y

pan
Cognitive illusions

Com
From magic to science

ing
Gustav Kuhn

lish
Magicians have developed powerful techniques to manipulate our perception

Pub
and awareness. Many of these techniques share similarities with phenomena
typically investigated by psychologists and neuroscientists. Here a novel ap-
proach to the study of consciousness is proposed which utilizes the magician’s
ins
expertise to complement more traditional experimental laboratory based
research. It is argued that this approach offers new and exciting insights into
jam

wide areas of consciousness, such as attention, visual awareness and top-down

modulation of perception.
Ben

Throughout history different theoretical and empirical approaches have been

used to study the human mind. For example, patients with localized brain dam-
ohn

age may exhibit specific cognitive impairments, which provide insights into the
cognitive function of the normal brain. The main interest here is not the lesion
itself, but what it can tell us about the “normal” mind. In contrast, rather than
J

investigating impairments, we may study individuals with particular expertise.

For example, it has been shown that people with domain specific expertise en-
code and represent domain specific information differently thus suggesting that
-

expertise results in altered perceptual experience, which give us clues on how

ofs

we learn to see (e.g. Chase and Simon, 1973). Perhaps the strongest and most
powerful examples of distortions in perception are those produced by magi-
pro

cians. Magicians have developed ways to distort our perception often resulting
in wonder and disbelief (Parris et al., 2009). Indeed distortions of awareness are
arguably the core definition of magic. Given this close link between magic and
ted

awareness, it is rather surprising that with few exceptions (e.g. Kuhn and Land,
2006; Kuhn and Tatler, 2005; Triplett, 1900) this relationship has been largely
rec

ignored by modern science. However, as I will argue here, and have elsewhere
(Kuhn et al., 2008), much can be learnt from the way in which magicians distort
cor

our perception. In this paper I will demonstrate how magic and science can be
Un
140 Gustav Kuhn

y
combined to form a science of magic. In particular I will argue that this rather

pan
novel approach promises new and exciting avenues to discovering more about
human consciousness.

Com
Magicians are not unique in manipulating our awareness. Several “experts”
have learnt to play tricks with our mind. For example, film directors have devel-
oped wonderful techniques to distort our visual perception, many of which share
similarities with those used by magicians (Kinlsey, 1993). Pickpockets manipulate

ing
their victim’s tactile awareness, using techniques that parallel those of the con-
juror. Advertising agents manipulate our decisions, whilst first-rate sports men/

lish
women frequently deceive their opponents thorough deceptive movements (e.g.
Jackson et al., 2006). The only difference between the conjurer and these other ex-

Pub
perts is that magicians have made distortions of awareness the centre of their art.
They have experimented with these principles for several millennia (Christopher,
2006) and have built up a solid tradition of distorting reality. The overlap between
ins
the conjurors’ principles of deception and those used by others demonstrates that
these techniques don’t merely apply to the limited context of the magic perfor-
jam

mance, but are general principles of cognition.

The first step towards a science of magic requires us to identify the link be-
tween the magic trick and the scientific principles involved. This process requires
Ben

translating the conjuror’s language into that of the scientist (Kuhn et al., 2008;
Macknik et al., 2008). For example, conjurors frequently use misdirection to
prevent the audience from detecting their secrets. They define misdirection as a
ohn

“deflection of attention for the purpose of disguise” (Sharpe, 1988). Visual scien-
tists on the other hand would describe misdirection as “attentional modulation of
J

visual awareness”. Whilst the languages may differ, the underlying principles are
the same. Rather than processing all of the perceptual information, we only select
the information that is of importance, trough systematic orienting of attention.
-

Whilst this attentional selection process is a very efficient way of dealing with vast
ofs

amounts of sensory information, it does come at a cost; namely unless we attend

to something, information about it is rapidly lost. This point is vividly illustrated
pro

by the change blindness phenomenon, whereby people often fail to see visually
salient changes to a scene when the change occurs with a visual disruption of the
scene. This demonstrates that whilst our conscious experience of the world is one
ted

of rich sensory detail, our conscious representation is far less complete (Rensink
et al., 1997). What we see is therefore largely determined by what we attend to,
rec

and this attentional orienting can either take the form of overt eye movements
(overt attention), or allocating additional processing resources in the absence of
cor

eye movements, known as covert attention. Misdirection may therefore provide

insights into areas such as attentional orienting and awareness, which lie at the
Un

heart of the study of consciousness.

 Magical insights into consciousness 141

y
Once this link between magic and science has been established, we can take ad-

pan
vantage of the magician’s real world experience to investigate questions that are of
scientific interest. Many conjuring techniques, such as misdirection, are complex

Com
multi-layered principles, which often makes them rather difficult to investigate
scientifically. The second challenge involves finding ways of breaking down some
of these techniques so that they can be studied in a scientifically meaningful way.
Whilst observations and descriptions are important starting points, the true value

ing
of magic is only unleashed by incorporating conjuring techniques into the scien-
tific paradigm. The approach we propose involves identifying the principles used

lish
by magicians and then using a more reductionist approach to investigate it scien-
tifically so as to learn more about the underlying mechanism. These findings can

Pub
then be used to guide the more traditional experimental research, thus closing the
gap between magic and science, and thus forming a science of magic.
Over the past five years we have developed several paradigms in which con-
ins
juring techniques are employed to manipulate people’s visual perception. I will
talk about two of these, namely misdirection and illusions.
jam

Misdirection
Ben

Misdirection is a complex multilayered process in which people’s attention is sys-

tematically orchestrated using a wide range of techniques. In order to study mis-
ohn

direction scientifically we have developed a very simple “magic trick” in which

misdirection is used to prevent people from seeing a fully visible event (Kuhn and
J

Findlay, 2009; Kuhn and Tatler, 2005; Kuhn et al., 2009; Kuhn et al., 2008). In one
of these tricks a magician is seen making a lighter disappear, an effect that was
achieved by distracting people’s attention at the moment in which the lighter is
-

dropped into the magician’s lap (Figure 1). The simplicity of this trick allows us to
ofs

study misdirection in a scientifically meaning full way, by both identifying the fac-
tors that contribute to its effectiveness, as well as manipulating it experimentally.
pro

In order to study something scientifically, we must be able to measure its ef-

fectiveness. The most obvious measure of “effectiveness” is whether it succeeds
in preventing people from seeing a “to-be-concealed” event. In these paradigms,
ted

we ensured that the “to-be-concealed” event was fully visible, which means that
participants’ detection of the event (i.e. whether they saw the object drop) pro-
rec

vides an index of its effectiveness. Moreover, by assuming that detection of the

drop requires covert attention (attention without eye movements), we can use
cor

participants’ detection of the event as an index of covert attention (Tse, 2004).

Whilst this awareness probe informs us about the effectiveness of the attentional
Un

manipulation at one particular point in time, we can use eye tracking techniques
142 Gustav Kuhn

y
pan
Com
ing
Figure 1 shows the sequence of events in the misdirection trick (Kuhn and Findlay,
2009). The effect in this magic trick was the disappearance of a lighter. In this trick the

lish
magician picks up a lighter with his left hand (a) and lights it (b). He then pretends to
take the flame with his right hand (d), and gradually moves it away from the hand that

Pub
is holding the lighter (d, e). During this maneuver the magician is looking at the right
hand. Once it has reached the other side, he snaps his fingers, waves his hand and reveals
that it is empty (f). At the same time the lighter is dropped into the lap. The drop takes
place in full view. The magician now directs his gaze to his left hand (g), raises it, and
ins
snaps his fingers to reveal that his left hand is now also empty and the lighter has disap-
peared (h).
jam

to measure where people are looking through the entire presentation of the trick.
Ben

The eye tracking data therefore provides us with an online measure of where par-
ticipants are attending overtly.
The misdirection paradigm allows us to study aspects of the magician’s misdi-
ohn

rection in a naturalistic, yet scientifically controlled way. It therefore provides us

with a valuable tool to investigate various aspects of attention and awareness from
J

a rather novel angle. For example, we have used this paradigm in “real world” situ-
ations in which participants faced a real person performing the misdirection trick
live (Kuhn and Tatler, 2005). Most of the participants who watched the misdirec-
-

tion trick for the first time failed to see the dropping object even though the event
ofs

took place right in front of their eyes. The misdirection employed by the magician
was, therefore, effective at preventing people from seeing the event. Importantly,
pro

participants did not miss the event due to low visibility. For one, when the trick
was repeated, most of the participants immediately saw the item being dropped,
thus demonstrating that it was indeed visible. Moreover, eye movement records
ted

revealed that participants’ ability to detect the drop was not related to where they
were looking at the moment of the drop, and therefore is not determined by the
rec

low level visual input received by the visual system. Misdirection is a powerful
demonstration of “inattentional blindness”, in which people who are engaged in
cor

attentionally demanding distractor tasks frequently fail to see unexpected yet ful-
ly visible events (Mack and Rock, 1998). However, instead of drawing attentional
Un
 Magical insights into consciousness 143

y
resources away through the use of an implicit distractor task, misdirection works

pan
by implicitly orienting attention away from the to-be-concealed event.
In later work we used pre-recoded versions of similar misdirection tricks

Com
which allowed us to control certain aspects of the stimuli more tightly. Whilst
these pre-recorded versions of the trick were still effective at misdirecting people’s
attention, its effectiveness was somewhat reduced, thus demonstrating that the
engagement in a one-to-one interaction plays an important part in attentional

ing
orientation (Kuhn et al., 2008). However, many of the participants still failed to
see the visibly dropping object even though it happened in front of their eyes.

lish
Rather surprisingly, again detection of the event was independent of where par-
ticipants were looking thus demonstrating a striking dissociation between covert

Pub
and overt attention. Whilst it is clear that under experimental conditions overt
and covert attention can be dissociated, it is generally thought that under free
viewing the main purpose of covert attention is to plan the next saccade, and does
ins
not seem to serve any additional role (Findlay, 2005). However, the current results
demonstrate that covert attention is instrumental in determining what we see or
jam

miss. It illustrates how what we see is not necessarily determined by what we look
at, but by what we attend to covertly, and thus demonstrates the important role
that covert attention plays in visual awareness. The effectiveness of the misdirec-
Ben

tion and the high consistency in eye movements also demonstrate that misdirec-
tion is very effective at orienting peoples’ overt and covert attention.
Given these strong and reliable effects we can then move to ask which aspect
ohn

of the misdirection is driving this attentional orienting process. For example, we

have investigated the role that social cues (i.e. where the magician was looking)
J

have in misdirecting attention (Kuhn et al., 2009). By experimentally manipulat-

ing the misdirection trick, we demonstrated that gaze cues are very important
at misdirecting attention. Moreover, as the effects found in this paradigm were
-

rather larger than those typically observed in more standard laboratory attention
ofs

paradigms (e.g. Friesen and Kingstone, 1998), it suggests that we may be under-
estimating the importance that social cues play in the real world. This real world
pro

approach to attention research also allows us to identify features or events that

are important in orienting attention that are typically ignored by mainstream sci-
ence. Thus far our empirical work on misdirection has predominantly focused on
ted

systematic orchestration of spatial attention, and thus relatively “low” level sen-
sory information. However, the magician’s concept of misdirection also relates to
rec

manipulating attention directed towards thoughts, and thus involves higher levels
of awareness. By identifying some of these principles, and developing paradigms
cor

analogous to the misdirection trick we may learn much more about the mecha-
nisms of this type of misdirection and gain new insights into consciousness.
Un
144 Gustav Kuhn

y
Illusions

pan
Our conscious perception of the world requires a transformation of the light that

Com
falls onto the retina into recognizable objects. Our visual experience is largely de-
termined by the way in which these signals are interpreted by the brain (Gregory,
1968). However, there are certain situations that present specific difficulties in
interpreting this information which then give rise to systematic errors. Most of

ing
the time, these errors are unnoticed. However, when we do become aware of them
they are experienced as visual illusions. Systematic errors in perception can serve

lish
as clues as to how the brain generally solves perceptual problems. For example
much of what we have learnt about depth perception comes from investigating

Pub
our susceptibility towards visual illusions (Gregory, 1968). Illusions lie at the heart
of magic. Many of the illusions employed by magicians do not necessarily differ
from the tricks the mind plays on us in everyday life. What makes them special are
ins
the theatrical ways in which they are presented. Moreover, magicians have learnt
to maximize their effectiveness by stage-managing the context in which they are
jam

viewed. The systematic investigation into these illusions and particularly the way
in which conjurors enhance their effectiveness could therefore provide valuable
insights into how the visual system solves some of these perceptual problems.
Ben

The illusions typically performed by magicians are often rather complicated.

The key to studying magic illusions scientifically involves finding illusions that are
both simple yet effective, and to discover ways of measuring their effectiveness.
ohn

An illusion that ticks all of the above boxes is the vanishing ball illusion (Kuhn
and Land, 2006; Triplett, 1900). In this illusion a magician is seen throwing a
J

ball up and down a couple of times. On the final throw, he merely pretends to
throw the ball when instead it remains secretly concealed in his hand (Figure 2).
Rather surprisingly, people often claim to see a ball leave the magicians hand,
-

even though it remained in the hand. This illusion is simple (i.e. merely involves
ofs

someone pretending to throw a ball in the air), but very effective. Moreover, par-
ticipants’ experience of the illusion (i.e. whether they saw the imaginary ball or
pro

not) can be used as a simple measure of its effectiveness.

Using this vanishing ball illusion we have shown that even though the ball
did not leave the hand, over half of the participants were convinced they had seen
ted

it move towards the top of the screen (Kuhn and Land, 2006). This finding dem-
onstrates how people’s perception of an event is based on the expected outcome,
rec

rather than the low level sensory input. Perceptual processes that are based on ex-
pectations or prediction have their advantage, as they allow for rapid action, which
cor

is particularly useful when fast responses are required. However, as demonstrated

by the vanishing ball illusion, reliance on such strategies can lead to errors, and
Un

illusory percepts. Whilst illusory percepts during a magic show can be rather fun,
 Magical insights into consciousness 145

y
pan
Com
ing
Figure 2 shows the sequence of events in the vanishing ball illusion. In the Social Cues

lish
Pro-illusion condition (top panel) the magician’s gaze followed the trajectory of the
imaginary ball (final pretend throw). In Social Cues Anti-Illusion condition the magi-

Pub
cian looked at his hand, rather than the illusory trajectory. Figure taken from Kuhn et
al. (2008).
ins
their effects could have detrimental consequences in other situations such as driv-
ing. Thus understanding how we distinguish between illusion and reality may have
jam

rather important theoretical, as well as practical, consequences. If our conscious

perception is based on expectations it is important to understand the processes
Ben

that drive these expectations. Conjuring illusions often rely on manipulating ex-
pectations, and magicians have learnt to identify situations in which our percep-
tion is particularly sensitive to anticipation, and developed ways of manipulating
ohn

these expectations. The magic literature tells us that social cues (i.e. where the ma-
gician is looking ) are particularly important in manipulating people’s attention as
well as intention (i.e. expected outcome) (Sharpe, 1988). Indeed Kuhn and Land
J

(2006) showed that if the magician looked at his hand, rather than the trajectory of
the imaginary ball, the illusion’s effectiveness is greatly reduced, implying that the
social cues did indeed influence expectations, and therefore what was perceived.
-

Moreover, unbeknown to the observers, the eye movement records revealed that
ofs

participants strategically utilized the social cues to predict the final location of the
ball. Rather than merely keeping an eye on the ball, participants glanced at the face
pro

prior to fixating on the ball once in the air. The use of adaptive heuristics to solve
complex perceptual problems has also been reported in other everyday domains,
ted

such as driving (Land and Lee, 1994) or cricket (Land and McLeod, 2000). It is
indeed thought that vision in general involves a complex set of problem solving
rec

strategies, and the way in which magicians trick these processes may provide im-
portant insights into how these strategies work.
cor

Magicians have developed an astonishing expertise in manipulating our

awareness. As scientist interested in consciousness and awareness, it would
Un

be a shame to ignore this enormous expertise. Luckily the first steps towards
146 Gustav Kuhn

y
recognizing the link between science and magic, as well as the development of

pan
paradigms that allow us to test these principles scientifically have been made
(Kuhn et al., 2008; Lamont and Wiseman, 1999; Macknik et al., 2008). Whilst

Com
the science of magic is indeed a rather novel approach, the concept of using real
world expertise to study aspects of the mind is far from new. For example, artists’
experimentation with perceptual effects has taught us much about the percep-
tion of colour and perspective (Pinna, 2008). Similarly magicians may enlighten

ing
us about aspects of attention and awareness. Conjurors are in no way unique, in
that they have expertise that could be of scientific interest. However, what makes

lish
them special is the dedication that they have given to understand and perfect
these perceptual distortions.

Pub
Our aim as cognitive scientists is to explain and predict human behaviour in
the real world. In cognitive psychology as well as in neuroscience, the trend has
been to simplify the issue of investigation by making the experimental context
ins
both impoverished and controlled, so as to discover the causal relationships be-
tween one factor and the other. The idea that we can learn things about the mind
jam

by isolating it in a box and bombarding it with artificial stimuli has however been
challenged (Kingstone et al., 2008). In particular, the cognitive ethology frame-
work suggests “that it is by starting with real-world observations and individual
Ben

variation that one is most likely to generate subsequent research questions for
investigation that may lead to general principles of cognition that have relevance
to naturally occurring phenomena” (Kingstone et al., 2008). According to this
ohn

approach, “the initial job of the researcher is simply to observe, describe, and
measure what people do and experience in the situation of interest” (Kingstone et
J

al., 2008). In many ways this “job” of observing, describing and measuring human
behaviour is something magicians have already done. By utilizing the experience
developed by magicians we may, therefore, shortcut this initial, often difficult task
-

of real world observation.

ofs

The science of magic as an approach to study consciousness and the human

mind in general has several advantages. For one the principles used by magicians
pro

are incredibly robust and effective, which means that they are an ideal starting
point for scientific explorations. Moreover, many of the techniques and theories
are well documented which makes them an ideal starting point for scientific ex-
ted

ploration (e.g. Sharpe, 1988). Most of the principles used by magicians have been
tried and tested in front of live audiences, which means they are likely to hold
rec

up in the real world, which has major practical applications. Finally, this field
of research is still very much in its infancy, but by identifying and investigating
cor

some of the techniques used by magicians that may not necessarily be known to
the scientific community, this approach has the potential to result in discoveries
Un

of new principles.
 Magical insights into consciousness 147

y
Much of what we have learnt about consciousness comes from identifying

pan
the types of tricks that the mind plays on us. For example, our susceptibility to-
wards change blindness, and inattentional blindness has taught us about the lim-

Com
ited conscious visual representation. Similarly investigations into visual illusions
such as the hollow mask illusion are powerful demonstrations of how top down
information influences our conscious perception. The ease with which magicians
can distort our perception is a stark reminder of just how subjective our con-

ing
scious perception truly is. However, more importantly, by developing a science of
magic, which investigates these distortions scientifically, we may gain novel and

lish
important insights into some of the factors that are responsible for this conscious
representation.

References
Pub
ins
Chase, W. G. and Simon, H. A. (1973). Perception in Chess. Cognitive Psychology, 4, 55–81.
Christopher, M. (2006). The Illustrated History of Magic. New York: Carroll and Graf Publish-
jam

ers.
Findlay, J. M. (2005). Covert attention and saccadic eye movements. In L. Itti, G. Rees and
J. Tsotsos (Eds.), Neurobiology of Attention (pp. 114–117). London: Academic Press.
Ben

Friesen, C. K. and Kingstone, A. (1998). The eyes have it! Reflexive orienting is triggered by
nonpredictive gaze. Psychonomic Bulletin and Review, 5, 490–495.
Gregory, R. L. (1968). Perceptual Illusions and Brain Models. Proceedings of the Royal Society
ohn

of London Series B-Biological Sciences, 171, 179–296.

Jackson, R. C., Warren, S. and Abernethy, B. (2006). Anticipation skill and susceptibility to
deceptive movement. Acta Psychologica (Amst), 123, 355–371.
J

Kingstone, A., Smilek, D. and Eastwood, J. D. (2008). Cognitive Ethology: A new approach for
studying human cognition. British Journal of Psychology, 99, 317–340.
Kinlsey, D. (1993). Georges Méliès: Magier der Filmkunst. Stroemfeld: Roter Stern.
-

Kuhn, G., Amlani, A. A. and Rensink, R. A. (2008). Towards a science of magic. Trends in
Cognitive Sciences, 12, 349–354.
ofs

Kuhn, G. and Findlay, J. M. (2009). Misdirection, Attention and Awareness. Inattentional blind-
ness reveals temporal relationship between eye movements and visual awareness. Quar-
pro

terly Journal of Experimental Psychology, 1–11.

Kuhn, G. and Land, M. F. (2006). There’s more to magic than meets the eye. Current Biology,
16, R950–R951.
ted

Kuhn, G. and Tatler, B. W. (2005). Magic and fixation: Now you don't see it, now you do.
Perception, 34, 1155–1161.
Kuhn, G., Tatler, B. W. and Cole, G. G. (2009). You look where I look! Effect of gaze cues on
rec

overt and covert attention in misdirection. Visual Cognition, 19, 925–944.

Kuhn, G., Tatler, B. W., Findlay, J. M. and Cole, G. G. (2008). Misdirection in magic: Implica-
cor

tions for the relationship between eye gaze and attention. Visual Cognition, 16, 391–405.
Lamont, P. and Wiseman, R. (1999). Magic in theory. Hartfield: Hermetic Press.
Un

Land, M. F. and Lee, D. N. (1994). Where we look when we steer. Nature, 369, 742–744.
148 Gustav Kuhn

y
Land, M. F. and McLeod, P. (2000). From eye movements to actions: how batsmen hit the ball.

pan
Nature Neuroscience, 3, 1340–1345.
Mack, A. and Rock, I. (1998). Inattentional blindness. Cambridge, MA: MIT Press.
Macknik, S. L., King, M., Randi, J., Robbins, A., Teller, Thompson, J. and Martinez-Conde, S.

Com
(2008). Attention and awareness in stage magic: Turning tricks into research. Nature Re-
views Neuroscience, 9, 871–879.
Parris, B. A., Kuhn, G., Mizon, G. A., Benattayallah, A. and Hodgson, T. L. (2009). Imaging the
impossible: An fMRI study of impossible causal relationships in magic tricks. Neuroimage,

ing
45, 1033–1039.
Pinna, B. (2008). Art and Perception. Towards a Visual Science of Art, Part 1. Leiden: Brill.

lish
Rensink, R. A., O’Regan, J. K. and Clark, J. J. (1997). To see or not to see: The need for attention
to perceive changes in scenes. Psychological Science, 8, 368–373.
Sharpe, S. H. (1988). Conjurers Psychological Secrets. Calgary AB: Hades Publications.

Pub
Triplett, N. (1900). The psychology of conjuring deceptions. The American Journal of Psychol-
ogy, 6, 439–510.
Tse, P. U. (2004). Mapping visual attention with change blindness: New directions for a new
method. Cognitive Science, 28, 241–258.
ins
jam
Ben
ohn
- J
ofs
pro
ted
rec
cor
Un