game’s gameplay mechanics are geared towards manipulat-
ing the physical objects in an environment, it is appropriate to provide interac-

21
As reported in Andrew Rollings and Dave Morris, Game Architecture and Design: A
New Edition, New Riders, Berkeley, CA, 2004, p. 61.
22
Also often quoted as interesting “decisions.”
23
To get an idea of the dangers, read Robert Pirsig’s Zen and the Art of Motorcycle Main-
tenance.
World Building 297

tion possibilities that allow for this. Those kinds of actions are meaningful within
the context of that particular game and are desirable, since it is our task as level
designers to present our audience with a game that provides meaningful choices
and interactions.
What follows shows a number of ways in which we can do so.

Interesting Choices
At the most basic level of our level designs lie some very simple choices that we
present to the player, often without much thought and without realizing that
they are choices. If a room has two exits, the player has to make a choice. If an
area features enemy creatures, players can choose to fight them or avoid them.
A pickup can ask if the player wants to replenish his or her health. This is almost
mundane stuff and as a consequence is often not thought about. But by doing
this, level designers miss opportunity after opportunity to make their levels in-
teresting, or at least filled with interesting and meaningful choices.
Let’s look at a very basic but typical scenario: a t-section.
Figure 13.2 shows a simplified layout that presents the player (P) with the
choice of choosing one direction over another. With a bit of investigation, the
player learns that both directions are guarded by an enemy creature (E).
Figure 13.3 shows the same setup, but one direction is more heavily
guarded than the other, and the player can spot treasure pickups behind the
guardians (t).
The first scenario offers an interactive challenge presented as a choice, in this
case one based on engagement with enemy AI. But is it an interesting choice?
What does it matter in which direction the player chooses to go? All the player
knows is what is in front, and in this case the choices are equally good or bad.
This means that they aren’t interesting, but arbitrary.

Figure 13.2. T-Junction with two Figure 13.3. T-Junction with unequal
equal enemies. enemies.
298 Designing Gameplay

The second figure shows the same scenario, but with very little effort, it has
been made intrinsically interesting. First, the player is shown that there are dif-
ferent enemy encounters to be engaged in, depending on the choice. This is in
itself not that interesting; most players would pick the easy enemy. However, the
player is also shown an alternative reward that is based on what is chosen.
Now the player has to make a judgment call: risk more for a bigger reward,
or risk less for a small one. This IS an interesting choice, and one that ties in to
all kinds of considerations that the player has to make. How much health does
the player have? How well armed is the player? How did the player fare last time
against enemy one, as opposed to enemy two? And so on.
If we always ask the question, “Is this an interesting choice?” we can trans-
form the most basic and mundane gameplay scenarios into interesting ones by
providing meaningful interactions.

Meaningful consequences
The above scenario with the two directions shows an example of consequence.
Engaging in combat with one enemy over another will yield a different re-
ward, in other words a different consequence. Consequences themselves can
be meaningful as well and can therefore greatly add to the quality of the level
design.
The interesting choice presented in the second figure shows two awards that
the player can earn: one of 100 gold pieces and another of 10. But what if the
player already owns 100,000 gold pieces?

Formal Rewards
By formal rewards24 I mean in-game, tangible rewards, like gold coins or pickups,
not abstract rewards like “a sense of achievement,” although rewards of that kind
are of course essential. What I am talking about is simple interactions leading to
direct rewards. Destroying a crate will yield a powerup, killing a baddy makes the
baddy drop the loot, opening a chest will give the player treasure. Video games
are littered with gameplay of this sort. Reward dispensers like this are extremely
easy to implement, which explains some of their popularity. They provide a way
for level designers to hand out rewards in places they find appropriate. Players
understand the convention perfectly, which makes it a favorite tool. Addition-
ally, it is just fun to interact with of one of these reward dispensers and see what
you are going to get.

24
Yes, I made that term up, as well.
World Building 299

Intermittent reinforcement and the slot machine effect

There is a lot of scientific data that explains, at least partly, why this is such an
effective reward mechanism. Much of this data is in the field of behavioral sci-
ence.25 Although there is some controversy associated with the subject, this kind
of intermittent reinforcement has been used by the creators of gambling ma-
chine; hence it is sometimes referred to as the slot machine effect.
The so-called slot machine effect is a well-known psychological phenom-
enon and is understood well enough that applications of it have become quite
sophisticated. There are many recognizable examples of this in the real world.
The chocolate eggs of Kinder Surprise have hidden gifts, there’s a toy at the bot-
tom of the cereal box, and of course, there are the slot machines themselves. The
application of this knowledge in entertainment is not good or bad intrinsically,
but there are serious ethical implications that need to be considered, especially
when done for commercial gain or when aimed at children.
I am not suggesting that as level designers or game designers we cannot
make use of intermittent reinforcement. Nor am I saying that the players of
our levels are helpless and without personal responsibility in this regard. How-
ever, I do want to make the connection clear that similar processes are at work
in activities that are known to be habit forming, like gambling. We need to be
aware of our own responsibilities in deciding when and when not to employ
this principle.

Exploration
Sometimes the simplest and most rewarding interaction with an environment
comes from enjoying it by traveling through it. In other words, exploration can
be a meaningful interactive activity. Uncovering the joy of new environments
works as a reward in its own right. This hints at certain lessons that need to be
learned when one creates levels that have to foster enjoyable exploration. Un-
locking a new level area should not be a punishment. Allow the player time and
safety to really enjoy and benefit from the exploration. This will make it much
more likely that the player will want to keep on exploring.

World Building Conclusion

Seen through the prism of world building, level design touches on almost any
aspect of the craft we can think of. It takes notions of gameplay narrative and

25
See, for example, the concept of operant conditioning.
300 Designing Gameplay

classical narrative and asks the designer to construct an experience that is deep
and immersive and holds up to scrutiny when thoroughly explored by the player.
It deals with the rules of the world, as well as the details in the world. It demands
a consistent and intelligent stance on a level’s interactive elements and the way
in which we populate it with AI inhabitants. Finally, it gives us a framework in
which we can explore the exploration of the world we have built, and the way in
which this exploration can be enjoyed.
World building overlaps most aspects of level design. While the two are not
the same thing, level design can be greatly helped by an understanding of the
world building needed to make it effective.

Practice

Example 13.1: Rules of the World—

Internal Logic and Consistency
Summary
The rules of the world govern the behavior of the content of the world. This in
turn forms a framework in which player actions take place. Since we have to do
our level designs based on these parameters, it makes sense to incorporate the
rules of the world into the level design at a fundamental level.

Game Genre
This technique is suitable for most games.

Goals to Achieve
• Link the rules of the world to level design scenarios.
• Emphasize the coherence of the levels through internal logic.
• Apply consistency to internal logic so the player can formulate their own
mental map of the rules of the world.

Description
(Example type: General)
Level designers can manipulate the player experience by interpreting the game
rules that are defined in the game design. This is an essential part of what we do,
World Building 301

and something that is required for all levels. We can expand on this, however, by
interpreting these mechanical rules by trying to formulate an internal logic that
can be interpreted by the player. Indeed, sometimes we can create internal logic
by bending the mechanistic game design rules to fit our level design goals.
For example, if the game design rules spell out that the world uses a predict-
able gravity model, and that falling objects do damage to what they land on, then
it is very easy to take that and create an internal logic that the player can have
great fun with—especially if we help this along by making sure our levels feature
many opportunities for the player to drop things on targets, like enemies or pres-
sure plates, or whatever else would benefit the player.
It is important to make the distinction between mechanistic rules and in-
ternal logic as the latter is the one that the player is aware of, while the former is
the one that the level designer is aware of. Within the framework of world build-
ing, this gives us an opportunity to create levels that incorporate and interpret
the rules of the world and apply them to, or utilize them alongside, consistent
internal logic. The result is a coherent, deep, and detailed gameplay environment
for the player.
This principle goes beyond the literal nature of the gravity example I have
given.

Fabricated internal logic. I’d like to give an example of internal logic that
has to be created by the level designer, which through consistent appli-
cation in a level becomes part of the rule set of the game in the mind of the
player.
In a game based on athletic ability, like a platforming game for example, we
often motivate the player by showing them hard to reach areas and enticing him
or her to somehow find a way to reach those areas. We can apply a consistent
internal logic to these challenges that adds to the understanding of the game for
the player. To do this we define our own level-design-based logic rules. Let’s take
a 2D platform game set in a sci-fi environment:

Internal logic rule 1. If the platform is far away, floating out of reach of every-
thing, the player has to perform a task, possibly locate a switch, for it to float into
a position where the player can reach it.
Internal logic rule 2. If the platform is close to other platforms, then the play-
er can get to it by conventional means—possibly by jumping on the heads of
enemies in a timed sequence, or by dropping onto an adjacent platform from
above.
If we show the player these two rules consistently early on—for example, by
making sure that early examples are all solvable within the same screen, and only
302 Designing Gameplay

slowly making the solutions appear further away in subsequent iterations—then

the player won’t have a hard time learning these rules.
Additionally, further internal logic lessons have been taught in the process:
• obscure platforms feature valuable rewards,
• jumping on enemies is a valid technique to reach areas,
• switches control platforms,
• all platforms with rewards can be reached in one way or another.

All levels do this to some agree, but if we plan these things out beforehand, or
at least think about the possibilities this offers to create interesting content, our
levels will end up more interesting and coherent.

Further Notes
Internal logic can also be abstract, or surreal, or even humorous. Imagine a run-
ning gag in a game, or disturbing visions being the result of specific actions.
There is plenty of new ground to be covered in this area.

Example 13.2: Implied Detail

Summary
In level design, implied detail can take on various guises and forms and can be
used for various purposes. Aspects of it can be used in a world-building context,
specifically by using localized detail to imply a larger and detailed world in which
the level is placed.

Game Genre
This technique is suitable for non-abstract games.

Goals to Achieve
• Perform effective world building.
• Maximize the use of art assets in the process.

Description
(Example type: Original)
Implied detail. We can add greatly to the perception of detail in a level by in-
cluding detail that implies further content, without the level designer having to
World Building 303

create it. This technique is based on the idea that if we add this detail in strategic
ways and places, the players will assume this detail exists throughout the levels
and the game in general.
Take, for example, a window that looks out onto an alley. The player can’t
get to it, but he or she can see a number of details from their viewpoint. These
details can be of such a nature that they imply additional content and detail just
out of view. Here are some examples:
• a street sign that lists a multitude of other street names,
• an opposite building with a window that can be peered into,
• street sounds,
• posters detailing other events in the game world like concerts or political
rallies,
• weather effects.

All these things point at a much more detailed and larger environment, full of
variety and life. Yet, the level designer only has to include a few assets to create
this perception—parts of a building, signs, sound—and the player will reconcile
these hints at other things into a mental image of an environment that is detailed
and robust.

Further Notes
These are just some examples but it should be clear that the actual us-
age depends on the detail that the level designer wants to imply. This
will be different for all games, but the principle is clear and easy to
reproduce.

Example 13.3: Built-in Meaningful Choices

Summary
We know that an important aspect of level design lies in confronting the play-
er with meaningful gameplay choices. This can be done as part of the level
design scenarios that we create piece by piece, but this can also be a goal of
the world-building process in a way that is fundamental to the fabric of the
level itself.

Game Genre
This technique is suitable for most games.
304 Designing Gameplay

Goals to Achieve
• Make the likelihood of interesting gameplay choices part of the world
building process.
• Do this with a minimum of necessary variables.

Description
(Example type: Existing game/general)
When we create a level, we can ask of every gameplay scenario that we create
that it presents a meaningful and interesting gameplay choice to the player. This
is the basis of much of our design work. We can, however, also ask this same
question of the level environment itself. The structure and fabric of the world
we build can ask these questions on its own, if we design it with such goals in
mind.
Dyson’s built-in, meaningful choices. I have used this principle extensively
in the game Dyson26 for which among other things I did most of the design. In
Dyson the environment consists of asteroids that contain resources and can be
conquered by the player or by opposing AI teams. These resources are repre-
sented by energy, strength, and speed attributes, and these differ from asteroid to
asteroid. These resources are mined and used to create the game’s units, which
are exploration and conquest entities that the player can use to conquer more
territory and attack opponents. A key part is that seedlings produced on aster-
oids take on the resource attributes of that particular asteroid, which defines
their capabilities. So, a seedling produced on an asteroid rich in strength and
energy but poor in speed will produce units that are good at fighting and very
durable, but are very slow movers. This will have consequences on their chances
of success when engaging with opponents or navigating and interacting with the
level space. Levels are built in such a way that there is a rich variety of asteroids
of varying sizes and consequently of very diverse attribute composition. Within
these asteroid fields key details are built in, for example, to do with enemy place-
ment, optimal routes, and asteroid distribution.
Taking these things into account shows us that the level design has incorpo-
rated a style of world building that maximizes the chances of meaningful game-
play choices.
We have simple parameters incorporated into the level itself, and they pro-
vide constant choices on immediate gameplay consequences and actions:

26
www.dyson-game.com. Developed by Alex May and Rudolf Kremers. Music and audio
by Brian Grainger.
World Building 305

• where to send units,

• which asteroids to conquer,
• what units to breed,
• what units to use to engage enemies,
• which asteroids to defend,
and so on.
This is possible because there are only a few basic variables that determine
the makeup of a level, but they all affect gameplay in one way or another. There
is almost nothing in the fabric of the level that does not contribute to this, and
the result leads to interesting complexity, and interesting choices, as the default
basis for most gameplay actions.

Further Notes
This is a technique that is extremely useful in games where we would like to
encourage emergent gameplay. Sandbox games are especially helped by this,
since they need by their very nature to be engaging at a fundamental level. And
that means that the world building aspects of sandbox levels are fundamental
to the success of such games. The game world itself needs to support interesting
choices.
Puzzles
14
W hy do we enjoy puzzles so much? In almost all cases, puzzles are not
real and do not offer a direct benefit. They are often taxing and require a
serious amount of concentration. They can cause severe frustration, even anger.1
You would expect that people have better things to do with their time.
Yet playing with and solving puzzles is a human activity that people are en-
gaged in all over the world. Some puzzle types seem so universal that they pop up
and are understood within societies that culturally could not be further apart.
Pretty much all types of puzzles present the player with a challenge. They
have a solution or require a certain strategy, or a particular skill, in order to be
completed or at least enjoyed.
In many ways they behave exactly like games, in fact, quite often puzzles are
classified as games.

Concept

A Monkey, a Termite Hill, Some Salt, and

a Bit of Rope
Picture the following puzzle scenario in a game, for example a point-and-click
adventure:
The player is located in a desert and has to find water within a certain time
frame. The tools to solve this puzzle include
• a monkey,
• some rocks of salt,
1
I wonder how many Rubik’s cubes have been destroyed in fits of anger.
307
308 Designing Gameplay

• a termite hill,
• a length of rope.
All items are to be used to obtain water.
This sounds like a typical non-realistic, arbitrary puzzle scenario. It could
have appeared in one of many of the classic point-and-click adventures of the
past, like Escape from Monkey Island or Sam and Max Hit the Road.”
Strangely enough, it is in actuality a real-world puzzle of survival that is faced
by the bush people in the Kalahari dessert. I once saw a documentary; I think
it may have been called Animals Are Beautiful People, which features a scene that
has stayed with me all my life. It shows how some Kalahari bushmen have found
a solution to the problem of finding enough water to survive.
The documentary includes a scene where one of the bushmen plays with a
bit of salt in front of a monkey. The monkey, curious and fond of salt, becomes
extremely interested. When the bushman is secure in the knowledge that the
monkey is interested, he walks up to a termite hill. He finds a cavity in the hill
just large enough for his hand to enter, and drops the salt inside. He then walks
off, out of the monkey’s view.
The monkey can’t contain itself. When it thinks the coast is clear, it quickly
runs up to the termite hill. It finds the cavity that holds the salt and grabs it tight-
ly in its paw. But the entry hole is too narrow for the monkey’s fist, as long as it
is holding the salt. Until it lets go, it is trapped. But stubborn as it is, the monkey
will not relinquish its prize. At this point, the bushman comes out of cover and
quickly approaches the monkey. Although it could easily escape, it keeps hold-
ing onto the salt and is easily captured.
In the next phase of this strange plan, the bushman tie the monkey to a
stake, but lets him keep the salt. Sure enough, the monkey starts licking it. Since
the monkey has been tied down in the hot sunlight, it soon becomes extremely
thirsty. At this point the bushman releases the monkey, who immediately shoots
off, closely followed by the man. The monkey runs straight for a hidden reservoir
of water.

Problem Solving Skills

The point of this amusing but slightly bewildering anecdote is that there is a cor-
relation between real life challenges and our ability to understand complex and
often abstract connections.
Have you ever seen the TV show MacGyver? It was an ‘80s show in which
the hero found himself in a jam in every episode and used his knowledge of sci-
ence to find a solution. If a villain locked him in a cell, he would “simply” com-
Puzzles 309

bine charcoal and bird droppings to create gunpowder and blow the lock with a
small controlled explosion2 (or some equally fascinating solution).
A certain amount of lateral thinking is needed to come up with the solutions
described in the previous section. Although the examples seem far-fetched and
irrelevant to most of our lives, it’s true that we engage in this kind of puzzle solv-
ing. Or at least, we use the same skills and processes, on a very regular basis.
The Kalahari Bushmen and MacGyver examples present exaggerated exam-
ples of the use of the problem solving skills I am talking about. Nonetheless, they
showcase a very real phenomenon, that human beings are very good at making
cross connections and seeing relationships that are not apparent on the surface.
This is key to problem solving in general:
Although all problem solving relies on a largely shared cortical network,
the sudden flash of insight occurs when solvers engage distinct neural
and cognitive processes that allow them to see connections that previ-
ously eluded them.3
And if we can use these abilities to solve survival problems, we can also use them
in other situations that require problem solving.

Survival skills
As we have seen with the purpose of “play” in general, it is not unreasonable to
make a connection between survival and training for difficult challenges through
the use of puzzle play. We can greatly improve our chances of survival by training
our mental and physical abilities by training our puzzle-solving skills.
Let’s look at some typical skills associated with solving puzzles or problems:
• pattern recognition,
• lateral thinking,
• hand-eye coordination,
• abstracting/conceptualizing a problem,
• memory usage.
There are many others.
All of these can come into play even when we are confronted with very mun-
dane or simple puzzles, such as jigsaw puzzles or skill puzzles. Nonetheless, all of

2
Gripping stuff !
3
Mark Jung-Beeman, Edward M. Bowden, Jason Haberman, Jennifer L. Frymiare,
Stella Arambel-Liu, Richard Greenblatt, Paul J. Reber, and John Kounios, “Neural Ac-
tivity When People Solve Verbal Problems with Insight,” PLoS Biology, http://biol-
ogy.plosjournals.org/perlserv/?request=get-document&doi=10.1371%2Fjournal.
pbio.0020097&ct=1, 2004.
310 Designing Gameplay

these are important abilities, and must be trained and sharpened throughout life.
Realizing which of our abilities and senses are used the most when we engage
in problem solving is very helpful. It can add to our understanding of creating
puzzles and is therefore useful within a context of level design.

Concept Summary
Puzzles provide a safe and controlled way of testing one’s skills and abilities in
very specific and precise ways. They can be started, restarted, or abandoned with-
out serious consequences to the person involved. Because of the playful nature
and the fact that they use formal rules, it is fair to say that puzzles provide an-
other form of gameplay, or at least are a subset of the “play” family of activities.
They can also be incorporated in the kind of teaching gameplay we talked
about in Chapter 2, where we saw that they can provide a number of very impor-
tant and specific functions.

Theory
The list of puzzle types is endless: block puzzles, pattern puzzles, skill puzzles,
mechanical puzzles, door and key puzzles, pressure plate puzzles, and so forth.
The level designs of most games littered with different types of puzzles. Some
puzzle types are well known outside of video games, while some others are
unique to the form. But why are they so ubiquitous? Why are they so useful
that level designers keep using puzzles, game after game? Why do people enjoy
engaging in puzzles?
It is clear that puzzles provide some very useful functions, or their usage
wouldn’t be so enormously widespread. This chapter will list a number of rea-
sons why, and ways in which, their useful qualities can be maximized.

Puzzles as Formal Tests

Early in this book, we looked at the concept of games as teaching mechanisms.
We concluded that teaching gameplay skills also requires the level designer to
test the player’s progression in his matter:
Teaching mechanisms are meaningless unless that which is taught is
tested and put into practice. There has to be a way to test the player’s
knowledge or proficiency within the game, or the game will lack a
Puzzles 311

purpose. This is another intrinsic goal of level design. If the gameplay is

taught well and the player gets tested in an enjoyable manner, the level
designer has done a good job.
I believe that one of the reasons we always keep reaching for puzzles in our level
design is that puzzles constitute a fantastically practical and enjoyable mecha-
nism for testing skill and knowledge. Puzzles can be used as doorways or gate-
keepers. Access is granted if the player knows how to solve the puzzle. This ap-
plies in a physical sense: we can literally stop progression. Or we can apply it less
rigidly: solve this situation (puzzle) and receive a reward. Puzzles are formal
tests of the player’s gameplay knowledge and skills. And as tests go, puzzles are
good ones, because they ask all the right questions.

Gate Keepers or Funnels

A side effect of this function is that puzzles can be used as gate keepers that only
allow the player to progress after gaining the necessary skills to solve the puzzle.
Alternatively, the game can withhold the means of solving the puzzle unless the
player has performed certain other actions. This creates a funnel through which
the player has to pass before the level opens up further or before progression is
allowed.

Game survival skills

A specific gatekeeper function is found in puzzles that test for skills that are neces-
sary later in the game. Ideally these puzzles are presented in a non-threatening
manner so the player learns important skills in a playful manner, without being
under undue pressure. If the puzzle is presented as a progression puzzle, possibly
with a secondary reward, the player can be forced to master a skill without real-
izing that it will play a more vital part later in the game.
Teaching important game survival skills or strategies in this manner is very
effective, and it lets the level designer create subsequent challenges with accurate
knowledge of the player’s skill set.

Dangers
It is very dangerous to make these kinds of skill gates absolute. Players may not
be able to grasp a certain skill or may be made to feel inadequate if the difficulty
of the puzzle is misjudged. If at all possible, allow for multiple ways to solve the
progression puzzle, or allow the difficulty level to decrease as time passes.
In addition, we must be careful not to abuse gateway puzzles and create ex-
cessive or arbitrary requirements of time investment for the player. There are
312 Designing Gameplay

serious issues in many games with so-called fetch gameplay, or what I term col-
lectathons.
Fetch gameplay is a situation in which the player’s progress is frustrated at a
specific physical location in a level and the player is told to go find certain items.
Within reason, this kind of gameplay is fine, but often we see that the level dura-
tion is padded out excessively by blatant fetch quests that don’t add anything to
the experience. A clear sign that a level design is weak is when one fetch quest
triggers another, with little difference between the two except for the name of
the item that must be collected.
Fetch gameplay can turn into a real nightmare when misguided level de-
signers suddenly realize that they can ask their players to fetch not one, but
many, items before progressing. Even that is not necessarily wrong if it fits the
goals and motivations that are carefully fostered in the player. But this is so
easy to misjudge that designers need to be very careful about employing such
methods.

Puzzles as Information Conduits

Almost all puzzles stop the player and ask questions. For example, they may ask
things like:
• Have you mastered this skill?
• Are you strong enough to perform this task?
• Do you understand this danger?
• Have you encountered and absorbed that earlier lesson?
• Are you ready for the next bit?

Puzzles can be used for all kinds of tests. They can function as benchmarks or
bottlenecks or exams, among many other applications.
In some ways, games are like schools, teaching us gameplay (survival)
skills, where the level designs are the curriculum, and puzzles are some of
the questions being asked in exams and tests. (And the level designers are
teachers).
Although this comparison is slightly strained, it is nonetheless a useful one.
We need to make sure that we provide a good environment for teaching and
learning, and that the curriculum covers the right building blocks for future les-
sons. It needs to be fun to learn.
We have seen that these properties mean that puzzles can be used as a for-
mal testing device, but it is also good to note that this means that they can con-
vey or receive information.
Puzzles 313

Information for the Level Designer

An often-overlooked aspect of puzzle design and implementation is the oppor-
tunity they offer for gathering information about the player. How a player deals
with a situation can tell the level designer a great deal, and in the context of a
puzzle, that information can be structured and specific. Good examples of his
can be found in data mining and in reading the outcome of conditional puzzles.

In-game data mining

If the game’s technology allows for player data to be captured and measured, a
designer can find all kinds of useful applications for it. Data mining in this regard
is in principle no more than measuring player actions and improving the game
experience by interpreting the recorded data and putting it to good use.
Level designers may alter the spawn rate of certain items by reading the play-
er’s inventory at certain moments, or they may decide to allow progression in an
area of the game when it becomes clear that the player has spent too much time
trying to solve a gameplay puzzle.

Conditional level design and puzzles

A level designer can offer multiple outcomes based on the way the puzzle has been
solved. Puzzles can be designed to offer diverse progression possibilities, based
on certain conditions being met, taking full advantage of this knowledge. For ex-
ample, if a puzzle is solved by dexterity, the level designer may open up a path
that offers a large number of athletic possibilities. The level designer knows that
the player can cope with this because of the way that person has manipulated the
puzzle. Another possibility lies in testing for certain objects. If the player solves a
puzzle by using a certain item, the level designer may opt to remove it from the
rest of the level and replace it with something that is still desirable to the player.

Information for the Player

There is also an opportunity to feed information back to the player, often in a
very natural way. A puzzle solution may be based on gathering information from
the game’s environment and putting it to use in the puzzle itself. This can be
fairly explicit, like teaching players enough about the backstory of the game for
them to answer a specific question.4
The application can also be indirect, where information given to players al-
lows them to interact with the game in ways they wouldn’t know about before.
4
This kind of application is often seen in survival horror games.
314 Designing Gameplay

For example, the puzzle may teach the player a new way to use a weapon or a
handy item.
Subtle as this technique may be, it can really pay off to have a good look at
ways in which you can communicate important information to the player, as this
is one of the key tasks that a level designer has.

Puzzles as Reward Mechanisms

To quote Raph Koster again:
[F]un from games arises out of mastery. It arises out of comprehension.
It is the act of solving puzzles that makes games fun.
In other words, with games, learning is the drug.5
We know that much enjoyment in gameplay comes from the learning process:
it is enjoyable to master new skills. Additionally, puzzles can be engineered in
such a way that solving them gives the player a tangible award, like delivering a
powerup or opening up a new area of the level.
The sky is the limit, really, when it comes to this particular use of puzzles;
there are too many examples to name them all. To highlight one, however, a very
useful application is found in exploration gameplay.

Reward for Exploration

A great way to reward the player for exploration is by placing independent puz-
zles all through an environment where they can be played with at the player’s
leisure, without any negative consequences. Because the puzzle would in this
context be purely used as a reward dispenser, the player can invest as much time
into it as desired.
A note of caution: it is very important in exploration gameplay to not expose
the player to time pressure, as this punishes the act of free exploration itself.

Purely self-contained puzzles

The above example acts as a reward for exploration gameplay. However, it is en-
tirely justified to design puzzles as completely independent reward dispensers.
A slogan for this might be, “Solve the puzzle, receive a prize.”
This may be a simple tactic, but it can be very pleasing indeed for the player, and
it gives the level designer a way to teach the player that the game world is equipped

5
Raph Koster, A Theory of Fun for Game Design, Paraglyph Press, Scottsdale, AZ, 2005,
p. 40.
Puzzles 315

with these kinds of reward mechanisms. This is an important aspect of rewarding

exploration gameplay, or increasing the enjoyment level of the game in general.

Puzzles as Agency Enhancers

It isn’t just the learning of skills that is enjoyable: we also revel in that mastery by
applying those skills in fun ways. When one learns how to play the guitar, it is fun
to actually end up playing songs. We have seen throughout the book that a sense
of agency, the ability to act within the game world, is fundamental to a number
of level-design goals, such as a deepened sense of immersion and a willingness to
engage with the interactive setting of the game.
A certain positive reinforcement can be achieved by making sure that play-
ers get to test the gameplay skills that have been taught to them by the level de-
signer, in various ways and settings. It is great fun to try out new abilities and
learn to master them, or to feel the thrill of successfully beating a difficult (game-
play) puzzle.

Gameplay Puzzles
Sometimes it is useful to see typical level design tasks in a different light in order
to keep their application fresh and inspired. Within this context, I can recom-
mend applying some of the subject matter discussed in this chapter to formally
designed gameplay scenarios. Many level designs feature detailed and contained
game-play scenarios that have a number of specific ways in which the player can
tackle them. It is perfectly reasonable to see these encounters as puzzles, and to
then try to apply the positive applications of puzzle design to these situations.

Puzzles as Pacing Devices

As already discussed in Chapter 14, puzzles are also very useful as pacing devices.
For example, a level’s action level can be affected by the placement of more con-
templative puzzle gameplay. If the player has just had a long and tense sequence
of enemy encounters to contend with, it can be a great relief to be able to take it
easy and do some leisurely work to solve a nice rewarding puzzle.
Depending on the type of puzzle employed, the level designer can directly
influence the pacing of a level in order to break up the gameplay, for example, so
that it isn’t all fast action all the time.
Puzzles are really useful devices to calm down players after an exciting
part of the game and focus on the details of the world around them. We
316 Designing Gameplay

build all this detail into the world and in many cases the player whiz-
zes through it at a breakneck pace. Puzzles and “Downtime” are like a
sorbet in a multi course meal, in that they allow the player to better ap-
preciate whatever action comes next. Without those pacing contrasts,
everything becomes a numbing blur of relentless action, which winds
up being fatiguing and not fun after a while.6
If used in this way, puzzles become a useful modular gameplay object, where the
precise nature of the puzzle is less important than the content of the puzzle itself
(within reason, of course.)

Theory Summary
Puzzles are an enormously helpful tool for level designers. Players are already
familiar with their use on a fundamental level, including the training and testing
of important survival skills or as part of enjoyable activities. In some form or
another, we engage in puzzles on a nearly daily basis.
The diverse ways in which they can be used in level design reflects this fact.
Puzzles are useful for testing player skills, providing a chance to interact within
the game world, influencing the pacing of a level, or providing enjoyment on
their own accord.
Puzzle gameplay is fundamental to level design. Indeed, entire games have
been based on puzzle gameplay. It is therefore very useful to build up an arsenal
of techniques and methods with which to incorporate them into game levels.
The next chapter will show very practical ways in which this can be done.

Practice

Example 14.1: Puzzle as Information

Conduit—Investigative Puzzles
Summary
Puzzles can be used as a discreet challenge, where the enjoyment comes from
the focus on the puzzle itself. Puzzles can also be used to tie into the game world
6
Josh Weier, Half Life 2: Raising the Bar, Valve/Prima Games, Roseville, CA, 2004.
Puzzles 317

on a wider level, by providing information about the game world to the player in
the process of solving the puzzle.

Game Genre
This technique is suitable for games that allow the player to explore a representa-
tive game environment and gather information.

Goals to Achieve
• Design and implement an original puzzle.
• Increase the player’s knowledge of the game world.
• Deepen immersion and world building.

Description
(Example type: Original)
Investigative puzzles are puzzles that require the player to investigate their envi-
ronment to solve a puzzle. This can apply in many diverse ways, depending on
the specific circumstances of the game.
Examples can be given where the puzzles are part of the challenges incor-
porated in the actual environment: In an exploration game, the player may need
to investigate the behavior of certain creatures to get past them. In a crime ad-
venture game, the player may have to find certain written facts spread through
various written accounts located in the game, perhaps to be able to open a safe.
A fantasy game may require the player to find a specific mix of herbs in order to
create a specific spell that allows the player to proceed.
Other examples can consist of actual formal puzzles, encountered in the
game. This occurs frequently in survival horror games, as they have become a
convention of the genre. With these kinds of games, nobody will bat an eyelid to
find a puzzle box that requires a series of passwords to be entered in the correct
order, or some other fairly contrived puzzle construct. The correct words and
order of words may only become available if the player investigates and learns
about the game’s back-story.
What matters most in all of these examples is that the process of solving
these puzzles is not one that occurs in isolation of the rest of the level. If done
well, they are in fact deeply integrated and can be used with great effect to teach
the player all kinds of extra information about the level or the game as a whole.
A level designer can construct investigative puzzles in a variety of ways, but
it helps if they create a mental map of what it is they are doing. Generally these
puzzles are made up of a number of parts:
318 Designing Gameplay

• Solving the puzzle rewards the player with a consequence.

• The player has to construct a solution from a number of components.
• (Optional) The components need to be manipulated to form the correct
solution.
• The player has to go out and investigate the environment in order to find
these components.
In the crime adventure game example this process unfolds as follows.
Solving the puzzle rewards the player with a consequence. The player
has to unlock a safe to find a suspect’s diary.
The player has to construct a solution from a number of components.
The safe combination is the birthday of a long lost friend of the suspect. This is
broken up into three numbers (day, month, year) that need to be found.
(Optional) The components need to be manipulated to form the cor-
rect solution. The date has to be put in the right format (dd/mm/yy).
The player has to go out and investigate the environment in order to
find these components. The player needs to go and talk to people who know
the suspect, research the suspect’s life study and the town’s history, and so on, in
order to discover the clues that eventually reveal the exact age of the suspect’s
lost friend.

Further Notes
Investigative puzzles offer much to level designers, because they incorporate the
level as a whole. It provides the player with a deep and immersive context from
within to explore their skills as well as the level’s content, and—despite its con-
structed nature—it can provide a natural pretext for gameplay.

Example 14.2: Puzzles as Formal Tests—

Teaching Mechanic
Summary
Puzzles can be used to test if a player has mastered a certain skill and provide an
opportunity to acquire that skill if lacking.

Game Genre
This technique is suitable for all games that employ puzzles.
Puzzles 319

Goals to Achieve
• Test if a player has acquired a specific skill.
• Teach the player that skill if they haven’t acquired it.
• Do this as an optional gameplay encounter.

Description
(Example type: Original)
If a level wants to check if a player has obtained a specific skill, and wants to pro-
vide an incentive for the player to lean it if necessary, then the level designer can
formalize this into a puzzle design.
The puzzle’s solution is only viable by the application of a specific skill, and
the puzzle itself provides the teaching mechanism to acquire that skill. Let’s take
as an example a gameplay scenario where the player can see an award, but is pre-
vented from accessing it unless they solve a puzzle. This can take place by show-
ing the player a locked treasure chest, protected by a complex locking mecha-
nism that consists of words that need to be aligned in an order that spells out
a specific sentence. The problem is that the sentence is written in an encrypted
language. The player needs to learn how to decrypt the language, and then sub-
mit the correct answer in encrypted form.
Let’s say that in our example the player once again has to input correct num-
bers. There are 4 slots, and two of them are already given:
Input: 15 / 16 / – / –
The player may find out through hints that the first two digits stand for O and P,
which gives the player the following sequence:
Input: 15 / 16 / – / –
Translation: O / P/ – / –
Through logic reasoning, or through trial and error (or both), the player
should be able to figure out that the numbers apply to the positions of the
letters in the alphabet, and that the correct word/number sequence spells
“OPEN”:
Input: 15 / 16 / 5 / 14
Translation: O / P/ E / N
When successful the player will not only receive the contents of the treasure
chest as a reward but will also have acquired the ability to decipher similarly
encrypted text throughout the game, which can provide additional gameplay for
future level design scenarios.
320 Designing Gameplay

Further Notes
Please don’t use this example literally unless appropriate. It has been used many
times in the past and has become a cliché. I only used it as it provides a good
demonstration of the principle.

Example 14.3: Puzzle as Pacing Device—

Player Controlled Pacing
Summary
Puzzles can be used as a variation to previous gameplay, slowing the pace of
the game down and giving the player a chance to do something different, until
they solve the puzzle and carry on along the game’s path. This can be done in a
binary fashion, the puzzle is solved or not, but it can also be done in a slightly
different way, one where the player has to decide if they “want” to solve the
puzzle.

Game Genre
This technique is suitable for all games that employ puzzles.

Goals to Achieve
• Take control of the pacing in a level.
• Create a puzzle situation within a gameplay scenario.
• Tie various aspects of the puzzle to changes in level pacing.
• Let the player feel like he or she is in control of pacing choices.

Description
(Example type: Original)
Laser fence dilemma. Imagine a situation where the player is in trouble be-
cause they are accosted by an ever-larger group of enemies, perhaps an ever-in-
creasing zombie horde. Eventually the player has no choice but to run for safety
and ends up racing into an abandoned prison complex. The player is just able to
activate its defenses before the zombie horde gains entry. The defenses consist
of a fence made up of four laser beams that repel anybody trying to enter or leave
the complex. The player is now safe, but also trapped. The same laser defenses
also lock a number of zombies within the compound. The player has to figure
out how to disable the lasers, yet doing so will unleash the zombies, both in the
Puzzles 321

compound as well as eventually the zombie horde outside. In effect, the player
has to solve a puzzle.
We can break this situation down into sections (sub-puzzles) where the
player needs to figure out how to disable all four individual lasers one by one. In
addition, we show the player a series of weapons lockers with clearly visible, use-
ful weapons or items, protected by individual laser beams that correspond with
those in the fence. Disabling one laser means unlocking a new weapon locker,
containing weapons or items that allow the player to deal with the unleashed
zombies.
It is now up to the player to decide which lasers to disable first and when to
do so. In practical terms, the pacing of the setup is in the hands of the player, but
with moments of actions when the player has to act and accept the consequenc-
es of their actions. Eventually a choice has to be made when to disable the final
laser, which makes the entire fence disappear and lets in the zombie horde.

Further Notes
The specific solutions in this scenario don’t really matter that much, that will
be up to the level designer based on the circumstances of the level or the game
he or she is working on. What is key to this setup, however, is that the level de-
signer has a number of great opportunities to affect pacing , while leaving the
final choice with the player.
Artificial Intelligence
15
A s noted earlier in this book, it is clear that interaction with artificially in-
telligent agents in levels is an important factor in level design. This is a
statement that is easily made, but the implications of that statement are not as
easily grasped. What do we actually know about Artificial Intelligence? It is an
incredibly rich area of scientific and academic interest, to the point where it has
spawned whole new areas of research and application. Before we can talk about
formulating theories and techniques for implementing level-design interactions
dealing with artificial intelligence, we should have a look at the field as a whole.

Concept

What Is Artificial Intelligence?

Some Notes and Concepts
Many a book has been written that tries to answer that question, so I am not going
to pretend that this chapter will offer more than the briefest glimpses into the sub-
ject matter. First and foremost, I would like to make an important distinction be-
tween two types of artificial intelligence: “true” artificial intelligence and approxi-
mated artificial intelligence. These are not formal definitions from the field, but are
rather ways of looking at some of the areas of AI that people are working on.

True AI
First there is the romantic notion of scientists finding the magic spark necessary
to create a truly intelligent self-aware artificial life form. This is quite close to the
concept of artificial life, and it is an attempt to solve all the problems of artificial
323
324 Designing Gameplay

intelligence in such a way that the outcome is a creature or construct that pos-
sesses actual intelligence. This kind of AI covers the popular meaning of the term,
and as such it frequently pops up in films and books and other popular media.
However, it is just a subset of the field as a whole, along with many other areas of
AI that are equally relevant to the concept.

Limited AI
Then there are the myriad attempts to create or simulate aspects of intelligence
or life, and ways to improve on them. This in reality is where most of the focus
of current AI work can be found, be it in academic, commercial, or even military
terms. This area suggests that perhaps we cannot create true AI at this point in
time, but we can work on all the individual requirements that have to be met
for it to work in the future. This chapter will show that surprising advances have
been made in that respect.

Focus
There are many aspects of AI that are of immediate consequence to the general
subject matter of this book. Specifically, there are areas of AI that deal directly
with the same issues as those that level designers deal with when implementing
AI in their levels. In fact, game AI is a respectable field in its own right, even
though others don’t always acknowledge that.
So, to avoid getting lost in the immensity of the fields of AI research and ap-
plication, I want to highlight some specific areas that I feel are most relevant to
level design and game AI, starting with the infamous Turing Test.

The Turing Test

Alan Turing1 famously said:
If a machine acts as intelligently as a human being, then it is as intel-
ligent as a human being.
This statement acknowledges the idea that for a machine to have human-like
intelligence, it must be able to reason or behave in such a way that its intelligence
is recognizable by other human beings. This assertion is often referred to as the
Turing Test. Since human beings cannot resist an impossible challenge, the Tur-
ing Test has become one of the most popular pursuits in the field of AI. It has
1
The famous British mathematician. See http://www.scienceray.com/Technology/
Applied-Science/Artificial-Thought-A-Virtual-Being.514325.
Artificial Intelligence 325

lead many people to an approach where they try to create machines that display
behavior that is indistinguishable from that of human beings.

Talking Machines
One of the directions in which to take the Turing Test is to create machines with
the ability to converse believably with human beings. The best test for this con-
sists in creating a situation where the AI converses with a human being with-
out the person being aware that the conversation partner isn’t human. This may
seem like science fiction on the surface, and indeed the whole quest for AI has
the ring of the impossible to it, but this has never stopped people from trying. It
is remarkable, therefore, that with regard to the above version of the Turing test,
scientists have achieved very good results relatively easily.

Eliza
To some degree this feat has been performed many times already, sometimes to
the dismay2 of researchers. An early example of this is creation of Eliza by Joseph
Weizenbaum at the Michigan Institute of Technology in 1966. Eliza is a com-
puter program that masquerades as a psychotherapist. People can talk to her via
a computer interface, typing out and reading questions and answers. I had a chat
with “her” myself and this is what came out of it:
>> Hello.
How do you do. Please state your problem.
>> I need to finish writing my book on level design
Your book on level design ?
>> Yes, I am currently writing about artificial intelligence
You seem to be quite positive.
>> Well, yes. I hope it will be of use to many people.
You are sure.
>> I can hope!
You say you can hope ?
>> Yeah, one can never be sure with these matters.
But your book on level design .
>> Yes sorry I got distracted. I need to know more about you Eliza
I see.
>> I hope you don’t find it rude if I ask you if you are intelligent?
(From a java applet that faithfully reproduces Eliza’s code here: http://
www.chayden.net/eliza/Eliza.html)

2
Dismay at the gullibility of human beings.
326 Designing Gameplay

I won’t reproduce our entire conversation, but while playing the role of
somebody who could not have known3 that Eliza was artificial we kept up the
conversation for a while. Not bad for a little program that does some simple lan-
guage parsing based on a parody of a psychotherapist’s conversation style.
It is fairly successful because the psychotherapy setting creates patterns of
expectation that enhance the validity of certain answers. Normal conversations
may not go like this:
>> Hello.
How do you do. Please state your problem.
>> I am anxious.
Is it because you are anxious that you came to me ?
>> Yes, I am told you can help me
You seem to be quite positive.
>> Well, maybe on the surface, but I am quite sad
You don’t seem quite certain.
>> Ok, well I am really quite sad.
I am sorry to hear that you are sad.
>> Can you help me?
You believe I can help you don’t you ?
>> Yes, that is why I am here!
As long as the fiction of the session doesn’t get too disturbed this seems a re-
markable success. It is easy to be fooled by this conversation as the answers con-
form to the expected response from a real psychotherapist. Now this may not
be a very good indication of actual AI, but it is a startling demonstration on the
tendency of people to make sensory input fit their patterns of expectation. The
brain is quite happy to gloss over discrepancies to a certain degree. It will only
go so far however.
>> Hello.
How do you do. Please state your problem.
>> Eliza please explain to me how you work.
Why do you ask ?
>> Because I am writing a book about level design and a chapter in it is
about the Turing test
Is that the real reason ?
>> Yes
You seem to be quite positive.
>> That is not relevant
I’m not sure I understand you fully.

3
Which I think is respectful of the true spirit of the test.
Artificial Intelligence 327

>> Can you explain how you convince people that you are intelligent?
You believe I can explain what I convince people that I are intelligent
don’t you ?
>> That sentence makes no sense
Are you saying no just to be negative?
>> What?
Does that question interest you ?
>> You don’t seem intelligent to me
Why do you think I don’t seem intelligent to you ?
As willing as the brain is to help the illusion along it still doesn’t take much for it
to shatter completely. Huge improvements have been made however since Eliza
was created in 1966. Today’s “Chatterbots” are employed with serious levels of
success in all kinds of settings.4
The original Eliza was still a huge success in its own right, to the point where
professor Weizenbaum was so disconcerted by the ease with which it was pos-
sible to fool people that he spent a considerable amount of time exploring the
subject from different angles. Amongst other things this led to his book Comput-
er Power and Human Reason: From Judgment to Calculation in which he further
examines the tension between the mechanical and human elements within this
field.
Despite Eliza’s tendency to fall over fairly quickly when approached out-
side of its limited area, it is still a shocking success that has led many people to
re-examine their relationship towards AI and human beings’ reactions to (semi
intelligent) machines. The conclusion remains that it is relatively easy to con-
vince people, or rather to fool people, to a certain degree, that the entity they are
dealing with showcases intelligence it does not actually possess. As we have seen
throughout this book, the mind fills in the blanks, which is a phenomenon that
can both enthuse and scare, depending on its implementation.

Independence
To avoid the intelligent machine going off the rails completely, a certain amount
of independent reasoning is required. It has often been argued that for real artificial
life to be convincing, it as to have the ability to act independently of its creators.
Whatever application or future there would be for an AI construct, it would be
much enhanced if it could still function independently, especially if some kind
of intelligent reasoning is part of the process. Independence of this kind been
approached by some of the most famous thinkers of the last century.
4
See Simon Laven’s website dedicated to chatterbots: http://www.simonlaven.com/.
328 Designing Gameplay

Von Neumann Machines

John von Neumann (1903-1957) was an incredibly influential mathematician
whose work has had significant impact on fields as diverse as nuclear power,
quantum physics, game theory, economics, computer science, and most applica-
ble to us, artificial life.5 Von Neumann theorized the concept of independent self-
replicating cellular automata or, to put it in more easily digestible terms, machines
that can create their own offspring independently. Such machines are more pop-
ularly known as Von Neumann machines. Von Neumann’s work was theoretical;
he never built self-replicating machines (although Lionel and Roger Penrose
later captured the principles of this concept in simple mechanical models made
of plywood). However, Von Neumann machines capture two important con-
cepts in artificial intelligence studies: procreation and independence. Most life
as we know it is able to make autonomous decisions and is able to safeguard its
continued existence through procreation. These are attributes that are hard to
achieve in practice.

Freeman’s Astrochicken
Another famous thinker and mathematician who has contributed to related
areas is Freeman Dyson. Professor Dyson’s work has been as influential as von
Neumann, and the breadth of the subject matter he has written about is impres-
sive. He is mostly known in popular circles for the concept of Dyson spheres,6
but the concept most appropriate to this chapter is his imaginary Astrochicken.
This is a theoretical space exploration robot that takes the concept of von Neu-
mann’s self-replicating automaton and places it in the context of space explora-
tion. Astrochicken is a small spacecraft that hatches from an egg laid in space by
a regular spacecraft. It then develops extremities that help it gather solar power
and uses this to travel to explore planets and other celestial bodies of interest. It
would even have the ability to take “nutrients” from any atmosphere it would
encounter in order to stay operational.
The point of Astrochicken isn’t just that it can act independently of its hu-
man creators, but that it can flourish and explore for a reason. It is given a goal,
a purpose in life for which to use its intelligent attributes. Were one to observe
a colony of space chickens of this type, one would see an independent society of
at least limited artificial intelligence. This independence matters greatly, as it is
needed in situations where the machines have to act autonomously.
5
Specifically, the field of cellular automata.
6
Dyson Spheres are a theoretical concept. They describe a sphere surrounding a solar
system in order to harvest the star’s energy in an efficient way.
Artificial Intelligence 329

Mechanical Life
“The incredible adventures of Astrochicken in space!” is a fascinating concept, but,
as already noted, it is also purely theoretical. Where are we now then when it
comes to the actual construction of artificially intelligent machines? Have we
made any serious steps towards this goal?

Vaucanson’s Mechanical Duck

There have been some interesting attempts in the past to create convincing me-
chanical AI’s. Jacques Vaucanson created a mechanical duck as early as 1738 that
at the time was considered very lifelike (see Figure 15.1). It even had a rudi-
mentary digestive system of sorts. It could be fed and would deposit fake drop-
pings to show that it had “digested” its food. It is important to note that in our
technologically sophisticated times the duck may seem primitive, but in the time
it was built, it was considered a sensation. Again, we see that an approximation
of intelligence elicits strong reactions in people, especially if visually based on
biological principles.

Figure 15.1. Jaques Vaucanson, Canard digérant, from Le monde des automates
(1928).7

7
“A postulated interior of the Duck of Vaucanson (1738–1739) by an American ob-
server,” Wikipedia, http://en.wikipedia.org/wiki/Digesting_Duck, 2009.
330 Designing Gameplay

Space Exploration Machines

Some of our most successful attempts to create independent mechanical AIs
come from necessity. In the case of space exploration, we require probes and
such to do our work for us, since we cannot function that well in a vacuum our-
selves. Space exploration isn’t entirely theoretical after all; we regularly launch
space probes and satellites. Some of the machines we launch are expected to be
fairly autonomous for long periods of times. The Voyager missions are a good
example of space probes whose mission life has been extended due to the fact
that they continue to operate with minimal human interference.
Research in this area is very important, as we cannot predict all the challeng-
es that lie ahead when it comes to missions into the unknown.8 The developers
of the failed Mars expedition of the Beagle 29 probe are sure to appreciate that.

Robot Toys
Space machines are impressive pieces of engineering, but they are incapable of
convincing anyone on a level of human interaction. Surprisingly, we have seen
some impressive results in that area in the form of everyday commercial endeav-
ors, namely advances in the toy industry. The clearest example of this can be
found in robot toys, which combine sophisticated robotics, rudimentary AI, and
human psychology. Much can be learned about human behavior by studying the
successes in this area, and the ability of some of these toys to create feelings of
empathy in the people who interact with them.
Sony’s AIBO (Artificial Intelligence Robot) is quite capable of generating af-
fection, even though it is not convincing as a dog. Furby and Pleo are two further
examples that create serious amounts of empathy, especially from children.
Furbies create a mystique of their own through their non-human language,
Furbish, and are cuddly to the touch. They imitate several human emotions and
can converse with other Furbies. Pleo is shaped like a dinosaur, is capable of
some surprisingly lifelike movement, and reacts to being petted and stroked.
Both robotic toys are commercially very successful. Although nobody will mis-
take them for true life forms, they nonetheless make people react to them as if
they were alive.
Having recognizable emotions or shapes that create certain expectations (if
it looks like a duck and talks like a duck) has once again shown to be a potent
means to convince people of intelligence where there isn’t any.

8
It is by definition the unknown; at best, we can expect the unexpected.
9
See http://www.beagle2.com/index.htm.
Artificial Intelligence 331

As before, however, the visuals or the setting can only take the illusion so far.
Eventually, real life will create a scenario that intrudes on the believability of the
construct and shatter the illusion of intelligence. And therein lies an interesting
proposition: if real life causes the illusion to break, why not remove it as a factor
entirely? Why not create virtual life?

Virtual Life
There is a way of getting around many of the bothersome engineering prob-
lems encountered in the physical construction of artificial life. We can create a
world that is predictable and completely controllable by its makers. We can even
change the rules of the physical universe and write our own laws that govern
our environment. This is done through the creation of computer-based virtual
worlds, and as a logical extension, virtual life.

Evolutionary AI and Virtual Life

Scientists have been creating or capturing elements of virtual life in computer
simulations and habitats for a long time. This has been especially important in
the field of evolutionary biology, which used to be hampered by the fact that it
could only study the effects of evolution, and not evolution itself, since evolu-
tion on earth occurs too slowly to notice. This ended in the sixties, when people
started to capture and accelerate evolutionary principles in computer models.
One of the early pioneers of this approach was John Henry Holland, who wrote
a genetic algorithm that allowed a computer to run a program that incorporates
basic aspects of artificial life, like self-replication.

Tierra
The field really took off in the early nineties, when people were starting to use
similar approaches from different viewpoints, aided by faster and more accessible
computers. A notable and influential example is the work done by Thomas Ray,
an ecologist who became frustrated with traipsing through rainforests to study
evolution and ecological systems. He decided to create a computer program,
which acts as an ecosystem for artificial life forms and incorporates enough at-
tributes and environmental data to form a basic Darwinian system. The program
he wrote is called Tierra,10 and it can be downloaded and run to this day. Tierra
is not much more than a program that allows virtual creatures to live, die and
replicate, with mutations, in computer memory.
10
See http://life.ou.edu/tierra/.
332 Designing Gameplay

To quote a Tierra webpage:

Life on Earth is the product of evolution by natural selection operating
in the medium of carbon chemistry. However, in theory, the process of
evolution is neither limited to occurring on the Earth, nor in carbon
chemistry. Just as it may occur on other planets, it may also operate in
other media, such as the medium of digital computation. And just as
evolution on other planets is not a model of life on Earth, nor is natural
evolution in the digital medium.
The Tierra C source code creates a virtual computer and its Darwin-
ian operating system, whose architecture has been designed in such a
way that the executable machine codes are evolvable. This means that
the machine code can be mutated (by flipping bits at random) or re-
combined (by swapping segments of code between algorithms), and
the resulting code remains functional enough of the time for natural
(or presumably artificial) selection to be able to improve the code over
time.11
Even with these simple parameters, interesting behaviors evolve or develop.
Creatures evolved into parasites if their program length was too short to allow
procreation (latching on to other longer programs in memory to do so), and
hunter-prey cycles between the parasites and host programs began to appear.
From an ecology research perspective, this turned out to be very interesting in-
deed, and the results were accepted as impressive in many ways. The results are
easy to study and can be reproduced easily, due to their digital nature.

Karl Sims
Another good example of evolutionary biology and AI meeting in virtual life
simulations is to be found in the work of Karl Sims. He has created a number
of Darwinian “contests”12 for virtual creatures, where they are placed in a vir-
tual environment and programmed to compete with each other for certain re-
sources or abilities. For example, two creatures on land (including gravity and
friction) have to compete to get to a block of “food” as quickly as possible.13 The
winner gets to evolve to a next generation, thus being rewarded for being fitter
than its opponent. Sims’ creatures use a virtual brain of sorts, and virtual limbs
and muscles that are subject to a process of evolution to define their shape. The
resulting creatures are fascinating to behold. Not all evolved creatures end up
11
“What Tierra Is,” http://life.ou.edu/tierra/whatis.html, 2004.
12
See http://www.karlsims.com/papers/alife94.pdf.
13
An essential text can be downloaded from his website: http://www.karlsims.com/pa-
pers/alife94.pdf.
Artificial Intelligence 333

using solutions recognizable in earth’s biology, although many do. But almost all
the creatures appear lifelike.14

Virtual Life in Entertainment

The examples given so far mostly explain the scientific viewpoint with regard to
virtual life, but there is just as much to be learned from commercial or entertain-
ment expressions in this field. Throughout the years, we keep seeing virtual life
being used in video games and related industries, sometimes with astounding
levels of commercial success. Many of these are interesting from a research point
of view.

Tamagotchi
Bandai’s Tamagotchis were at one point some of the most popular pets in the
world. Launched in 1996, they sold at ridiculous rates.15 The “original virtual
pet” was so successful partly because it managed to create real empathy in the
children who owned one. A Tamagotchi is a tiny portable virtual chicken in an
egg with an LCD screen. It demands constant nurturing and care or it will die
of neglect. Kids have to feed it, play games with it, and in later versions can let it
communicate with other Tamagotchis.

Nintendogs
The incredible success of the Tamagotchi is no fluke. It has been repeated suc-
cessfully several times, including more recently in Nintendo’s puppy simulator
Nintendogs for the Nintendo DS.
Nintendogs quite adroitly manages to portray virtual puppies and give the
player several ways of interacting with them in a context of care and responsibil-
ity. The adorable puppies can be exposed to voice commands via the handheld’s
microphone, so they can learn to do tricks or react to their names. You can play
with them directly and show affection through “physical” contact. What really
makes this a successor to the Tamagotchi, though, is that they need regular at-
tention or will suffer negative consequences. They get dirty, for example or feel
neglected and get sad. Nobody wants to treat puppies poorly. Just as with Tama-
gotchis, a golden vein of addictive protective gameplay has successfully been
mined with the Nintendogs.
14
Beautiful examples of his work can be found on his website here: http://www.biota.
org/ksims/blockies/#video.
15
Over 40,000,000 by July 2005, according to some: see http://www.ubergizmo.
com/15/archives/2005/07/bandais_tamagot.html.
334 Designing Gameplay

Little Computer People,The Sims, and Virtual Girlfriends

This principle doesn’t only apply to fantastical computer creatures or animals.
Not surprisingly, there are many examples of similar games based on the imita-
tion of human behavior. This goes all the way back to the exploits of Little Com-
puter People in the 1985 Activison game. It was so successful that it was made for
several classic formats, including the Commodore 64. It offered a side view of a
virtual house in which a computer person moves around and goes about his or
her business. Players are approached completely in character as an onlooker, in-
dependently from the virtual person, although they can interact with each other
in several ways.
This concept was expanded on and taken to dizzying commercial heights
by Will Wright and his team through the creation of The Sims. At its core, the
game provides the same concept as others, but the behavior of the AI entities
in the game, and the levels of interaction awarded to the players, are massively
enhanced. The Sims has become one of the most successful game series ever and
has crossed the boundary into mainstream acceptance. Interestingly, there are a
lot of female players of the game, without it being specifically targeted at a female
demographic.
Having a large female player base is something that cannot be said for the
mostly Japanese phenomenon of the virtual girlfriend.16 These appear in dedi-
cated programs in which players try to keep their virtual girlfriend happy, to
more elaborate dating SIMs that try to encompass the whole spectrum of the
dating experience, from the wooing stage to relationship issues and more. This
sub-genre of virtual life may be not as intellectually inspiring as some others,
but in its practicalities there are as many lessons to be learned as something like
Eliza.

Concept Summary
As this chapter shows, there are many aspects to artificial intelligence and arti-
ficial life. All are interesting in their own right, and a surprisingly diverse set of
approaches to understanding or even creating artificial life has evolved.
“True artificial intelligence” has not yet been achieved, but there are a large
number of projects that have been able to imitate certain aspects of artificial
life, and there are myriad examples of largely independent constructs with ru-
dimentary reasoning powers. The elements of imitation and independence may
of themselves not provide real AI, but they are instrumental for showing that it

16
Maybe somebody will create a virtual boyfriend soon.
Artificial Intelligence 335

is possible to fool people into believing there are intelligent processes at work. It
is especially successful when combined with the ability to create empathy or so-
licit other strong emotions from the people interacting with the AI. Even when
people rationally know that they can’t possibly be interacting with an actually
intelligent artificial life form, they still react to it as if it were one.
This is a crucial achievement and leads to the creation of believable virtual
“life,” a concept central to video games and therefore to level design.
Pioneers of this dark art can be found all trough the history of video games.
This kind of virtual life may not be driven by scientific goals per se, but it does
employ scientific principles. The field has been incredibly successful in learning
about and applying psychological principles that allow people to interact and
empathize with virtual creatures.

Theory

Artificial intelligence is a big part of almost any video game ever made where the
game controls or simulates behavior other than the player’s. It is interesting that
an activity so much associated with non-serious matters has so much experience
dealing with real and practical aspects of successful AI application.
This naturally affects level design in many ways, psychological and game-
play-specific, and leads us to consider the subject from a number of angles.

AI and Level Design

It is probably clear by now that there is much overlap between AI for games and
AI in science or other fields. If we take the Turing challenge for example, admit-
tedly a tiny focus, we see huge similarities between the separate fields. This is
especially true in its focus on trying to convince participants that they are deal-
ing with real human intelligence. In other areas, the likeness may be less clear.
Games don’t need to create viable independent spacecraft to be sent away for
many years to operate outside of the protection of human guidance. Yet it would
be nice if game AI entities could deal with a number of situations like this and
convincingly react to diverse player stimuli.
It seems a good idea, then, to identify some of the goals that are specific to
level design, as well as to see where they require unique solutions and where we
can use ideas from traditional science.
336 Designing Gameplay

Basic AI Goals for Level Designers

Let’s start out by organizing our thoughts around some clear fundamentals. In
most cases, AI in level design fulfils two purposes: to assist gameplay scenarios
and to enhance immersion.
The two often feed off each other each other and can be hard to separate.

Assisting Gameplay
There are of course many kinds of AI present in video games. We can have AI
opponents in a real-time strategy game, or competing race car drivers in a racing
game. Some of these types of AI fall under the category of general AI and aren’t
necessarily relevant to this book. But there are many instances where the level
designer has to take some control over the behavior and/or placement of spe-
cific AI entities. When this occurs, we are talking about something of enormous
importance to the final success of the levels.
This kind of use of AI agents is what most level designers first think of when
discussing AI implementation in level design. AI becomes a useful tool and
mechanism for creating all kinds of level design and gameplay scenarios. To do
this well, we need to make sure that we have sufficient tools at our disposal and
understand the ways in which we can use them.
What is the psychology of creature placement? How do we make AI be-
havior readable? Or scary? How do we keep AI behavior from becoming too
predictable? These are all classic questions that level designers constantly have
to deal with.

Enhancing or Protecting Immersion

In tandem with the more practical gameplay aspects of a level, AI also has enor-
mous impact on the overall immersion in the game. This is especially clear in
games that require some kind of world building by the level designer, and in
games where it really matters that the player be immersed in the game world. If
it is important that a player buy into the game world through immersive mecha-
nisms, we will have to take extra care with the implications of our AI usage.
No matter how carefully we have crafted a convincing and immersive expe-
rience for our levels, it can all be for naught if the AI presence can’t keep up or
even sabotages the effort. Imagine a game that is all about employing difficult
strategic tactics and maneuvers to get close to a key AI enemy, which is terrible
at path finding and gets stuck on every bit of scenery in the level while uttering
the same phrase over and over again.
Artificial Intelligence 337

Or imagine something as simple as disturbing a flock of birds perched on a

wire when you approach them. The birds settle on a different perch, but return
to the original one when you have moved far enough away again for things to be
“safe.”
When designing levels, the immersive aspects of AI must be taken into
account, as well as the play-oriented ones. We have learned from the previous
chapter that there are many things that help determine the functionality and hu-
man response to AI entities. So, let’s see if we can translate these things to level-
design considerations. The following sections all discuss techniques or areas of
interest that deal with direct gameplay implementation or the perception of the
AI by the player. Sometimes they deal with aspects of both.

Believability and Closure in AI

As with the Turing Test, we need our game AI entities to represent some level
of believability. They need to be natural inhabitants of the spaces we design and
not feel so out of place that they give away the artificiality of the environment.
Unsurprisingly, and unfortunately, this is actually quite difficult to achieve, due
to the limited level of sophistication that we have reached with regard to creat-
ing believable AI behavior. We should not despair, though, since this problem is
even worse outside of games, and we have some great techniques at our disposal
nonetheless.

Allowing the Mind to Fill in the Blanks

Human imagination is a powerful ally to have in the arts, and it is something that
can be especially helpful when dealing with the subject of AI. One aspect where
this really counts is when the mind engages in closure. As mentioned in Chapter
13, closure is a fantastically important aspect of any artistic endeavor where im-
ages or concepts need to be resolved among themselves. What I mean by that is
that anywhere there are empty spaces to fill in—between the panels in a comic
book, between cuts in a film, between chapters in a book and so on—the artist is
using his or her skills to tempt the mind of the audience to resolve those spaces
into an understanding that fits the artwork. Take for example a scene in a movie
that shows a hand holding a match, followed by a scene showing a burnt-out
house. The space that has to be filled between those scenes is enormous, yet
without any problem, the audience will resolve the two disparate scenes into a
logical narrative that leads them to conclude that the match was used to light the
fire that burned down the house.
338 Designing Gameplay

A similar thing can be made to occur in video game AI, and in the way we
use AI as level designers. We have already seen in the example of Eliza the virtual
therapist that this goal is very much achievable.

Goals and Motivations

In many cases, we attribute the existence of goals and motivations to the exis-
tence of some form of intelligence. Non-intelligent entities have no needs and
wants; at best, they just react in a rigid, unintelligent way to set parameters. Some
of the programming involved in game AI deals with giving the AI entity some
simple needs and reasoning ability. So, level designers must choose which needs
have to be acted upon and in what way that will be done. This is something the
level designer needs to be aware of, as it will tell them how the AI entity will react
in specific situations.
Armed with this knowledge, we can enhance these core abilities by setting
up situations that showcase them to the player or even exaggerate the existence
of these motivations and goals.
Thus, we can set up an event that shows a robot fleeing a fight when it runs
out of bullets and retrieving new ammo elsewhere. Or we can engineer a situ-
ation where kindness to an animal will make it stick close to the player, while
cruelty will make it attack.
Many scenarios can be thought up, but it is helpful to demonstrate to the
player that these AI entities have goals and motivations that they act on.

Complexity
Another indicator of intelligence is the ability to deal with complex situations,
or the ability to show a complex range of strategies to cope with diverse chal-
lenges.
One (low cost) way to present this is by creating scenarios that hint at a be-
havioral complexity that simply doesn’t exist. Behavioral complexity is generally
seen by the audience as a sign of intelligence: if we convince the audience that
the AI behaves in a way that is inherently complex, then we convince the audi-
ence that the AI is inherently intelligent. We can also hint at deeper complex-
ity by showing goals and motivations that are completely independent from the
player’s actions. This hints at an autonomous world in which the AI entity has its
own role to play.
We can either show off real complexity in explicit ways, or we can simulate
complexity by manipulating the player’s perception of the AI in question.
Artificial Intelligence 339

Explicit complexity
Level designers often do not realize how complex AI code can be, and how hard
it is to program entities that can deal convincingly with the challenges they have
dreamed up. But once we recognize this, we can show off this depth through our
level designs in ways that make the levels much better.
Explicit complexity is the kind of complex behavior that automatically
emerges from the AI entity’s programming and reactions to outside stimuli.
While we have little control over the programming, we can still manipulate the
environment and the context and challenges in which gameplay occurs.
For example, sometimes it is possible to dress up a small number of game-
play scenarios in such a way that they keep appearing fresh and original. We can
hark back to some of the principles mentioned in the section on “Object-Ori-
ented Level Design” in Chapter 4 and reuse the AI-based challenges in new and
complex ways.
In my previous examples of Tierra or the Karl Sims “contest” I have shown
that sometimes a few simple parameters can lead to incredibly complex results
and even lifelike behavior. The simple goals shown in those examples feature
situations dealing with basic abilities, terrain or environment rules, a need to eat,
adversaries, and so on. These are very similar to factors that we have to deal with
in level design:
• Basic abilities become character/entity controls.
• Terrain/environmental rules become the game physics.
• A need to eat becomes resource management.
• Adversaries are game opponents.
Throw these factors in a blender, and we can create very complex scenarios com-
ing out of very basic factors. Even though individual components can be simple,
we can make our levels into interestingly complex systems, where small changes
in specific parameters can have interesting and even chaotic results.
Perceived complexity
Perceived complexity is the behavior as the player interprets it. This is a very
powerful concept that we can play with to enhance the appreciation of apparent
sophistication of the AI in question.
As before, we can reuse AI-based level design scenarios, but this time with-
out doing much to actually add complexity to the AI in question. For example, if
our AI creature is no good at dealing with diverse types of terrain, we just make
sure it is never in a situation where it has to do complex path finding. Instead, we
put it on balconies, in towers, on top of a roof, and so forth, but we give it access
to different guns and catchphrases.
340 Designing Gameplay

Again, as long as the player believes that the responses are complex, that’s
good enough.
We can “cheat” by suggesting all kinds of complex reasoning that in fact isn’t
very complex at all. Here are some ideas on how to do that:
• Give the AI entity knowledge that does not come from its own senses.
• Script it to explain its reasoning in a way that sounds more intelligent than
it really is.
• Adjust very simple basic attributes of the entity in a way that changes its
abilities drastically, without altering its reasoning.
• Place the entity in areas of the level that make it appear able to cope with
all kinds of environments.
Others can be found, as well.

Dangers in seeking complexity

There are a number of things we need to look out for when it comes to seeking
complexity.
In many cases, our dealings with in-game AI are based on encounters that
provide challenge and adversarial gameplay. The player’s skills are often tested
against the challenges presented through AI entities. What needs to be remem-
bered, though, is that the AI needs to serve the game’s higher goals, and not
become a goal in its own right. There isn’t much point in seeking to create in-
credibly sophisticated AI to use against the player if that AI opponent is too
smart for the player. The same is true if it has internal workings that are fairly
intelligent but unclear to the player, which prevent the player from devising en-
joyable play strategies.
We also need to make sure the AI is deemed to be fair. Any time we embark
on a path of tricks and smoke and mirrors to make the AI look smarter than it is,
we run the risk of the player finding out that we are “cheating.” This is something
that needs to be avoided, because once the player thinks the game is unfair, he or
she will not trust it anymore and the unwritten entertainment contract between
game and player is broken.

Imitation
We can also try to create AI behavior that imitates the actions of intelligent enti-
ties. This is a valid approach to take and offers all kinds of tools for level design-
ers. If the imitation is good enough to fool the player into believing in a high level
of intelligence, we have reached our goal as convincingly as if we had actually
created the intelligence we wish to portray.
Artificial Intelligence 341

We can engineer situations where the AI’s actions mimic processes that we
KNOW to be intelligent. By doing so, we can make the player assume that ac-
tual intelligence is being used when it isn’t. This is a method that is especially
useful in cutscenes and scripted events, where we have more control over cir-
cumstances.

Independence
Independence can be hinted at by showing the AI performing autonomous tasks,
even mundane ones. After all, if an AI can lead an existence where even the mun-
dane is acted upon, it must surely have a high level of independence. This is often
used to great effect in games where we can observe the AI without interference,
for example in stealth games where the player spends a lot of time in hiding,
studying the behavior of AI entities on patrol.
Most stealth games have the AIs perform quite arbitrary or random inde-
pendent actions. They stop at a vending machine, chat to another guard, whistle
a song, and so on. These actions are not accidental; they are specifically put in
to showcase that the AIs have a “life of their own,” even though quite often these
actions are quite unintelligent and are just hard coded to occur at specific loca-
tions. Little touches like these can make a level stand out, sometimes with mini-
mal effort.

Empathy and Antipathy

Alternatively, we can try to create an emotional response in the player that is
strong enough for a reaction that makes the player attribute all kinds of mental
capabilities to the AI. This can be subtle or brutally direct; the trick is to make
the actions emotionally powerful.
An AI enemy is just a bit of code, but say for example we show a human NPC
club a seal to death or slap a defenseless person in the face. Chances are that the
player will associate all kinds of negative human emotions (and therefore intel-
ligence) to this creature. The opposite principle works equally well; empathy can
be just as powerful as antipathy.

Gameplay and Environment

Considerations
As much as we can try psychological tricks to make player interaction with
game AI as meaningful and convincing as possible, we still need to be able to
342 Designing Gameplay

place the AIs in an environment and use direct gameplay application in a way
that works.

The Right Environment (Context)

So how do we achieve all these things? The tips, techniques, and theories men-
tioned above will certainly help, but it is best to try to put them all into an overall
system of logic, a balanced AI ecosystem that performs like clockwork no matter
what it encounters.
For AI to work well, it is best to try to create a context that allows for ap-
propriate behavior and that plays to the AI’s strengths. To go back to the concept
part of this chapter, we can see this principle at work in the Eliza AI. Eliza was
very successful because it provided an appropriate context for AI interaction
that was conducive to expectations that the AI could easily meet, and that had a
setting that played to its strengths.

Appropriateness
In Eliza’s case, its framing of a psychotherapy question-and-answer setting im-
mediately creates a situation where vague searching answers and questions don’t
feel odd. The participant may think some question or answers are a bit strange,
but will often ignore these peculiarities because the psycho-probing expected in
the session explains it. We assume there are goals and motivations involved that
are appropriate due to the setting.
We can do the same thing within our level designs. By creating an environ-
ment that makes the actions of the AI look appropriate, we can create a much
more believable experience. We can place defensive turrets or robots with lim-
ited AI in a damaged factory setting, and suddenly their “glitches” are given a
believable context.

Fitting the AI abilities

Eliza’s environment is that of the program and another person communicating
through a keyboard and monitor interface. It fits the AI’s limited abilities because
it doesn’t ask it to show any physical representation of itself and allows it to use
the medium it is best suited for: written language, without involving the danger
of spoken language, which it could never pull off well.
In retrospect this is a fairly obvious approach, but isn’t everything obvious
in retrospect? Sometimes the obvious is worth stating, regardless.
In level design it is often of the utmost importance that we make sure that the AI
we employ is used in settings where it operates well and avoids its weaknesses.
Artificial Intelligence 343

• If your game’s AI entities are not very good at path finding, don’t place
them in a complex terrain.
• If they are not very good at dealing with situations up close, don’t throw a
bunch of them in a cramped environment.
• If they don’t have many lines of text to say, don’t make them talk constant-
ly.
I am sure you get the picture and are able to add dozens of such cautions of your
own.

Scripted Events and Cutscenes

Finally, as mentioned briefly before, when all other things fail, we can use brute
force and create prescripted responses that allow us to direct AI the way we want.
This is often done in scripted events and cutscenes and can range from subtle
actions to taking all power from the player and letting AIs do things they have
no business doing.
We can make the AI say things that go beyond its basic vocabulary or tra-
verse environments it can’t deal with in-game. Or we can let the AI use its ac-
tual capabilities, but we assert control over the circumstances and timing of the
scripted event. Either way, due to the potentially intrusive nature of events like
these, we should treat this approach with caution.

Theory Summary
It seems that we can spend as much time dealing with the actual capabilities of
game AI as on managing the perceptions of these capabilities. Both aspects are
indeed valid, and are rooted in sound principles and ideas that can be taken from
existing fields of AI research or implementation. Games also have their own
knowledge to contribute. The mixture of these elements makes for a potent set
of tools to use to improve our levels.
We can make our levels and the AI used in them stand out by using elements
of AI, showing off AI capabilities with the use of real or perceived complexity. We
can demonstrate intelligent independence and showcase the use of goals and
motivations. We can direct feelings through imitation, empathy, or antipathy.
And we can direct entire gameplay scenarios through cutscenes and scripted
sequences.
There are many other examples that could be named, but the previous
ones should illustrate the general principle and they should apply to many lev-
els.
344 Designing Gameplay

Practice

Example 15.1: AI Believability—Imitation

of Emotions
Summary
A classic way to make AI believable is to engineer situations where the AI is seen
to do seemingly intelligent things. One way to do this in a fairly subtle way is to
provide moments when the AI “betrays” an emotionally complex “inner life.”

Game Genre
This technique is suitable for games that allow moderate scripting of AI entities.

Goals to Achieve
• Enhance the believability AI characters.
• Achieve this by making them appear to suffer from human weaknesses.
• Put this in a practical level setting.

Description
(Example type: Original)
When dealing with level design it is all too easy to get lost in the practical needs
with which the AI will be confronted: pure gameplay requirements like the abil-
ity to navigate the environment or its effectiveness as a challenge to the player’s
skills. These subjects are indeed massively important, but at times it is also vital
that the AI contributes to the level’s immersive qualities, or instills an emotional
bond with the player.
There are, for example, many games that require the player to interact with
a helpful sidekick or similar type of friendly AI. It is very helpful if the player is
convinced by the AI’s behavior that its intelligence is fairly sophisticated. It is
even more helpful if the player sees the AI as a creature with a personality, and
not just a tool with which to progress through the level. That way, the user is able
to form some kind of emotional connection to the AI that leads to a more im-
mersive or interesting level.
Artificial Intelligence 345

We can achieve this outcome by hinting at a sophistication that isn’t nec-

essarily there, but imitates sophisticated intelligence to such a degree that the
player accepts it’s there nonetheless. With some minor scripted touches we can
achieve a lot, and there are some excellent existing examples to draw inspiration
from.
Emotional outburst. In Bungie’s Halo: Combat Evolved17 the player is regu-
larly aided by helpful NPCs that help the player combat enemy alien creatures.
The NPCs have many lines of dialogue that portray them as gung-ho marines to
firmly establish them within the genre’s conventions. There are moments, how-
ever, where they go beyond shouting one-liners and perform actions that can’t
help but provoke a response in the player. One such moment occurs right at the
end of a challenging battle, when one of these marines runs up to an already dead
alien, and shoots it again, gleefully yelling at it.
Nothing is served by this scripted sequence in gameplay terms but the play-
er is unlikely to forget such a moment, and the general enjoyment of the level as
a whole is much enhanced.
Humor/fear. Alien vs. Predator 218 is a particularly scary first person shooter. In
one of the game’s sequences, the player and a group of marines investigate an
off-world installation. The marines are separated from each other but they can
communicate through radio. The radio chatter is used by the level designer to
convey NPC dialogue that enhances the sense of personality and intelligence of
the NPCs, as well as adding to the level’s tension as a whole.
Here is an example of the radio chatter:
A scared and nervous sounding marine:
“Is that a boot? Is that a boot!? That better be a boot!!”
A minute or so of silence….
“Hey, it’s a boot!”
The example is very simple but nonetheless very effective in humanizing the AI
in this level.

Further Notes
There are many other human character traits that can be incorporated in a game’s
AI like nervousness, mischievousness, irritation, and so on.

17
Published by Microsoft Game Studios, developed by Bungie Studios, released Novem-
ber 15, 2001.
18
Published by Fox Interactive, Inc. – Sierra On-Line Inc., developed by Monolith Pro-
ductions Inc., released October 22, 2001.
346 Designing Gameplay

Example 15.2: Appropriate

Environments—Ambush Scenario
Summary
To get the most out of AI, it is best to use them within an environment where
they are best able to perform to their capabilities. A classic ambush scenario pro-
vides an opportunity to take that principle even further.

Game Genre
This technique is suitable for games that allow moderate scripting of AI entities.

Goals to Achieve
• Showcase the use of AI in a way that maximizes their impact and minimizes
their apparent weaknesses.

Description
(Example type: Original)
The ambush scenario. All AI entities suffer from weaknesses, and at times
that makes it hard to use them in a way that does not endanger player immer-
sion. They may be bad at navigating complex terrain, or have difficulty acquiring
targets, or have problems choosing from too many actions, and so on. Gener-
ally, that means that we design our levels in such a way that they hide these
deficiencies: the AI is not placed in a position where their inefficiencies shine
through.
We can, however, take this one step further and place the AI in a position
where their inefficiencies are hidden and their capabilities are exaggerated—
especially if done in such a way that is not possible in general gameplay circum-
stances. The idea is to design a level design pretext that allows the player to use
the AI in such a way without destroying immersion.
One such method is the classic ambush situation. It gives the designer a
number of advantages that don’t often occur in regular gameplay:
• The AI movement can be logically restricted to specific locations that
maximize their efficiency.
• The AI can be placed in locations that stop the player from direct interaction
with them.
• The AI can be scripted to perform specific actions.
Artificial Intelligence 347

Take, for example, an ambush in a mountain pass. AI entities can be

placed away from the player, on a higher road, in the mouth of a moun-
tain cave set in a cliff face, on top of a large and steep set of boulders,
and so on. None of these locations are reachable by the player, yet within
the context of an ambush all these locations are explained. Impact can be
maximized through specific scripted actions. A far away AI entity can be
restricted to using long range weapons, the AI entity on the higher road
can run up and down, evading player attacks while returning fire, and the
AI on top of the boulders can throw sticks of dynamite when the player
comes too close. All things that can be achieved with limited amounts
of scripting, and all showing off the AI as capable of a range of actions
requiring intelligence.
This can be made even more effective if we signpost the AI’s intentions to
the player, even if those intentions are faked by us, for optimal believability.

Further Notes
This example features a static set piece, but variations can include a more dy-
namic scenario. A player can be stalked by an AI moving along a higher, unreach-
able, path, and taking specific actions at well-designed intervals.

Example 15.3: AI Believability—

Emphasizing Ability
Summary
The more natural and capable an AI behaves within a gameplay context, the more
it contributes to the success and immersive qualities of a level. We therefore try
to showcase AI in its best light.

Game Genre
This technique is suitable for most games.

Goals to Achieve
• Achieve closure by manufacturing the appearance of
■ complexity,
■ independence,
■ emotions.
348 Designing Gameplay

Description
(Example type: Original)
It is common sense to try to use AI in a way that does not detract from its believ-
ability, for example, by avoiding situations that show up its weaknesses. That kind
of solution is valid and we often will end up taking exactly that approach. It is, how-
ever, a somewhat passive approach, as it is focused on not doing certain things.
An alternative and often more effective approach is to create gameplay sce-
narios and situations that are built around AI strengths. So, rather than avoiding
things the AI is bad at, we end up researching those things the AI is great at and
deliver the most fun for the player. Once we have isolated those strengths, we
can construct gameplay sequences that are purely built around those strengths.
We may end up with a list of good and bad points from which we can start
to draw inspiration. In this example we will look at a robot cat sidekick in an
exploration platformer. Our list may look a bit like this:
AI abilities and noteworthy behavior:
• The AI gets confused by more than five onscreen characters.
• It will always chase mice if they are around.
• It can stand up to small dogs.
• The AI is no good against the game’s larger animals .
• The AI is very good at jumping (better than the player).
• It is very bad at swimming.
• It won’t go outside in the rain.
From this we may derive a mix of skills and behaviors that make us create a
gameplay sequence that involves the following elements:
• few enemies,
• no mice,
• some dog encounters,
• no larger animals,
• a series of platform jumps that the cat can perform but the player can’t,
• no water,
• sunny weather.
Simplistic as this may sound, it is nonetheless a good exercise to go through
when deciding on what ingredients to use for a specific level design situation.

Further Notes
This approach requires the designer to spend a serious amount of time play test-
ing scenarios and circumstances in feature-rich test levels or test beds to be able
Artificial Intelligence 349

to judge the strengths and weaknesses accurately. In a way, it is about “finding the
fun” and stress testing the AI to find circumstances under which its abilities and
impact is at an optimum. This can only be done in the context of representative
gameplay, as gameplay mechanics tend to influence each other and do not occur
in a vacuum.
Challenge
16
W e have encountered the concept of challenge numerous times now through-
out this book. We have spoken of positive and negative challenges in life and
how conquering them can lead to a sense of empowerment. We have spoken of the
willingness of players to deal with challenges if they form part of a reward system, and
we have spoken of formal challenges put on a person in order to test knowledge.
These are three distinct ways of looking at the same word, which is an indi-
cation that it may not suffice to think of “challenge” as a single discreet concept.
This chapter therefore looks deeper into the various aspects that are important
to understanding the concept of challenge as a whole.

Concept

Challenge Revisited
Although we already spent a fair amount of time on the subject of challenge in
the “Challenge and Empowerment” section in Chapter 7, I would still like to
revisit some aspects of this subject. I think it deserves to be approached from
several different angles, since it is so important to level design, and because it is a
complex subject that warrants multiple observations.
To recap1 some of the points made, here is an observation I made earlier:
There are positive challenges. If you go running for an hour every day,
you will become very fit. If you study a language diligently for ten years,
you will become fluid in its use. If your work is better than that of your
colleagues, you will get that promotion. (Well, you should, anyway.)

1
I know it is bad form to quote one’s own text, but it saves you from having to flip back
and forth between chapters.
351
352 Designing Gameplay

And there are negative challenges. Stand up to that bully and he will stop
harassing you. Deal with the tragedy of a loss for a long enough time
and it will hurt less and less. Resist smoking for a year and your cravings
will be gone.
These are all good examples where there is a sense of empowerment, a reward
for meeting the challenge successfully. There are other well-known uses for chal-
lenges as well, for example in an educational context.

Problematic Conceptions of “Challenge

for Challenge’s Sake”
The conclusion that presents itself from all of the examples previously discussed
is that we need challenge in life in order to measure our strength of character or
to test our skills at a certain activity, and so forth. There are many applications for
challenge that are worthwhile and rewarding. While this can be true in its own
right, it often leads to the conclusion that therefore challenge is a desirable goal
in itself. This is dangerous and often untrue.
If challenge were seen as a goal in its own right, than just inserting one or
more new challenges into the equation should increase the value of the expe-
rience. Indeed, many people have done so in the past. Unfortunately, this ap-
proach can lead to serious problems.
To understand this, let’s look at some examples.

Lack of Meaning
When an artist challenges his or her audience, it generally is expected that there
is a reason to do so. If the audience is presented with challenge for challenge’s
sake, it will eventually come to realize that there is no meaning behind the act.
This is a very dangerous thing to court, because once an audience realizes that
it has been put through its paces for no good reason, it loses its motivation to
stick around for the ride. This problem can occur across the spectrum of cre-
ative expression. Challenging but pointless dissonance in a composition will be
experienced as off-key or cacophonous. Harsh and disorientating edits in a film
without serving the content of the shot will just dizzy and upset an audience.

Creating Resentment
A similar problem with challenge for challenge’s sake lies in the likelihood that
it will build up resentment in the audience. If you were to poke somebody in the
Challenge 353

chest for no good reason, that person would probably object. Maybe not the first
time, but do it a lot and even your best friend will start to resent the action. It has
to be understood that if challenge is not placed within a context, it is just hard-
ship. And who wants to be subjected to arbitrary hardship?
An artist and their audience have an unwritten contract. The artist presents
creative output and invites the audience to experience it, even if the experience
can become unpleasant or challenging. The audience will go along with this,
based on this unwritten contract, which states that there is some value or payoff
in doing so. If it turns out that there is no good reason for the audience’s discom-
fort, that contract has been broken.

Desensitization and Boredom

Additionally, there is the danger of just overloading the audience. If there is no
particularly good reason for being harassed by the artwork and the audience
catches on to this, it will either stop taking it seriously or just stop feeling an
emotional connection to the work. Both are clearly disastrous results.

Concept Summary
If challenge is to be incorporated into art and entertainment, it needs to serve a
higher goal.
This is just as true for video games as it is for other art forms. Generally, this
kind of artistic challenge exists to eventually reach a positive outcome. In video
games, challenge is used more than in almost any other artistic medium, because
challenges are a great way to test the player’s skill level, which is an intrinsic part
of level design. In video games, much enjoyment comes from learning how to
use the gameplay mechanics and from being confronted with new and enjoyable
ways to test one’s ability.
Challenge is a formal invitation to the player to overcome adversity through
applying his or her gameplay skills.

Theory
Few words are used so freely, but misunderstood so much in discussions about
level or game design as challenge. Worse yet, people always think that everybody
agrees on the exact meaning, or even worse, that they agree on the right applica-
354 Designing Gameplay

tion of it in level design. The reality of the situation is that almost nothing does
more harm to players’ enjoyment of games than the misuse of the concept of
challenge.

Challenge and Difficulty levels

Challenge as a concept is often used in conjunction with difficulty levels. While
the two can have a relationship, they are not the same thing. Challenge is a for-
mal invitation or demand on the player to overcome adversity. Difficulty, on the
other hand, is much more subjective and simply describes how well the player is
equipped to do so. In other words, a level can offer the same challenge to different
players, and they may have a completely different appreciation of the difficulty
level of that challenge.

Frustration and Boredom

One of the biggest dangers in level design is misjudging the abilities of the player
in such a way that a level becomes too hard or too easy. The emotions that come
into play when we get this wrong are frustration and boredom, and both of them
can severely damage a player’s enjoyment of the game.
For one last time, let’s go back to the diagram of Csiksczentmihalyi’s flow
theory. (See Figure 16.1.)

Figure 16.1. Flow theory diagram.

Challenge 355

The main danger in the implementation of challenge is spelled out right in

front of us. If challenge and skill are not appropriately balanced, the player leaves
the zone of optimal experience and becomes either anxious (frustrated) when
the game is too hard, or bored when the game is too easy.

Frustration
Of the two, this emotion is potentially lethal to the success of a game. Anywhere
in life where an activity is beyond the skill level of participants, they will be liter-
ally frustrated in their ability to enjoy it. If that frustration is too fundamental
and cannot be overcome, participants will become so unhappy that they will
cease the activity.
The same thing occurs in video games. A player who becomes frustrated
with the difficulty level of a game will eventually give up playing all together,
likely for good. Why would someone want to spend time on an activity that frus-
trates the desired outcome, such as an enjoyable or worthwhile or rewarding ex-
perience? Challenges are often used to affect the game’s difficulty level and learn-
ing curve. This can be appropriate in many cases, but when the difficulty level is
too challenging, and the player has no prospect of this changing, the game will
have gone too far.

Boredom
Boredom has its own unique problems and can lead to the same negative re-
sult. Interestingly, one of the justifications for harsh challenges in a level is to
stave off boredom. The argument often goes something like “The game has to be
challenging or the player will become bored. Therefore, we must make the game
more challenging over time or the player will lose interest.” This is a subtle but
serious mistake. As we already established, challenge is simply a tool, no more
than a means to an end. In this case the tool helps avoid boredom. So far so good,
but boredom is not necessarily avoided by escalating difficulty levels. Just upping the
difficulty will not automatically push the player back into the flow channel, the
zone of an optimal game experience. If an activity doesn’t require the use of new
skills or refine the use of current skill sets, its activity will become less interesting
and certainly less rewarding.

Apparent Exceptions
In some rare cases, games actually provide frustration as a gameplay goal. There
are games that propose an almost masochistic test to the player, where it is a given
356 Designing Gameplay

that although harsh and unforgiving treatment will be doled out, in exchange the
player receives a very particular kind of satisfaction that comes from successfully
overcoming the extreme challenges of such games. This kind of gameplay often
occurs in specific taste niches or areas of specialization and is typically some
kind of challenge-based game.

Bullet Hell
An amusing example of this kind of specialized, challenge-based game is a phe-
nomenon found in some types of classical shootemups. The level or mission
design for these games boils down to the player trying to maneuver into ever
decreasing areas of the screen where there are no enemy bullets. This is entirely
acceptable, even expected, in these kinds of shootemups. It becomes an endur-
ance test to players who are already used to being tested by these conventions of
gameplay.
Unfortunately, this also often occurs in inappropriate places. One example
is in third-person action games where the player literally cannot set a foot down
wrong without being shot to pieces or otherwise hindered from playing the
game.

Unfairness and humor

There is a case to be made for recognizing these principles and using extremely
unfair challenges for comedic appeal. This works well to varying degrees. If the
whole game is based on that premise, players will be forewarned and may take
on the challenges with a good sense of humor. A recent example of this can be
found in the indie game I Want to Be the Guy, which prides itself on providing un-
fair and frustrating gameplay in a series of brutally hard and challenging levels.
The game maker says the following in the “how to play” section of its
website2:
This game is very hard. You could find that getting past the very first
few screens is taking you the better part of an hour, or more. Do not
be deterred. It’s supposed to be unfair. Realize this before all else; any
and all situations can kill without warning. Take nothing for granted.
If it appears safe; you can be nearly certain that it is a trap. Understand
that you can only get hit once before you die in a particularly gruesome
manner. Be careful!
The player has certainly been forewarned in this particular case!

2
“I Wanna Be the Guy! Manual!” IWBTG!, http://kayin.pyoko.org/iwbtg/manual.php,
2007.
Challenge 357

Narrow flow channel

It can be argued that these are exceptions to the rule that too high a challenge
leads to anxiety, as these games openly court frustration. On closer inspec-
tion one could come to a different conclusion. These games don’t present a
challenge level that is too high, but one that is appropriate to the game genre
itself.
The zone in these kinds of games is a very narrow flow channel, but the same
principles still apply that we saw before. It is just that the intended audience is
willing to deal with much higher and more specific challenge levels than players
of more accessible games.

Appropriate Challenges
There are two major aspects of challenges in level design that should never be
ignored. As already shown in the flow diagram3, skill needs to be in balance with
challenge. Additionally, the balance definition itself, the positioning of the flow
channel, is subject to considerations of context.

Skill
Skill issues in level design are the easiest to recognize, yet also the most frequent-
ly misjudged. We have already discussed the potential dangers of challenging
the player too much by requiring unreasonable skills. Let’s look at ways that this
danger can be avoided.

Dynamic Difficulty Adjustment

Dynamic difficulty adjustment is the practice of allowing the game to react to
the apparent skill level of the player and adjust difficulty levels accordingly.
A classic example of this would be measuring player death at specific loca-
tions in levels. If a player dies too many times in a conflict area, for example a
shootout scenario, the difficulty of the encounter can be adjusted downwards,
perhaps by making enemy AI shooting less accurate. There are many ways of
implementing this principle, and if done without the player being overly con-
scious of getting a helping hand, this can yield good results. However, there
are some real dangers and disadvantages attached to this kind of difficulty
adjustment.

3
See Figure 16.1.
358 Designing Gameplay

Data limitations
To make the decision to change the difficulty level of the game, the dynamic dif-
ficulty adjustment (DDA) system needs some way of evaluating the need to do so.
Performance data needs to be gathered and interpreted before any change can be
made. For example, the player’s ability to hit targets in a first person shooter or
the number of hit points left after specific encounters, need to be known.
The problem with collecting this kind of data is that it gives us very limited
information on which to base important decisions. What if the player enjoys a
playing style that results in being very close to death most of the time? What if
the player enjoys shooting the scenery as well as shooting at enemies? Since we
cannot gather data that objectively tells us the player’s real emotions and experi-
ences directly, this kind of reactive or passive difficulty adjustment will always be
based on subjective guesses. And that means that the interpretation of the data
can yield incorrect conclusions, resulting in inappropriate adjustments.

Passivity
However, let’s assume that the DDA system gets it right. Take a series of jumps
to reach the exit in one of our levels in a platforming game. The system is alerted
that the player keeps failing the last jump; in response, it may lower the diffi-
culty level by allowing a collapsing tree to create a bridge between the two jump
points.4 To a degree, the DDA has worked correctly, and the resulting interven-
tion has allowed the player to progress through the game. This is all true, but
it did not happen until after the player has left the flow channel and had to be
guided back into it. Because the system can only react to data, actions always oc-
cur after the fact. And levels that constantly have to correct themselves in order
to make an unhappy player happy again are far from ideal.

Jenova Chen–Flow In Games

Thankfully, there are ways in which these problems can be avoided or minimized.
Jenova Chen has done great work in this field by taking the basic principles of
Csiksczentmihalyi’s flow theory and incorporating it into a specific flow theory
for games. His thesis, “Flow in Games,” is a must read for all serious students
of level and game design theory. He has much to say on the subject of flow and
what he calls “player centric dynamic difficulty adjustment.”
So why don’t we give the players choices in a video game and let them
navigate their Flow experience? In order to create a game like this, …5,
4
Yes, I realize this is a very silly example.
5
I removed a redundant reference to an image in the original thesis.
Challenge 359

the game needs to offer a pool with a wide spectrum of activities and dif-
ficulties for different types of players to swim inside. Based on players’
tastes, each individual will choose different choices and work at a dif-
ferent pace to navigate through the game. Once a network of choices is
applied, the Flow experience is very much customizable by the players.
If they start feeling bored, they can choose to play harder, vice versa.6
What Jenova Chen is proposing here is proactive difficulty adjustment con-
trolled by the players themselves. This is an intriguing concept that is worth ex-
amining in more detail.

Player controlled difficulty adjustment

As opposed to trying to design a self regulating difficulty system that tries to read
player behavior or preempt player actions, Chen argues for a system that puts
these choices in the player’s hands. Who else is better able to judge what is too
hard or frustrating or fun? A logical consequence of this philosophy is a game
that has based its core game mechanics on player-controlled difficulty adjust-
ment. This is what Chen and Nicholas Clarke have done with fl0w. This game
has incorporated these choices seamlessly into the core gameplay, and this has
led to a remarkably successful project.7
Fl0w is based on a simple premise: the player controls a primitive marine
organism that has to find food in order to evolve to more evolved states. To do
this, the creature has to dive to depths that are more dangerous than shallower
water. The game therefore allows the player to find a gameplay space that he or
she is the most comfortable with.

Context
Challenge has been shown to not only be subject to personal skill levels, but also
to context. This context comes from the multifaceted dimension of the whole
notion of challenge. We can challenge players to test their shooting skills, or to
affect pacing , or we can give them something by which they can measure their
own strength. All of these add a unique contextual dimension to the challenge in
question, and therefore to the experience of the player. The last boss in the game
presents a different context and experience than an idle puzzle along the way.

6
Jenova Chen, “Flow in Games Thesis,” Flow in Games, http://www.jenovachen.com/
flowingames/Flow_in_games_final.pdf, 2004, p. 13.
7
The flash version of the game downloads numbers in the hundreds of thousands. There
are a PS3 version and a PSP version, and other games based on the same principles are
in development.
360 Designing Gameplay

Challenge as a Level Design Tool

What these aspects of appropriateness highlight is that we should never forget
that challenge is a tool for achieving something specific. It is meaningless as a
goal in its own right, and level designers who claim that all gameplay needs to
be challenging completely misunderstand this basic principle. It is those level
designers who implement challenge for challenge’s sake who commit some of
the worst offenses to be found in gameplay implementation.

Arbitrary Difficulty
One of the common dangers that creeps into designs is arbitrary difficulty that
provides pointless challenge. A good analogy is a lopsided bicycle tire. Once in
a while you see these contraptions at carnivals or in similar places. Sometimes
there is a bicycle with a very lopsided tire, and an obstacle course to be traversed.
In the right context (a few minutes of silly fun), this can work okay; but for most
uses, a bicycle is not improved by making the wheel lopsided. In fact, when you
want to go somewhere, it is incredibly annoying to have a lopsided wheel. Yet
there are plenty of games out there that insist on taking a useful game mechanic
and making it awkward without any good reason whatsoever.

Prohibitive Difficulty
Some games are just too hard for most people to enjoy. They require a commit-
ment unlikely to be deemed enjoyable. Some surpass the skill level of most play-
ers, who will never be able to do certain things in the game. Unless a game is made
specifically for people who seek a challenge like that, this is a serious problem.
In level design this problem often occurs because level designers have spent
so much time playing a game that they have reached an atypical skill level, and
are trying to entertain themselves in their level design. This is a fatal error since
their job is to entertain their audience, not themselves.

Unfair Challenges
This is a very strange one to understand, but it keeps occurring on a regular basis.
Some level designers seem to think of the player as the enemy, and feel their job
is to kill the player as harshly as possible. If you are one of those people, stop that
immediately please!
Gameplay situations can test a player’s skills. No issues there, as that is what
gameplay learning is all about. It is enjoyable to use new skills in interesting ways,
Challenge 361

and good level designs provide players with fun ways to test their skills. Chal-
lenges can also be fun if there is an interesting or enjoyable sense of achievement
to be had, especially if it ties in contextually to the game’s overall objectives.
What is not appropriate to players, though, is to be challenged to do tasks or over-
come obstacles that are presented in an unfair way. This can include situations like
• instant death (unannounced),
• requiring payers to do something extremely hard without training them first,
• obscure nonsensical puzzles,
• extended memory-based gameplay,
• overly harsh restrictions (impossibly short time trials),
• lack of save points,
• one-way doors that close behind the player without warning.
There are many, many more.

Theory Summary
At the beginning of the book, we established that if we are to teach players how
to enjoy a game through our level designs, we also have to test their skills and
progress on occasion. To truly enjoy the fruits of their gameplay journey, players
need to be able to put their gameplay skills into practice and enjoy their mastery
of the abilities required to solve the problems set up by the game.
One of the ways this can be done is by devising gameplay challenges to the
player, against which they can measure their skills. This is something they will be
happy to engage in if the challenge is appropriate and there is some kind of decent
reward. A challenge without a purpose or reward is very adversarial, and while
this can be okay once in a while, it can become quite off-putting to people who
want to enjoy their gameplay time and not be punished for trying to progress.

Practice

Example 16.1: Player Controlled

Challenge Levels—Varying Skills
Summary
A different way of allowing the player to control the challenge level
they face is by catering to different play styles and skills. This allows
362 Designing Gameplay

the player to choose those styles and skills he or she is most comfort-
able with.

Game Genre
The technique is suitable for games that allow for multiple solutions to gameplay
challenges.

Goals to Achieve
• Define challenge by skill type and play style.
• Set up a scenario that incorporates several play styles and skills.
• Allow the player to make a choice between them.

Description
(Example type: Original)
Laser fence dilemma 1. In the practice section in Chapter 14 (“Puzzles”),
I gave an example where puzzles can work as a pacing device and introduced a
hypothetical gameplay scenario based on the player being locked in a fenced-off
prison compound. The scenario was called the “laser fence dilemma,” and it can
easily be adjusted to provide player controlled challenge levels.
The basic setup of the scenario was as follows:
Imagine a situation where the player is in trouble because they are
accosted by an ever-larger group of enemies, perhaps an ever-increasing
zombie horde. Eventually the player has no choice but to run for safety
and ends up racing into an abandoned prison complex. The player is
just able to activate its defenses before the zombie horde gains entry.
The defenses consist of a fence made up of four laser beams that repel
anybody trying to enter or leave the complex. The player is now safe,
but also trapped. The same laser defenses also lock a number of zombies
within the compound. The player has to figure out how to disable the
lasers, yet doing so will unleash the zombies, both in the compound as
well as eventually the zombie horde outside.
We can break this situation down into sections (sub-puzzles) where
the player needs to figure out how to disable all four individual lasers
one by one. In addition, we show the player a series of weapons lockers
with clearly visible, useful weapons or items, protected by individual
laser beams that correspond with those in the fence. Disabling one laser
means unlocking a new weapon locker, containing weapons or items
that allow the player to deal with the unleashed zombies.
Challenge 363

Laser fence dilemma 2. We can adjust this scenario slightly to fit the context
of player controlled challenge levels. In this second scenario the player can find
two weapons spread out across the compound: a sniper rifle and a flamethrower.
Both have a limited amount of ammunition, so the player can try them out, po-
tentially on a few wandering groups of zombies already in the compound, but
they will soon run out of ammunition.
The weapons lockers from the original scenario are now more frequent,
and they house ammunition for the two new weapons, rather than the weapons
themselves. This means that the player can disable the four defensive lasers and
in doing so effectively choose which ammunition type becomes available.
In effect, the player can choose a preferred play style:
• Ammunition for the sniper rifle may allow for a play style that emphasizes
slow and deliberate play. The rifle is only effective at a large distance, is
slow and unwieldy to use, but it is devastatingly powerful. This will suit
players who like careful, deliberate strategies.
• Ammunition for the flame thrower requires players to run in amongst the
zombies, and affect whole groups of them at once, but the player is then
exposed to them at close range. This requires different skills and playing
styles, which will suit other players.
This provides an elegant and completely integrated way of allowing the player to
choose their preferred challenge level.

Further Notes
Again, the specific solutions in this scenario don’t really matter that much, just
that catering to different player abilities can provide an effective dynamic chal-
lenge environment, under full control of the player.

Example 16.2: Player Controlled

Challenge Levels
Summary
Dynamically adjusted difficulty levels are generally an outcome of a game’s core
design, and not so much the domain of the level design, which just interprets the
game design parameters. There are ways, however, in which we can set the levels
themselves up in such a way that the players can actively choose the challenge
they are comfortable with.
364 Designing Gameplay

Game Genre
This technique is suitable for most games but works best in games with open
environments.

Goals to Achieve
• Allow players to choose the challenge level they are comfortable with in
certain scenarios.

Description
(Example type: Original)
We can create a straightforward way of letting the player decide how much chal-
lenge they are comfortable with by providing a level environment that features
challenge or danger gradients. That is to say that the levels are open enough that
the player can traverse the level areas at will, but that progress is tempered by
escalating levels of challenge, for example, through increased danger.
Imagine a level that gets more challenging (or dangerous) the further the
player strays from the main path. This can be set in a valley, for example, where
progress becomes harder due to natural obstacles or in a forest where the deeper
forest hides more dangerous creatures.
This technique works especially well if the player is given proportional
awards when facing these challenges. The player can keep going further away
from the minimal challenge path until he or she finds that the awards are no
longer worth the difficulty in obtaining them. In effect, this is a physical level
design representation of the optimal flow channel discussed before, one where
the player can balance skill versus challenge.

Example 16.3: Skill Gates and Challenges

Summary
There are times when we don’t want the player to decide where to go unless we
are sure they are able to handle the challenges. To address this we can use skill
gates, which act as a hard barrier against unwarranted progress, while leaving the
choice aspect largely intact.

Game Genre
The technique is suitable for most games that feature skill-based progression.
Challenge 365

Goals to Achieve
• Prohibit the player from progressing beyond their skill level.
• Make this feel like it is under the player’s control.
• Balance skill and challenge within the level design rather than the game
design.

Description
(Example type: Original)
Similar to the previous example but less forgiving is the use of skill gates. The
same principle applies: the player decides which path to attempt through an area
with varying difficulty levels, but in this case the player needs to possess certain
skills to be able to enter the high-challenge area. A higher path through a level
may, for example, only be accessed if the player can execute a tricky athletic ma-
neuver, or is able to displace a guard of certain strength. In some ways this is a
lock-and-key situation: the key is, in fact, the player’s skill level and if high enough
it gives access to high-challenge areas in a level.
Think of it as a series of walkways each with a skill exam at the entrance to
the walkway. If the player passes the exam they can explore that particular route
and reap the rewards as well as face the higher challenge levels. If the player fails
the exam, it shows that the player is not yet ready for the level of challenges that
follow.
The advantage of this technique is that players are not able to end up in
situations they are not ready for, and therefore avoid excessive frustration. Ad-
ditionally, it gives the player something to aim for, especially if the skill gate is
clearly signposted. Few things are as motivating as locked-off areas that contain
rewards.

Further Notes
This technique also works with other factors prohibiting premature progress. It
can be the possession of certain inventory items, player hitpoint levels, or spe-
cific weapons. Other options can easily be found depending on the game.
Final Thoughts VI

“Le Bris Flying machine”, Wikipedia, http://commons.wikimedia.org/wiki/File:LeBris

1

1868.jpg, 2009.
367
Considering the Future
17
A while ago I bought the first eight years of the popular science and science
fiction magazine Omni and have been marveling ever since at some of the
incredibly fun content it contains. Omni started in the late 70s and had attracted
a number of great thinkers and scientists to contribute to its pages since the very
first issues. Some of the most fun to be had with such a retro chic magazine is
to examine their views on the future, and sure enough, Omni published several
futurist articles early on. They make entertaining sci-fi reading material now, but
not much of what was predicted has come to pass. Reading those predictions of
the future is a rather sobering experience actually as even the greatest scientists
and thinkers have very limited success in predicting the future1.
With that realization in mind I made sure that one of the goals I had with
this book was to provide a text that ages as little as possible. Something that is
useful even in 20 years time. As a consequence of that goal I have left out specu-
lative topics and I have minimized technology specific subjects. The latter be-
cause I wanted to write a book about design fundamentals, not about production,
but also because technology in games move so rapidly that it would have aged
the book faster than I could have written it.
Nonetheless, probably to the amusement of readers that pick up this book
when it is a bit older, I would like to point at some areas where I think improve-
ments to our craft and exciting new directions can come from.

Tools and Technology

Development tools and technology are some amongst a number of tools avail-
able to us to create levels and games. They are not the only tool, nor the most
important ones, yet at the time of writing this book there is still an excessive
focus on technology, especially as a means to produce photo realistic visuals.
Much time is spent in game engines and tools on rendering capacity and visual
1
Notable examples like Arthur C. Clarke buck the trend.
369
370 Final Thoughts

tricks and special effects, almost always aiming for “Hollywood style” graphics,
as if that should be the main goal of game development.
Yet, within the tools and technology there is almost no evolution towards
techniques that make game levels play better, or to fundamentally improve the
way levels are made. Quite often they are based on the same style of content
creation as 3D art packages, event though the content created by level designers
is completely different from that.
What we really need are tools that set designers free to try out level de-
sign ideas rapidly and independently. Tools that allow scripted sequences to be
mapped out coherently. We need technology that makes it easy to create envi-
ronments that can be adjusted to play testing results.
In other words, tools that focus on the improvement of the actual work a
level designer does.

Resources
Now that the information age has truly established itself it is becoming clear
that society as a whole has to reexamine its ways in which knowledge is cre-
ated, stored, and made available to others. Projects like Wikipedia have shown
that there are large gains to be made by pooling resources and information and
making this process open and transparent, even cooperative. Many subjects are
now starting to find their way to the public in ever increasing efficiency and
volume.
Yet apart from a limited number of books and websites there is very little in
the way of resources for level designers. With tremendous effort and patience
one can compile a good list of helpful data resources and programs but it is
a disparate affair that every individual level designer is forced to go through.
I think there are two major areas of improvement that can yield better future
results.

Education
Most people working in games are too busy to make games and learn the ropes
to be able to teach the subject to others. And consequently there is a real issue
with finding people of real experience to be able to teach level design or game
design classes. As the profession is still very young I suspect it is just a matter
of time for this to improve. Eventually current professionals will reach an age
and a level of accumulated wisdom that academic careers will become enticing.
This will take a while however and before this can really be an effectively taught
subject we will need to reach a broader consensus on many of the field’s often-
controversial subjects.
Considering the Future 371

Online Resources
As professionals and enthusiasts we should also be able to collect and distrib-
ute knowledge ourselves. One way in which this can be done is to start a level
design knowledge database, similar to Wikipedia, but based on level and game
design knowledge. This can be of immense use as a practical resource and source
of reference because game development is such a wide-ranging affair. It would
allow specialist knowledge to be collected into a central location, made avail-
able for general use. There are level and game designers who are in possession of
vital knowledge on the most diverse subjects: How to balance unit types in turn
based fantasy role playing games. Somebody else may know a lot about the use
of camera techniques in survival horror games or ambush scenario templates for
third person action games.
Once a project like this reaches critical mass I suspect it can become a price-
less aide to game developers all over the world.

Corporate vs. Independent Development

Most people think of game development through the prism of big console plat-
form holders like Nintendo or Sony, or big corporations like Electronic Arts.
The mainstream game industry model sees video games get developed by stu-
dios that sign publishing deals with big companies to fund the production of the
games that everybody knows and loves. The reality of this model is that most
game development is fuelled by commercial goals, and decisions are made with-
in a business model that is not always that suitable to individual artistic intent.
There is another area of future growth and excitement that has started to
assert itself quietly but clearly nonetheless. Independent game creators are start-
ing to make games based on their own personal desires and goals, and although
these games are not made by teams of a 100 developers with multi million dol-
lar budgets, they have something to offer that the big boys often can’t touch:
freedom from focus groups, financial dogma, shareholders, and so on. They can,
and often do, differentiate themselves from mainstream games with ideas and
concepts that are genuinely original and progressive to the art of making games.
I myself have recently snuck over to that way of working and I find it a very
exciting place to be. Hopefully I can take the things I have learned while writing
this book and use them to make a positive contribution to the grass on this side
of the fence.
Recommended Reading
& Materials

I n the process of writing this book, and in the years preceding it, I have been
exposed to an enormous amount of worthwhile reading material and other
inspirational content. Many extracts from this material have found their way into
this book as quotes or in footnotes. I encourage the reader to follow up on the
material referenced within the main body of text, but I would also like to high-
light some extra useful resources—not as an attempt to create an exhaustive list,
but to provide a good starting point to examine some of the things I found useful
while writing the book or while creating levels and games.
This section will therefore showcase some examples of what I found to be
inspirational or that I kept going back to on multiple occasions. I created some
loose categories, within which books appear in alphabetical order; publication
dates come from the editions I own or have read.

Architecture
Architecture: Form, Space and Order, Second Edition
Francis D. K. Ching
Published by John Wiley & Sons, Inc., 1996
This wonderful book demonstrates architectural principles from high-level con-
cepts to nitty-gritty architectural details and brings it to life in clear and informa-
tive examples. The book features hundreds of handmade drawings by the author
himself, and it is a joy to read as a reference work as well as an art book. It is one
of those rare books that will never grow old.
Fantasy Architecture: 1500–2036
Neal Bingham, Clare Carolin, Peter Cook, and Rob Wilson
Published by Hayward Gallery Publishing, 2004
373
374 Recommended Reading & Materials

This is a thin but remarkably rich little book that takes a look at fantastical archi-
tecture through the ages. It is rich with inspirational concepts, ranging from the
farcical to the truly majestic. If you want to be inspired by something a bit differ-
ent in architecture try this book.
In addition, I recommend works by Syd Mead and Lebbeus Wood.

Game Development/Level Design

A Theory of Fun for Game Design
Raph Koster
Published by Paragryph Press, 2004
This quirky little book combines concepts from flow theory with aspects of neu-
rology and couples that with the author’s own personal experiences with loved
ones and game design. The book may not necessarily present much new mate-
rial, but it is all useful and extremely readable and likeable.
Designing Virtual Worlds
Richard Bartle
Published by New Riders Publishing, 2004
This big beast of a book is a must-have with regards to creating virtual worlds and
multiuser game environments. Professor Bartle not only provides a thorough
and incredibly enlightening book, but it has since its original release been ac-
cepted as a foundation from which to build further knowledge. The book intro-
duces language and concepts that have become the state of the art, which makes
it a book that you can’t do without when interested in this subject.
Creating Emotion in Games
David Freeman
Published by New Riders Publishing, 2004
This book is a bit of a tour de force. It attempts to create a practical reference
work on how to apply the author’s self-coined methods of “emotioneering.” The
techniques are meant to create deeper emotional experiences in the games we
develop, not the least in areas of narrative and story. The book has been grum-
bled at by some people because of its somewhat exuberant and confident tone,
but it succeeds largely in providing the reader with a huge amount of inspira-
tional content.
Half-Life 2: Raising the Bar
Valve
Published by Prima Games, 2004
Recommended Reading & Materials 375

Raising the Bar is sometimes mistaken for a coffee table book filled with nice con-
cept art, and indeed the art is wonderful. That misses the true worth of the book,
though: it is a fantastically honest and insightful account of the design process,
including level design, that Valve went through to create its masterful Half Life
games. The book is filled to the brim with great insights, practical anecdotes, and
examples, and no level designer should do without it. Unfortunately, it is cur-
rently out of print and most copies, once tracked down, are rather expensive.

Various
Flow: The Psychology of Optimal Experience
Mihaly Csikszentmihalyi
Published by Harper-Collins Publishers, 1991
This is a groundbreaking work that combines years of exacting academic re-
search into the subject of happiness. The results are of massive importance to
game developers as they so easily translate to our work
Making Movies
Sidney Lumet
Published by Vintage Books, 1996
Sidney Lumet shares in a clear and insightful way many of the things he has
learned and practiced in his extinguished career.
On Writing: A Memoir of the Craft
Stephen King
Published by Pocket Books, 2000
A brush with mortality led Stephen King to write this engrossing hybrid between
biography and reference work. It works well on both levels.
The Design of Everyday Things
Donald A. Norman
Published by Basic Books, 2002
Last but certainly not least, this is a book that every level designer must own. The
design of everyday things is one of the best books on product design, usability,
and other key areas of interest to level designers. Out of all the books in this sec-
tion I would recommend this one the most.
Glossary

L evel design is at times a highly specialized and often technical field. This
naturally leads to a certain amount of jargon and field-specific terminology.
Although I have tried to keep the language used in this book as transparent and
accessible as possible, some notes on terminology are needed. The glossary pro-
vided here can be used as a handy dictionary of level design terms. Please check
this section any time a word’s usage seems ambiguous or unclear.

2D games: Games that make use of a two-dimensional representation of the

game world.
3D games: Games that make use of a three-dimensional representation of the
game world.
agency: The ability of an “agent” to act in (and make an impact on) the world.
AI: Artificial intelligence.
ARG: Alternate reality gaming.
bots: Robots; AI player characters that simulate human opponents.
bullet hell: Describes shoot-em-up games that drown the playing field with
bullets.
catharsis: The (rewarding) release of pent-up emotions.
chatterbots: AI constructs that are created to try to converse with humans.
cognitive dissonance: Tension that comes from holding contradictory beliefs.
CGI: Computer graphic imaging.
concept art: Artwork intended to visualize concepts, rather than illustrate ac-
tual subjects.
COOP: Cooperative play; a game mode where players band together to play a
level.
cutscene: Sequence, often cinematic, where the player has no control over the
onscreen events.
377
378 Glossary

deathmatch: Multiplayer game mode where players compete against each other.
double think: The ability to hold two contradictory beliefs.
experience points: Points awarded to players of role playing games for engaging
with the game’s content; the points are used to gain higher character stats.
fantabulate: Entertaining an audience through fantastical content (admittedly
a made-up word).
FPS: First person shooter.
graphic novel: Comic book that tells a story of substantial length.
hitpoints: Health points of a game character.
level up: Advance a character to a higher level of experience, typically by being
awarded experience points; higher levels bestow higher character stats on
the player.
MMO: Massively multiplayer online game.
MMORPG: Massively multiplayer online role playing game.
mechanic (gameplay): Planned gameplay aspect. For example the wall-jump
in a platform game.
mod: Modification. The ability to edit existing game content in order to create
new content. The result is also referred to as a “mod”.
multiplayer: Game mode where multiple players co-exist.
on rails gameplay: Completely linear gameplay.
OOLD: Object-oriented level design.
OOP: Object-oriented programming.
playtest: Testing the integrity of a level or a game through playing it.
point-and-click adventure: Classic adventure game genre that requires the
player to use the mouse pointer and click on onscreen objects to interact
with them.
prop: Decorative scenery.
retro games: Games that have received classical status.
RPG: Role playing game.
RTS: Real-time strategy game.
sandbox games: Game type or play style that allows players to come up with
many of their own solutions or challenges within the gameworld.
sense of wonder: Feeling that describes a sense of awe or indeed a sense of
wonder in science fiction or fantasy entertainment.
Glossary 379

serious games: Games that have a goal that goes beyond entertainment, typi-
cally education or information delivery.
SHMUP: Shoot-em-up.
shoot-em-up: Game type that has shooting as its central gameplay mechanic.
single player: Game type enjoyed by a single player.
survival horror: Horror game genre that pitches the player against an adver-
sarial environment.
suspension of disbelief: The ability to accept impossibilities in art and enter-
tainment to better enjoy the work itself.
units: Controlled entities in strategy games.
world building: The creation of a virtual world.