See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/249624620

Evolutionary Influence on Human Landscape

Preference

Article  in  Environment and Behavior · July 2010

DOI: 10.1177/0013916509341244

CITATIONS READS

112 1,614

2 authors, including:

John H. Falk
Oregon State University
112 PUBLICATIONS   6,880 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Broader Impacts View project

All content following this page was uploaded by John H. Falk on 06 July 2014.

The user has requested enhancement of the downloaded file.

 Environment and Behavior
http://eab.sagepub.com

Evolutionary Influence on Human Landscape Preference

John H. Falk and John D. Balling
Environment and Behavior 2010; 42; 479 originally published online Aug 7,
2009;
DOI: 10.1177/0013916509341244

The online version of this article can be found at:

http://eab.sagepub.com/cgi/content/abstract/42/4/479

Published by:

http://www.sagepublications.com

On behalf of:
Environmental Design Research Association

Additional services and information for Environment and Behavior can be found at:

Email Alerts: http://eab.sagepub.com/cgi/alerts

Subscriptions: http://eab.sagepub.com/subscriptions

Reprints: http://www.sagepub.com/journalsReprints.nav

Permissions: http://www.sagepub.com/journalsPermissions.nav

Citations http://eab.sagepub.com/cgi/content/refs/42/4/479

Downloaded from http://eab.sagepub.com at OREGON STATE UNIV LIBRARY on June 21, 2010
Article
Environment and Behavior
42(4) 479­–493
Evolutionary Influence © 2010 SAGE Publications
Reprints and permission: http://www.
on Human Landscape sagepub.com/journalsPermissions.nav
DOI: 10.1177/0013916509341244
Preference http://eab.sagepub.com

John H. Falk1 and John D. Balling2

Abstract
Individuals residing in the rainforest belt of Nigeria were shown photographs
of five biomes: rain forest, deciduous forest, coniferous forest, savanna, and
desert. Subjects overwhelmingly selected savanna scenes as representing the
most desirable place to live. These results, coupled with extensive American
data, support the hypothesis that humans possess an innate preference
for savanna-like settings, which then is modified through experience and
enculturation. Findings are discussed in relation to anthropological, biological,
and psychological research.

Keywords
landscape preference, evolutionary psychology, savanna, cross-cultural, human
evolution, habitat preference

The analysis of historic patterns of landscape design worldwide reveal some

striking parallels in the use and organization of certain landscape elements
(Hyams, 1971a; Jellicoe & Jellicoe, 1975; Newton, 1971). Particularly in large
gardens and estates, where aesthetic preferences could most easily be expressed,
park-like settings of short grass and scattered trees have consistently emerged.
The English landscape tradition, as typified in the work of “Capability”
Brown, is one of the most familiar examples (Hyams, 1971b; Stroud, 1975),
where random placement of trees gathered in clumps are dispersed through-
out an open field.

1
Oregon State University
2
Willamette University

Downloaded from http://eab.sagepub.com at OREGON STATE UNIV LIBRARY on June 21, 2010
480 Environment and Behavior 42(4)

This landscape tradition most closely resembles the savanna biome—

grasslands with more or less scattered dense areas of trees. Orians (1980) has
further demonstrated that horticulturists and landscape designers in Japan have
tended to modify the phenotype of native tree-forms toward those found among
trees existing in savanna environments. In a cross- cultural study of tree pref-
erence (Orians & Heerwagen, 1992), subjects from three countries rated as
most attractive those trees in which the tree canopy is relatively dense and the
trunk bifurcates low to the ground. These characteristics of tree-form typify
those found in the East African savanna. Sommer and Summit (1995) found
consistent findings, with preference for large canopies and small trunks. A
later study of tree species preference and context, Summit and Sommer (1999)
found that the acacia-type tree-form, typical of the African savanna, was pre-
ferred over oak, conifer, palm, and eucalyptus. Dutton (2003) has argued that
the landscapes depicted in calendar art tend to have similar themes throughout
the world and that these themes are the very ones that would be predicted by
a savanna preference hypothesis.
Theoretical frameworks founded in evolutionary theory have been pro-
posed that attempt to dissect landscape preference into predictive variables
(Appleton, 1975; Kaplan, 1987; Ulrich, 1977). Ulrich analyzed informational
qualities of various settings and identified five significant variables: focality
(coherence, unity), ground surface texture, depth, mystery (the sense that
more information can be gained at low risk) and complexity. Earlier studies of
visual preference focused on complexity as the significant mediator of visual
preference (Day, 1967; Wohlwill, 1968). These studies discovered an optimal
range of preferred complexity in the visual field. The complexity of urban
landscape has been shown to have a positive influence on preference (Han,
2007; Heath, Smith, & Lim, 2000). These studies isolated complexity with the
use of urban scenes or geometric design, with the exception of Wohlwill who
included natural scenes in his preference study. Wohlwill rated slides of non-
representational art, urban and rural scenes, and natural landscapes in terms of
the complexity of color, shape, line direction and texture. Subjects were then
asked to rate the slides according to preference. Results showed that visual
preference was an inverted U-shaped function of increasing complexity. Stimuli
judged to be of intermediate complexity were the most preferred, but for natu-
ral landscapes the effect was weak. Kaplan (1987) suggested visual content of
the natural landscapes to be of greater significance. Though not the primary
focus of the study, Kaplan found that natural landscapes were preferred over
urban scenes regardless of level of complexity. Landscape complexity
accounted for relatively little of the variance in preference. As have been
found by others, Han (2007) found that water was highly preferred.

Downloaded from http://eab.sagepub.com at OREGON STATE UNIV LIBRARY on June 21, 2010
Falk and Balling 481

Prospect refuge theory (Appleton, 1975), when applied to this landscape

preference data, supports the notion that the savanna-like landscape, consist-
ing of low shrubs and clumps of trees divided by open spaces, offers an attrac-
tive combination of prospect and refuge symbolism. Open spaces provide an
opportunity to detect potential hazards—to see, whereas the low shrubs and
clumps of trees provide places to hide and escape—to not be seen. Appleton
makes reference to many examples of landscape painting, including “Land-
scape with Apollo and the Muses,” by Claude Lorraine (1600-1682), which
depicts a rural scene with a vista open to the distant horizon, small clumps of
large trees in the foreground, and low grasses dominating the open spaces.
Claude’s work, known for conveying a sense of tranquility and security, is
described by Appleton: “Black shadows under massive canopies of foliage
balance long, open views, providing satisfying equilibrium of complementary
strategic components” (p. 123). Appleton implies that humans have an affinity
for settings that exhibit savanna-like characteristics, but he is careful not to
suggest that there exists an optimum landscape preference since the idea of
“optimum” implies that human needs are the same at all times. On the con-
trary, humans have varying requirements at different times, which effect how
we interact with our environment. In his autobiography however, Appleton
(1994) said “When I look back on my own preferred landscape types I find
that many of the places which have most strongly attracted me can be seen to
have at least some affinity with the savanna” (p. 222). Kaplan and Kaplan
(1989) put it another way. People should prefer settings that (a) make the
acquisition of landscape information relatively easy and nonthreatening and
(b) convey the notion that additional information can be acquired if the setting
is explored. The experience of landscape is one of understanding and explor-
ing (Kaplan, 1987). Understanding, which is the comprehension or making
sense of the scene, is predicted by legibility and coherence. Exploration, the
promise of additional information, is predicted by complexity and a sense of
mystery.
Two other constructs have been used to investigate landscape preference.
Hagerhall, Purcell, and Taylor (2004) correlated the preference ratings of out-
door scenes in Sweden, northern Italy and Australia with fractal properties.
Fractals are a rough or fragmented geometric shape that can be split into parts,
each of which is (at least approximately) a reduced-size copy of the whole;
fractals have been found to be common in nature and can be analyzed by their
specific properties such as scale independence, self-similarity, and complex-
ity. When scenes with visible water were removed, Hagerhall et al. found that
the most preferred landscapes were those calculated to be in the mid-range of
the fractal dimension. This peak in preference corresponded to the fractal

Downloaded from http://eab.sagepub.com at OREGON STATE UNIV LIBRARY on June 21, 2010
482 Environment and Behavior 42(4)

characteristics of savanna-like natural environments. Another construct for

investigating landscape preferences has been affordance, described by Gibson
(1979) as the quality of an object or an environment which allows an indi-
vidual to perform an action. According to some researchers (e.g., Kaplan, 1992;
Rourke, 2007), landscape preference reflects those physical attributes that are
perceived to be beneficial to survival. Without denying the influence of expe-
rience, a range of theoretical frameworks reflect a growing consensus that
a complete understanding of the human experience of natural landscapes
requires the use of evolutionary principles.
Our investigations of human landscape preference were similarly moti-
vated. Struck by the prevalence of human-created parklands in widely dispa-
rate societies, and mindful of a number of biological studies showing that some
degree of preference for specific habitats is at least in part genetically deter-
mined (e.g., Klopfer, 1969; Wecker, 1963). Falk (1977) hypothesized that
human landscape preferences could be strongly affected by our evolutionary
adaptation to life on the savannas of East Africa. Preferred landscapes would
be those subjectively perceived to contain features favorable to human sur-
vival; landscapes that (unconsciously) appeared to afford access to subsistence
resources and pose minimal threat. Our previous research revealed that a large
sample of Americans, drawn from a broad educational, socioeconomic and age
range, showed a visual preference for savanna environments at least as strong
as that for more familiar natural settings, namely deciduous and coniferous
forests (Balling & Falk, 1982). In fact, subjects under the age of 12 demonstrated
a significant preference for savannas over all other environments. Savanna and
the familiar environments were also significantly preferred by all age levels
over two other biomes: desert and tropical rain­forest. To account for these data,
we argued that an initial innate preference for savanna, most clearly expressed
in childhood, was modified and overlain by experience with certain other veg-
etative forms. Savanna, however, was never significantly less preferred than
the more familiar environments for any participant. Although the data supported
the savanna hypothesis, the possibi­lity existed that some cultural bias, rather
than some innate preference, provided the best explanation of the results.
Members of Western societies, Americans in particular, may differ from other
peoples in their preference for open forests and savanna environments. Perhaps
the structure, desirability, and commonness of the parkland typically incorpo-
rated into urban settings has influenced Westerners’ preference for savanna-
like landscapes.
Cross-cultural studies conducted since our initial findings have reported high
correlations between groups in their preference for varied landscapes, with
familiarity accounting for much of the variance (Herzog, Herbert, Kaplan, &

Downloaded from http://eab.sagepub.com at OREGON STATE UNIV LIBRARY on June 21, 2010
Falk and Balling 483

Crooks, 2000; Kaplan & Herbert, 1987). Herzog et al. replicated the earlier study
by Kaplan and Herbert, but included a broader sample of settings and examined
several subcultural groupings. Familiarity once again emerged as a significant
predictor of preference and is consistent with our earlier thesis that preferences
can be modified by experience and enculturation. Six landscape categories were
selected: including rivers, dry lake beds, flood plains, terraces, mallee (a type of
Eucalyptus) plains and cultural images. American college students were com-
pared to several Australian subcultural groups. Australian students were divided
further by age: primary, secondary, college students, and adults. College students
were further sub­divided into three groups and adults into four professional groups.
On a 5-point scale, participants rated the 60 slides for preference, defined as how
pleasing they found the setting, or how much they liked it. The two cultures
showed high correlations, with the greatest divergence coming from the Austra-
lian adults. In the cross-cultural perception comparison test, participants inde-
pendently divided the 60 slides into six of their own categories: vegetation, open
smooth, open coarse, rivers, agrarian, and structures. The two open catego-
ries, featuring little or no vegetation with vast open views, and the structures cat-
egory, represented by man-made materials present in the scene, were similarly
least preferred. Two categories, vegetation, consisting of near trees and bushes,
and agrarian, consisting of agricultural settings with trees present, were signifi-
cantly more preferred, but similar in preference to each other. The most preferred
category, significantly above all others, was rivers. Each of these six categories
was represented by one photograph each in the published study. Although none
of the photographs clearly represented a savanna-like setting, it is interesting to
note that the top three preferred categories included trees relatively close to the
observer with large open areas in view, with the highest preference ratings com-
ing from the youngest participants.
Our initial study (Balling & Falk, 1982), conducted in the eastern United
States, represented one culture from a region dominated by one biome, decid-
uous forest. The study reported here extends our research to a very different
cultural and environmental milieu and provides support for our previous
conclusions.

Method
In February 1981, J.F. sampled three populations of West Africans from Riv-
ers State, Nigeria. Two of the samples were drawn from “First Form” second-
ary school children, ranging in age from 12 to 18 years. Sample A (N = 27)
was taken from a coeducational public secondary school, in an upland rain-
forest area, roughly 100 km north-west of Port Harcourt. The dominant local

Downloaded from http://eab.sagepub.com at OREGON STATE UNIV LIBRARY on June 21, 2010
484 Environment and Behavior 42(4)

vegetation was slash and burn agricultural land interspersed with scattered
pockets of remnant rainforest. Sample B (N = 36) was from a girls’ parochial
school located in a small fishing village on one of the hundreds of small
islands in the Niger Delta. The dominant local vegetation was mature tropical
rainforest and mangrove forest; very little of the land was under cultivation.
A total of 87% of subjects from both samples A and B had been born in and
had never traveled beyond areas vegetationally comparable to their home
town area. Each participant was individually interviewed and asked to exam-
ine 45 pairs of 3″ × 5″ (8.6 cm × 12.8 cm) photographs of landscapes repre-
senting all possible unique binary combinations of 10 scenes, two each of
tropical rainforest, temperate deciduous forest, temperate coniferous forest,
tropical savanna, and mid-latitude desert.1 Except for vegetational density, all
pictures were as comparable as possible (see Balling & Falk, 1982). The
participant was asked, for each pair, to indicate which photo looked most like
a place where he or she would like to live. Although subjects varied greatly
in age, they were all at similar levels of schooling. Performance on this rather
hypothetical task, which demands making judgments without recourse to real
world knowledge, should tend to be more a function of years in school than
chronological age (Sharp, Cole, & Lave, 1979).
Sample C (N = 37), aged 20-39 years, was drawn from a class of students
at a technical college in Port Harcourt, Nigeria. These students’ homes were
located throughout Rivers State, both delta and upland areas. A total of 73%
had never traveled outside of the rainforest belt of West Africa. This sample
was shown 20 randomly ordered slides, representing four replicates of each
of the five biomes previously mentioned. This was the same procedure used
with most of our American subjects. Subjects were asked to indicate on a
5-point Likert-type scale the degree to which the scene depicted appeared to
be a favorable place in which to live.

Results
For samples A and B, the total number of times each slide was preferred to
every other slide was calculated. As indicated in Table 1, for subjects in
samples A and B, savanna was highly preferred relative to all other biomes
(Χ2(4) = 119.6, p < .001). Pair-wise tests revealed that rainforest, desert,
deciduous forests, and coniferous forests were not significantly different
from one another. As indicated by Thurstone scale values also shown in
Table 1, a positive preference was demonstrated only for savanna; subjects
were either neutral or slightly negative toward all other biomes. There were
no statistical differences between the two samples.
Falk and Balling 485

Table 1. Biome Selection With Corresponding Thurstone Scale Values

Biome

Rain Deciduous Coniferous

Savanna Forest Forest Forest Desert

Total number of preferred 794 559 498 482 502

responses
Thurstone scale values 0.493 -0.022 -0.147 -0.184 -0.141
Note: The total number of times each biome was selected as the preferred environment
in the paired-comparison task with Samples A and B (Total N = 63), plus corresponding
Thurstone Scale Values to illustrate the standing of each biome relative to a neutral point.

Table 2. Mean Preference Rating With Standard Error

Biome

Rain Deciduous Coniferous

Savanna Forest Forest Forest Desert

Mean rating 3.82 2.24 2.89 2.77 2.83

Standard error 0.11 0.13 0.13 0.11 0.15
Note: Mean preference rating (5 = very high) and standard error for each biome in the slide
judgment task with Sample C (N = 37).

A within-participant analysis of variance was performed on the rating data

from Sample C with biome as the major independent variable and the picture
replicates within each biome as a nested random variable. Based on the
expected mean squares, there was no appropriate error term in this design for
a test of the biome effect, but an F1 can be calculated (Myers, 1972). As can
be seen in Table 2, there were differences in preference among the five biomes,
thus the omnibus tests were significant (F1(4, 21) = 9.29, p < .001). Compari-
sons revealed that savanna was again significantly preferred over all other
biomes (F1(1, 21) = 9.03, p < .01). However, unlike the data from samples A
and B, rainforest was the least preferred biome (F1(1,21) = 7.31, p <.025).

Discussion
We have previously argued that the tendency to purposefully create savanna-
like landscape elements and arrangements is not accidental, but rather at
least partially the result of an innate preference for savanna-like settings
486 Environment and Behavior 42(4)

(Balling & Falk, 1982; Falk, 1977). Such a preference for the savanna-like
landscape would have theoretically developed over the course of human evo-
lution, initially as an adaptation to surviving in the savannas of East Africa.
Although life experience modifies preference, the affinity younger subjects
have for savanna-like settings seems to suggest that a preference for savanna-
like environments likely represents a vestigial trait of early human biology; a
trait which expresses itself in modern humans as an aesthetic preference.
The belief that human biology and behavior were significantly molded
by adaptations our early ancestors made in order to survive in savanna-like
environments grew out of a series of major hominid fossil discoveries in
southern and eastern Africa during the early part of the 20th century. Associ-
ated fauna and flora led anthropologists to the reasonable inference that the
evolution of modern hominids was guided by the change in habitats through-
out Africa. The increase in aridity was thought to have propelled early homi-
nids away from the comfort of tropical habitats into the more hostile savanna.
Recent studies indicate that the landscape of human evolution was not
exclusively savanna, but rather a mosaic of environments including grass-
lands, savanna, open and closed forests (Butzer, 1977; Foley, 1996; Kingston,
Marino, & Hill, 1994; Leakey, Feibel, McDougall, & Walker, 1995; Scott,
1990; White, Suwa, & Asfaw, 1994).
The growing body of information about the habitats in which early humans
evolved remains insufficient to make a strong case for or against the hypoth-
esis that humans evolved a habitat preference for savanna-like environments.
We do know that human ancestors lived in or near savannas. We also know
they lived in or near a wide range of other habitats. However, the presence of
hominid fossils in these other settings does not in and of itself inform where
these hominids lived, since remains may have been transported to other envi-
ronments by predators or through physical events (e.g., flash floods). Hence,
fossil remains alone provide insufficient evidence to say which, if any, of
these many environments were the preferred habitat of our immediate ances-
tors. Given the diversity of environments inhabited by early hominids,
Chamberlain (2000) argued that a human preference for savanna habitats
may have arisen relatively recently with the emergence of Homo sapiens.
Evidence in other living primate groups suggests that environmental prefer-
ences are a species-specific adaptation rather than primitive or generalized
traits that have been inherited by all members of a genus. Furthermore, as
advances in our understanding of the human genome reveal, modern humans
are exceedingly closely related (Davies, 2001; Ridley, 2000; Wells, 2002).
For example, any two modern humans, from anywhere on the globe, possess
greater genetic similarity than do any two chimpanzees; even those selected

Downloaded from http://eab.sagepub.com at OREGON STATE UNIV LIBRARY on June 21, 2010
Falk and Balling 487

from neighboring groups in Africa. These findings add credence to current

theories that all modern humans evolved from a very small founder popula-
tion, perhaps no more than a few thousand closely related individuals, living
in Africa some 70,000 years ago (Lander, 2001; Rogers & Jorde, 1995; Wells,
2002). Hence, if a vestigial genetic habitat preference for savanna exists
among modern humans, it is because it was present in this single founder
population of Homo sapiens. Whether or not a savanna habitat preference
existed in related populations living in similar or dissimilar environments in
other parts of Africa or the world is irrelevant.
Habitat preferences are found in all animals, and most biologists would
agree that they are adaptive, since fitness is likely to be higher in a preferred
habitat. Natural selection would maintain habitat preferences to the degree
that they have a genetic basis (Jaenike & Holt, 1991; Rausher, 1984). Accord-
ing to Petit and Petit (1996), “the ability to distinguish and select among
habitats is one of the most evolutionary significant traits possessed by any
organism.” A wide variety of investigations have found evidence that verte-
brate habitat preferences are at least in part, genetically determined (Holt,
1987; Jaenike & Holt, 1991; MacCallum, Nurnberger, Barton, & Szymura,
1998; Martin, 1998; Partridge, 1978; Petit & Petit, 1996; Rausher, 1984;
Wecker, 1963). One early study showed that distinct, genetically determined,
habitat preferences can exist even within races of a single mammalian
species (Wecker, 1963). More recently, MacCallum et al. used biochemical
markers to distinguish between the different alleles for at least one of the
genes linked to habitat preference in two closely related species of toads.
Previous studies have identified the genes responsible for habitat preference
in several insect species (Barker, Starmer, & Fogleman, 1994; Fox, 1993;
Singer & Thomas, 1988). Hence a growing body of data support the fact that
a complex behavior such as habitat prefe­rence can be genetically determined
and that a wide range of organisms, including vertebrates of all kinds possess
these genetic traits.
In contrast to an evolutionary and biological interpretation, it can be argued
that human landscape preferences is best understood as a continuous progres-
sion of aesthetic ideals, tempered by social convention, passed on from one
generation to the next through human culture. This view of the world, which
assumes that all human behaviors are the products of our culture and history has
a long history in the humanities and anthropology. More recently, this sociocul-
tural view of the world has become reinvigorated among psychologists with the
rediscovery of the writings of Vygotsky (1978). The basic premise of this group
of psychologists is that the world in which each of us lives is socioculturally
constructed (Moll & Greenberg, 1990; Wertsch, 1986); psyche and culture are

Downloaded from http://eab.sagepub.com at OREGON STATE UNIV LIBRARY on June 21, 2010
488 Environment and Behavior 42(4)

seamlessly interconnected. Several studies have been conducted which show that
even such “basic,” presumed biologically influenced human perceptions as color
recognition (Berlin & Kay, 1969; Garro, 1986; Heider, 1972; Kay & Kempton,
1984; Lucy & Schweder, 1979), facial expressions (Gerber, 1975, 1985; Levy,
1973, 1990; Lutz, 1982, 1985; Poole, 1985) and emotion (Wierzbicka, 1986;
Gerber, 1985; Lutz, 1986) are culturally bound. So why not landscape preference
as well? According to this view, all objects in the world, including landscapes are
cultural constructs and represent a socially mediated form of culturally specific
conversation between the producers of that medium and the user (Moll & Green-
berg, 1990); landscape preferences are not innate but acquired.
Against this mix of opinion, we try to make sense of the current data. The
three modern human populations sampled in this study represented a diversity
of sociocultural backgrounds and experiences (Agrarian, fishing and urban
professional; rural and urban). When compared to previously sampled Ameri-
cans, with dramatically different sociocultural and economic histories, we
find a striking convergence in the data. The results reported here are consistent
with our earlier findings, leading to greater confidence in the accuracy of our
earlier thesis: humans appear to possess an innate preference for savanna-like
settings, but that preference can and typically is modified through experience
and enculturation. Beyond a strong modal tendency to prefer savanna, how-
ever, responses to landscapes become much more difficult to predict in differ-
ent populations. We have earlier postulated that experience and familiarity
with nonsavanna environments can lead to an elevation of preference for these
landscapes (Balling & Falk, 1982); or as suggested by Sample C in this study,
an aversion. The data strongly support the idea that life experience modifies
preference (a variable rarely taken into consideration in much of the current
research on landscape preference where samples are drawn primarily from popu-
lations of self-selected college students). Such an interpretation is generally
reinforced in Samples A and B where rainforest was the second most preferred
biome. By contrast, Americans of comparable age gave the rainforest a low
preference rating and deciduous forest high ratings. However, in Population
C, rainforest dropped to the least preferred position despite the fact that sub-
jects were highly familiar with this biome. One explanation is that this aver-
sion was brought about by a purposeful rejection of their “bush” roots by these
newly urbanized youth. Whatever the reason for the aversion by the urban
Population C to rainforest, the main effect remains—all three of these groups
demonstrated an overwhelming preference for a savanna environment; an
environment with which most were totally unfamiliar. This evidence strongly
supports the hypothesis that some kind of an innate, biological preference
exists for the savanna landscape, despite familiarity with another biome and

Downloaded from http://eab.sagepub.com at OREGON STATE UNIV LIBRARY on June 21, 2010
Falk and Balling 489

regardless of prior experience with the savanna landscape. The data do sug-
gest that landscape preference is a complex amalgam of factors, with innate
preferences forming a foundation which is then overlain by both sociocultural
and personal experience factors.
Other theories can be used to supplement our understanding of these find-
ings, including Gibson’s (1979) affordances theory, the coherence, legibil-
ity, mystery, complexity theory of Kaplan and Kaplan (1989), and Ulrich’s
(1977) five factors theory. All of these hypotheses share an underlying con-
ceptual framework rooted in a Darwinian view of landscape preference. It is
highly likely that something as behaviorally complex as a habitat preference
involves multiple genetic loci, and as indicated above is strongly influenced
by the interaction of the phenotype with the environment. Accordingly, factors
such as affordance, coherence, legibility, mystery, complexity, and sensitivity
to the fractal properties of the visual environment cannot be ruled out as
important psychological dimensions associated with a genetically influenced
ability to discriminate between preferred and nonpreferred landscapes.
In conclusion, the results of this study plus the equally compelling data
from our earlier American research support our hypothesis that humans,
regardless of origin and experience begin life with a preference for savanna-like
environments. The data strongly suggest that human evolutionary history has
had a major influence not only on such physical features as human posture,
brain size and diet but also on some of our most fundamental behaviors and
preferences as well.

Authors’ Note
Special thanks to G. Wokeh, D. DuBey, and B. Junge, without whose assistance these
data could never have been collected. Thanks also to the Smithsonian Institution who
provided sabbatical leave for J.F., allowing conduct of this research. Finally thanks to
Steve Kaplan for encouraging J.F. to publish this research after all these years.

Note
1. Images were identical to those used and previously published by Balling and Falk
(1982).

References
Appleton, J. (1975). The experience of landscape. New York: John Wiley.
Appleton, J. (1994). How I made the world: Shaping a view of landscape. Elloughton:
University of Hull Press.
Balling, J. D., & Falk, J. H. (1982). Development of visual preferences for natural
landscapes. Environment and Behavior, 14, 5-28.

Downloaded from http://eab.sagepub.com at OREGON STATE UNIV LIBRARY on June 21, 2010
490 Environment and Behavior 42(4)

Barker, J. S., Starmer, W. T., & Fogleman, J. C. (1994). Genotype-specific habitat selec-
tion for oviposition sites in the cactophilic species Drosophila buzzatii. Heredity,
72, 384-395.
Butzer, K. W. (1977). Environment, culture, and human evolution. American Scientist,
65, 572-584.
Chamberlain, A. T. (2000). On the evolution of human aesthetic preferences.
Retrieved July 3, 2008, from http://www.assemblage.group.shef.ac.uk/5/chamberl
.html#paper
Davies, K. (2001). Cracking the genome: Inside the race to unlock human DNA.
New York: Free Press.
Day, H. (1967). Evaluations of subjective complexity, pleasingness and interestingness
for a series of random polygons varying in complexity. Perception and Psycho-
physics, 2, 281-286.
Dutton, D. (2003). Aesthetics and evolutionary psychology. In J. Levison (Ed.), The
Oxford handbook for aesthetics. New York: Oxford University Press.
Falk, J. H. (1977). The frenetic life forms that flourish in suburban lawns. Smithson-
ian, 8, 90-96.
Foley, R. A. (1996). An evolutionary and chronological framework for human social
behavior. Proceedings of the British Academy, 88, 95-117.
Fox, C. W. (1993). A quantitative genetic analysis of oviposition and larval perfor-
mance in the bruchid beetle Calloscobruchus maculatus. Evolution, 47, 166-175.
Garro, L. C. (1986). Intracultural variation in folk medical knowledge: A comparison
between curers and non-curers. American Anthropologist, 88, 351-370.
Gerber, E. (1975). The cultural patterning of emotions in Samoa. Unpublished doc-
toral dissertation. University of California, San Diego.
Gerber, E. (1985). Rage and obligation: Samoan emotion in conflict. In G. Whilte,
and J. Kirkpatrick (Eds.), Person, self and experience (pp. 121-167). Berkeley:
University of California Press.
Gibson, J. J. (1979). The ecological approach to visual perception. Boston: Houghton-
Mifflin.
Hagerhall, C. M., Purcell, T., & Taylor, R. (2004). Fractal dimension of landscape
silhouette outlines as a predictor of landscape preference. Journal of Environmen-
tal Psychology, 24, 247-255.
Han, K.-T. (2007). Responses to six major terrestrial Biomes in terms of scenic
beauty, preference, and restorativeness. Environment and Behavior, 39, 529-556.
Heath, T., Smith, S. G., & Lim, B. (2000). Tall buildings and the urban skyline: The
effect of visual complexity on preferences. Environment and Behavior, 32, 541-556.
Herzog, T. R., Herbert, E. J., Kaplan, R., & Crooks, C. L. (2000). Cultural and devel-
opmental comparisons of landscape perceptions and preferences. Environment
and Behavior, 32, 323-346.

Downloaded from http://eab.sagepub.com at OREGON STATE UNIV LIBRARY on June 21, 2010
Falk and Balling 491

Holt, R. D. (1987). Population dynamics and evolutionary processes: The manifold

roles of habitat selection. Evolutionary Ecology, 1, 331-347.
Hyams, E. (1971a). A history of gardens and gardening. London: J. M. Dent.
Hyams, E. (1971b). Capability Brown and Humphrey Repton. London: J. M. Dent.
Jaenike, J., & Holt, R. D. (1991). Genetic variation for habitat preference: Evidence
and explanations. American Naturalist, 137, 67-90.
Jellicoe, G., & Jellicoe, S. (1975). The landscape of man. New York: Viking Press.
Kaplan, R., & Herbert. E. J. (1987). Cultural and sub-cultural comparisons in prefer-
ences for natural settings. Landscape and Urban Planning, 14, 281-293.
Kaplan, R., & Kaplan, S. (1989). The experience of nature: A psychological per-
spective. New York: Cambridge University Press.
Kaplan, S. (1987). Aesthetics, affect, and cognition: Environmental preference from
and evolutionary perspective. Environment and Behavior, 19, 3-32.
Kaplan, S. (1992). Perception and landscape: Conceptions and misconceptions. In
J. L. Nasar (Ed.), Environmental aesthetics (pp. 45-63). Cambridge, UK: Cambridge
University Press.
Kingston, J. D., Marino, B. D., & Hill, A. (1994). Isotopic evidence for neogene
hominid paleoenvironments in the Kenya rift valley. Science, 264, 955-959.
Klopfer, P. H. (1969) Habitats and territories. New York: Basic Books.
Lander, E. (2001). Interview. In E. Arledge & J. Cort (Writers & Producers), Cracking
the code of life. Boston: NOVA, WGBH.
Leakey, M. G., Feibel, C. S., McDougall, E., & Walker, A. (1995). New four-million-
year-old hominid species from Kanapoi and Allia Bay, Kenya. Nature, 376, 565-571.
Levy, R. I. (1973). Tahitians, mind and experience in the Society Islands. Chicago:
University of Chicago Press.
Lucy, J., & Schweder, R. A. (1979). Whorf and his critics: Lingusitic and non-linguistic
influences on color memory. American Anthropologist, 81, 581-615.
Lutz, C. A. (1982). The domain of emotion words on Ifaluk. American Ethnologist, 9,
113-128.
Lutz, C. A. (1986). Emotion, thought and estrangement: Emotion as cultural category.
Cultural Anthropology, 15, 405-436.
Lutz, C. A. (1990). Unnatural emotions: Everyday sentiments on a Micronesian atoll
and their challenge to Western theory. Chicago: University of Chicago Press.
MacCallum, C. J., Nurnberger, B., Barton, N. H., & Szymura, J. M. (1998). Habitat
preference in the Bobina hybrid zone in Croatia. Evolution, 52, 227-239.
Martin, T. E.(1998). Are microhabitat preferences of coexisting species under selec-
tion and adaptive? Ecology, 79, 656-670.
Myers, J. L. (1972). Fundamentals of experimental design (2nd ed.). Boston: Allyn &
Bacon.
Newton, N. (1971). Design of the land. Cambridge, MA: Belknap Press.

Downloaded from http://eab.sagepub.com at OREGON STATE UNIV LIBRARY on June 21, 2010
492 Environment and Behavior 42(4)

Orians, G. (1980). Habitat selection: General theory and applications to human behav-
ior. In J. S. .Lockard (Ed.), The evolution of human social behavior (pp. 49-63).
Amsterdam: Elsevier.
Orians, G. H., & Heerwagen, J. H. (1992). Evolved responses to landscapes. In
J. H. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolution-
ary psychology and the generation of culture (pp. 555-579). New York: Oxford
University Press.
Partridge, L. (1978). Habitat selection. In J. R. Krebs & N. B. Davies (Eds.)
Behavioral ecology: An evolutionary approach. Sunderland, MA: Sinauer.
Petit, L. J., & Petit, D. R.(1996). Factors governing habitat selection by prothonotary
warblers: Field tests of the Fretwell-Lucas Models. Ecological Monographs, 66,
367-387.
Rausher, M. D. (1984). The evolution of habitat preference in subdivided populations.
Evolution, 38, 596-608.
Ridley, M. (2000). Genome: The autobiography of a species in 23 chapters. New York:
Perennial.
Rogers, A., & Jorde, R. B. (1995). Genetic evidence and modern human origins.
Human Biology, 67, 1-36.
Rourke, T. (2007). The application of affordance theory to explain the landscape pref-
erences of travelers. Unpublished Doctoral Dissertation, Clemson University, SC.
Scott, L. (1990). Palynological evidence for Late Quaternary environmental change
in South Africa. Proceedings of the IXth biennial conference of the Southern African
Society for Quaternary Research, 21, 259-268.
Singer, M. C., & Thomas, C. D. (1988). Heritability of oviposition preference and its
relationship to offspring performance within a single insect population. Evolution,
42, 977-985.
Sharp, D., Cole, M., & Lave, J. (1979). Education and cognitive development: The
evidence from experimental research. Monographs of the Society for Research in
Child Development, 44 (1-2, Serial No. 178).
Sommer, R., & Summit, J. (1995). An exploratory study of preferred tree form. Envi-
ronment and Behavior, 27, 540-557.
Stroud, D. (1975). Capability Brown. Lonon: Faber & Faber.
Summit, J., & Sommer, J. (1999). Further studies of preferred tree shapes. Environ-
ment and Behavior, 31, 550-576.
Ulrich, R. S. (1977). Visual landscape preference: A model and application. Man-
Environment Systems, 7, 279-293.
Wecker, S. C. (1963). The role of early experience in habitat selection by the prai-
rie deer mouse, Peromyscus maniculatis bairdi. Ecological Monographs, 33,
307-325.
Wells, S. (2002). The Journey of Man. Princeton, NJ: Princeton University Press.

Downloaded from http://eab.sagepub.com at OREGON STATE UNIV LIBRARY on June 21, 2010
Falk and Balling 493

White, T. D., Suwa, G., & Asfaw, B. (1994). Australopithecus ramidus, a new species
of early hominid from Aramis, Ethiopia. Nature, 371, 306-312.
Wierzbicka, A. (1999). Emotions across languages and cultures: Diversity and uni-
versals. Cambridge: Cambridge University Press.
Wohlwill, J. F. (1968). Amount of stimulus exploration and preference as differential
functions of stimulus complexity. Perception and Psychophysics, 4, 307-312.

Bios
John H. Falk is Sea Grant professor of free-choice learning at Oregon State University,
Corvallis, Oregon. Before joining the faculty at Oregon State University, he founded
and directed the Institute for Learning Innovation; was an early child science educator
at the University of Maryland; and spent 14 years at the Smithsonian Institution where
he held a number of senior positions, including special assistant for education to the
assistant secretary for science and director, Smithsonian Office of Educational
Research. He is known internationally for his research on free-choice learning, the
learning that occurs in settings like museums, parks, and on the Internet. He earned
his graduate and postgraduate degrees in zoology and a joint doctorate in biology and
science education, all from the University of California, Berkeley.

John D. Balling is vice president for Integrated Technology Services at Willamette

University in Salem, Oregon. Other information technology positions he has held
include terms as director of computing services at Dickinson College, head of
academic computing at Bowdoin College, and director of the information resource
center at the Smithsonian Institution. He has also served as a research psychologist
for the Smithsonian Institution and the Harvard Institute for International Development.
He was a member of the faculty in psychology at the University of Wisconsin–
Madison. He earned his bachelor’s degree in psychology from Northwestern University
and his postgraduate and doctorate degrees in psychology from the University of
Massachusetts–Amherst.

Downloaded from http://eab.sagepub.com at OREGON STATE UNIV LIBRARY on June 21, 2010

View publication stats