architectural lessons from environmental psychology… · architectural lessons from environmental...

Architectural Lessons From Environmental Psychology:The Case of Biophilic Architecture

Yannick JoyeFree University of Brussels

A review of findings from the field of environmental psychology shows that humans areaesthetically attracted to natural contents and to particular landscape configurations.These features are also found to have positive effects on human functioning and canreduce stress. However, opportunities for contact with these elements are reduced inmodern urban life. It is argued how this evolution can have subtle but nontrivial adverseeffects on psychological and physiological well-being. These can be countered byintegrating key features of natural contents and structural landscape features in the builtenvironment. Several practical proposals are discussed, ranging from literal imitationsof natural objects (such as plants) to the use of nature’s fractal geometry in anarchitectural context.

Keywords: biophilia, environmental psychology, fractal architecture, biophilic archi-tecture

Natural objects, shapes, and processes haveoften acted as a source of inspiration throughoutthe history of architecture. Perhaps the mostobvious example of this inspiration is ornament,which often contains representations that areclosely similar to, or reminiscent of, the animaland plant world. Besides such literal imitations,some architects, notably Antonı Gaudı, drewlessons from the structural forces governing nat-ural structures, resulting in efficient and eco-nomically built architecture (e.g., Sweeney &Sert, 1960). Today, there seems to be a renewedinterest in the relation between nature and ar-chitecture, especially in zoomorphic or biomor-phic architecture (e.g., Feuerstein, 2002). Morespecifically, such architecture makes use of dig-ital design software, which allows one to easilyrecreate the curvy shapes and geometry that arecharacteristic of natural entities (Lynn, 1999).

This article affirms the importance of naturalform as a perennial source of inspiration forarchitecture. In fact, the main conclusion of thisstudy is that nature-based forms and organiza-

tions in architecture are valuable for humanemotional and cognitive functioning. However,the exact way in which this conclusion isreached differs in an important respect from thenarratives and arguments proposed in theoriesof organic and biomorphic architecture, whichare often more philosophical (Lynn, 1998) oreven pseudophilosophical (e.g., Steiner, 1999)in nature. In contrast, the argument for nature-based forms in architecture in the current studyis mainly based on empirical findings from di-verse psychological subdisciplines. In particu-lar, the article starts with a concise review ofempirical findings from the field of environ-mental psychology and aesthetics. This surveyreveals that humans have an emotional relationwith natural elements and shows that contactwith natural form is in a sense good for humanpsychological and physiological functioning. Itis argued that, by architecturally mimicking nat-ural forms and structural organizations of natu-ral settings, these beneficial effects can betapped in a built context.

The Psychoevolutionary Framework

Different psychological subfields study thehuman relation with nature. For example, evo-lutionary psychologists argue for the existenceof cognitive modules that are specialized inconceptual and perceptual knowledge about

Yannick Joye, Business Economics and Strategic Policy(BEDR), Free University of Brussels, Belgium.

Correspondence concerning this article should be ad-dressed to Yannick Joye, Business Economics and StrategicPolicy, Pleinlaan 2, B-1050 Brussels, Belginm. E-mail:[email protected] or [email protected]

Review of General Psychology Copyright 2007 by the American Psychological Association2007, Vol. 11, No. 4, 305–328 1089-2680/07/$12.00 DOI: 10.1037/1089-2680.11.4.305

305

natural entities. Such cognitive devices areclaimed to have evolved to handle the survival-related challenges and opportunities that werepresent in the natural settings in which humanancestors lived (e.g., finding food; Mithen,1996; Pinker, 1994). Scott Atran (1995) argueshow cross-cultural similarities in human folkbi-ologies support the existence of such a system.In particular, Western and non-Western individ-uals seem to classify nature in similar ways andconsistently ascribe essences to the taxonomictypes of folkbiologies. Atran’s view is consis-tent with the literature on category-specific def-icits, in which people are reported to have de-ficient perceptual and conceptual knowledgeabout the category of living things. One of theinterpretations of the causes of such deficits isthat, under evolutionary pressures, specific neu-ral areas have become specialized in informa-tion about living entities (e.g., Caramazza &Shelton, 1998).

Perhaps the psychological field that has mostprofoundly studied the human (affective) psy-chological relation with natural entities is envi-ronmental psychology. This research area drawson numerous empirical studies and is, therefore,less speculative than the just-mentioned modu-larity thesis. One of the central issues of envi-ronmental psychology is how different types ofsettings can trigger different affective states inindividuals (e.g., liking or preference reactions).Two important proposals have been advancedwith regard to the specific process underlyingthese emotional states. In the preference matrix,developed by Stephen and Rachel Kaplan, theoccurrence of such states is to a large extent theresult of cognitively assessing whether certaininformational features are present in a setting(R. Kaplan & Kaplan, 1989). This contrastswith a central tenet of Roger Ulrich’s (1983)psychoevolutionary framework. In this model,which will form the backbone of the currentstudy, affective responses toward environmen-tal settings are not mediated by cognition butstem from a rapid, automatic, and unconsciousprocess by which environments are immedi-ately liked or disliked. These fast affective re-actions are claimed to be rooted in human evo-lutionary history and are essentially adaptive:They motivated the organism to quickly under-take actions that contributed to its well-beingand survival. For example, if early humanscame across a setting containing an important

risk (e.g., turbulent water, a predator), this trig-gered negatively toned affective reactions (e.g.,dislike), ultimately leading to avoidance behav-ior. On the other hand, if a setting offered goodopportunities for survival and reproduction, thiswould have caused liking reactions, leading toexplorative behavior.

In agreement with Ulrich’s (1983) model,empirical evidence shows that environments(e.g., urban vs. natural) are processed accordingto their affective valence. Moreover, this pro-cess seems to occur very rapidly, which sup-ports the immediate and automatic character ofthese affective responses (Hietanen & Korpela,2004; Korpela Klemettila, & Hietanen, 2002).According to the psychoevolutionary frame-work, survival chances further increased if theseemotional reactions had an inherited compo-nent: No precious time and energy had to bespent learning what kinds of environments wereeither beneficial or harmful (S. Kaplan, 1987,1988; Ulrich, 1983). With regard to the neuralorigin of these affective states, some researchersattribute an important role to subcortical areas,especially the amygdala. Because these struc-tures are also involved in modulating stress-related hormones, it provides an explanation ofwhy certain types of settings have a different in-fluence on autonomic stress responses (Parsons,1991; see also Joye, 2007, for an in-depth discus-sion of this issue).

Aesthetic Preference and StructuralLandscape Features

What is the character of the settings or ele-ments that can trigger such immediate affectivestates? The literature states that, on the onehand, these reactions can be provoked by sometypical structural landscape features. Althoughcoming from a different research field, geogra-pher Jay Appleton was one of the first to pro-pose a model addressing this issue (Appleton,1975). According to Appleton’s prospect–refuge theory, human beings’ preference forlandscapes correlates with two environmentalqualities: prospect and refuge. The notion ofprospect refers to settings or landscape elementsthat facilitate obtaining information about theenvironment. A typical example is a hill, whichaids to visually access and inspect the surround-ing area. On the other hand, refuge points tosettings that can provide shelter and protection.

306 JOYE

An evident example is a cave, which can protectagainst predators and weather conditions.

Ulrich’s psychoevolutionary framework listssome other visual cues that are associated withimmediate positive affective reactions: com-plexity, gross structural properties (e.g., pat-terns), depth properties, ground surface andtexture, absence of threats, and deflected vista(Ulrich, 1983). The predictors in Rachel andStephen Kaplan’s preference matrix overlap toa certain extent with the variables listed byUlrich. The Kaplans’ model (R. Kaplan &Kaplan, 1989; S. Kaplan, 1987, 1988) describestwo types of postures toward the environment.An individual can be actively involved in anenvironment: One can, for example, explore thesetting. Alternatively, an individual can try tounderstand the environment. The Kaplans arguethat these two attitudes are associated with fourstructural landscape properties, each of whichcorrelates with positive aesthetic evaluationsand positively influences landscape selections.The structural properties that facilitate involve-ment in the environment are complexity andmystery. Stephen Kaplan defines complexity asa measure for “how much is ‘going on’ in aparticular scene, how much there is to look at”(1988, p. 48). Mystery refers to settings whoselayout suggests that more information can beacquired if the scene is penetrated more deeply.An example of a mysterious landscape quality isa deflected vista, such as a winding trail. Struc-tural properties that facilitate understanding theenvironment are coherence and legibility. Co-herence refers to features that contribute to theorganization, understanding, and structuring ofthe landscape image, such as symmetries, re-peating elements, and unifying textures. Finally,legibility refers to landscape qualities that helpto predict and maintain orientation in the land-scape as one further explores it. Think, for ex-ample, of a prominent rock functioning as anorientation point.

The Aesthetic Appeal of Natural Contents

In addition to the previous structural land-scape features, the field of environmental psy-chology also studies the natural contents thatcontribute to the aesthetic qualities of settings:namely (calm) water features and vegetativeelements. The explanatory framework, again, isessentially evolutionary. These elements are

liked because they contributed to the survivaland reproduction of early humans. Flowers, forexample, signaled the presence of food sourcesand were cues for future foraging sites. Theyalso helped in differentiating between differentvegetation types, because plants that are notblooming often look quite similar (Orians &Heerwagen, 1992). Trees protected against sunand rain and offered early humans prospects onthe surrounding landscape and retreats frompredators (Appleton, 1975; Orians & Heerwa-gen, 1992; Summit & Sommer, 1999).

These benefits can explain why vegetativeelements and settings containing vegetation stillcause aesthetic or liking reactions. Differentempirical studies show that individuals consis-tently prefer natural, vegetated landscapes overurban settings without vegetation. When urbanenvironments are mutually compared, highestpreference is associated with urban settings con-taining some vegetation, especially trees, or a wa-ter feature (Smardon, 1988; Thayer & Atwood,1978; Ulrich, 1986). Such phytophilia is also clearfrom the observation that nonnatural environ-ments (e.g., home and working interiors) oftencontain actual vegetative elements or (decorative)references to natural content (Eibl-Eibesfeldt,1989; Heerwagen & Orians, 1986). Although em-pirical research on preferential reactions towardflowers is scarce, some studies show that theseelements are indeed associated with positive aes-thetic reactions (e.g., Haviland-Jones, Rosario,Wilson, & McGuire, 2005; Todorova, Asakawa,& Aikoh, 2004).

Still, a possible critique of preferential re-search into greenery is that when vegetatedlandscapes and nonvegetated (urban) architec-tural settings are mutually compared, the lattermost often involve representations of quitemodern buildings or at least buildings that arenot very rich in form. However, it can bepointed out that nature is often characterized bya typical sort of geometry, or fractal geometry(see later discussion). This type of geometryoften does not apply to modern buildings ormodern urban settings. A possibility that needsto be entertained is that the preference for veg-etated scenes is not due to the fact that it is anatural setting but, instead, that it must (to acertain extent) be drawn back to the underlyinggeometric features of the scene. It would, there-fore, be interesting to compare natural settingswith buildings or urban scenery that emulate a

307ARCHITECTURAL LESSONS FROM ENVIRONMENTAL PSYCHOLOGY

more natural geometry, such as Gothic cathe-drals. (Note that this critique does not under-mine our plea for nature-based, or biophilic,architecture but instead only strengthens it.)

The Savanna Hypothesis

In habitat theory, savannas are claimed tobe the settings in which early humans spent asubstantial part of their evolutionary history,and these seem to display an ideal mix of thepreviously discussed structural landscape fea-tures and natural contents (Van den Berg,2004). This type of biome can be broadlydescribed as low to intermediately complexsettings, having a relatively even and grassyground surface dotted with scattered trees ortree groups. Savannas contain a high degreeof biomass and meat, and these are relativelyeasily accessible for terrestrial beings (as op-posed to, e.g., tropical forests). Furthermore,the openness of savannas facilitates detectingpredators and game and is conducive ofmovement and a nomadic lifestyle (Orians,1980; Orians & Heerwagen, 1992).

The aesthetic preference for savanna-typelandscapes has been the subject of a fewempirical studies. Balling and Falk (1982)found that young individuals preferred savan-nas over other biomes without ever beingexposed to the former type of landscape. Theresearchers hypothesize that these findingscould well point to an innate (aesthetic) pref-erence for savannas (see also: Synek & Gram-mer, 1998; but see Coss, 2003, for contrastingfindings). Consistent with the savanna hy-pothesis, research indicates that people tendto prefer tree shapes characteristic of high-quality savannas: These typically have a lowtrunk, a broad canopy, and a moderate canopylayering (Orians & Heerwagen, 1992; Sommer& Summit, 1995). Inquiries into the evolutionof artists’ work (e.g., John Constable) show anincrease of conspicuous savanna features overtime (e.g., opening up views; Heerwagen &Orians, 1993). Furthermore, studies indicatethat when artificial changes are made to plantsand trees, these increasingly come to resemblesavanna-type vegetation (Heerwagen & Orians,1993). Finally, areas that are created for recre-ational or aesthetic purposes (e.g., parks or golfterrains) often resemble savannas (Orians,1980).

Within the field of landscape aesthetics, thesavanna hypothesis is often taken for grantedand has remained mostly undisputed. Neverthe-less, we find it troubling that almost no attentionis paid to discussions in the field of paleoan-thropology. For instance, Wilson (1993) argueshow our preferences for nature, savannas inparticular, are remnants of paleohominid andearly Homo evolution in this type of biome, aview shared by many in the field of habitattheory and landscape aesthetics. Yet there is noconsensus on the claim that the savanna is theunique environment of evolutionary adapted-ness. In his review, Potts (1998) sketches amore complex view that is supported by scien-tific environmental analyses. It evidences that,during the evolution of early hominids, therewas quite some variation in the environmentsthat were inhabited, ranging from forests tosavannas to open-canopy woodlands. Still, itcould be countered that the truth value of thesavanna hypothesis does not have any bearingon the finding that humans adapt to and displaypositive affective affiliations with natural envi-ronments. As Kahn (1999, p. 39) points out,“The evolutionary account can hold, but thesavanna hypothesis needs to give way to abroader account of genetic predispositions toinhabited landscapes.”

Naturalness and Stress Reduction

Besides causing liking responses, natural el-ements (e.g., vegetation and water features) arealso found to contribute to the restoration ofhuman individuals. Two major interpretationsof restorative responses have been proposed.The first, attention restoration theory (ART),was developed by the Kaplans (e.g., R. Kaplan& Kaplan, 1989). Essentially, ART interpretsrestoration as the recovery of directed attentionor the ability to focus. This capacity is deployedduring tasks that require profound concentra-tion, such as proofreading or studying. Naturalsettings have been found to be ideally suited torestore or rest directed attention (e.g., Hartig,Evans, Jamner, Davis, & Garling, 2003; HartigMang, & Evans, 1991).

The second major interpretation of restora-tion is a part of Ulrich’s psychoevolutionaryframework. In this view, restoration applies to amuch broader context than attentional capaci-ties (e.g., Parsons, 1991; Ulrich, 1993; Ulrich et

308 JOYE

al., 1991). More specifically, Ulrich under-stands restoration as stress reduction, and stresscan occur even when directed attention is notfatigued. Within Ulrich’s model, restorative re-sponses are explained by the fact that earlyhumans were often confronted with threateningand demanding situations (e.g., a predator). Asdiscussed, such confrontations lead to the quickonset of negatively toned affective reactions andcorresponding adaptive behavior. Ulrich (1993)notes that the immediate effects of such re-sponses are beneficial for the individual. Yetsuch reactions also have a certain cost in thatthey lead to physiological and psychologicalstress (e.g., high blood pressure, feeling de-pressed). When the threat has vanished, theindividual is in need of restoration from thestress that has been caused. The benefits of suchrestorative responses are “a shift toward a morepositively toned emotional state, mitigation ofdeleterious effects of physiological mobiliza-tion (reduced blood pressure, lower levels ofcirculating stress hormones), and the rechargingof energy expended in the physiological arousaland behavior” (Ulrich, 1993, p. 99). These re-storative responses typically occurred in naturalunthreatening (savanna-like) settings. Suchopen, low-risk environments often contained a(calm) water feature and sometimes had a smallfire. Restoration was also facilitated by the avail-ability of resources, which reduced stress relatedto the uncertainty of finding food (Ulrich, 1993).

The stress-reducing effect of nature is stilleffective today because those individuals whocould respond restoratively to stressful situa-tions survived better. In an often-cited article inScience, Ulrich (1984) discusses a study of hos-pital patients who had undergone gall bladdersurgery and had rooms with views of either asmall tree group or of a brown brick wall. Asopposed to patients with the brick wall view,patients with the tree view had shorter hospitalstays, received fewer negative comments fromthe nurses, required less moderate and stronganalgesics, and had slightly fewer postoperativecomplications. (For further research into therelation between stress reduction and nature,see, e.g., Custers, 2006; Hartig et al., 2003;Parsons, Tassinary, Ulrich, Hebl, & Grossman-Alexander, 1998; Ulrich, 1981; Ulrich et al.,1991).

The Value of Nature-Based, or Biophilic,Architecture

Although there is solid empirical evidencethat humans hold positive affiliations with aspecific set of landscapes and natural elements,this does not preclude that some natural featuresor occurrences also cause more negative andeven aversive reactions in humans (e.g., Mineka& Ohman, 2002; Van den Berg & ter Heijne,2005). Another issue is that the experimentaloutcomes are sometimes difficult to integrateinto a coherent, overarching conceptual frame-work. For example, what is the relation betweenabstract structural landscape features and thepreference for water elements? Other issues arethat differences in nature appreciation are oftenleft in the dark in these discussions (but see Vanden Berg, 2004), and it is still a matter of debatein which sense the genetic component of these(positive) affiliations should be understood(Cummins & Cummins, 1999). Still, whateverthe outcome of these matters, the general pic-ture emerging from the previous concise reviewis that humans have a (partly) hardwired emo-tional affiliation with certain classes of naturalobjects. Some researchers have argued aboutthe affective relation with natural elements andlandscapes in terms of biophilia (e.g., Kellert &Wilson, 1993; Wilson, 1984). Although the the-oretical merits of this term have been ques-tioned (Joye, 2007), in the remainder of thisreport the notions biophilic and biophilia willnevertheless be used as synthetic concepts.

The occurrence of biophilic respondingstands in sharp contrast with the observationthat there is increasingly less contact with na-ture in Western technologically oriented societ-ies. Wolff, Medin, and Pankratz (1999) foundthat such an evolution has nontrivial effects oncultural expressions of nature. In particular,they made a historical study of word use indictionaries and found that, from the 20th cen-tury onward, the use of (folk)biological termsdevolved, and their application lost precision. Incontrast, several nonbiological terms evolvedduring this period (e.g., books, clothes, furni-ture). Apart from being associated with an im-poverishing conceptual framework for naturalobjects, it is also plausible that reduced contactwith nature can be accompanied by a reducedknowledge of the rich variety of forms charac-teristic of natural entities. A probable artistic or


creative consequence is that the formal curric-ulum of artists and architects becomes nar-rower. The reason is that natural form can beconsidered as a creative or compositional gram-mar, which can be used for creating artwork, or,as Stephen Kellert put it, “The aesthetics ofnature can function as a kind of monumentaldesign model” (Kellert, 1997, p. 36). The loss ofthis monumental design model has its architec-tural counterpart in modern urban settings,which are increasingly governed by euclideangeometry and stripped of ornament, patterning,detailing, and color (Salingaros, 2004). Archi-tectural references to nature can help put an endto this uniformity. By encouraging architects tointegrate natural forms and patterns in theirwork, they are motivated to study nature’sshapes and compositional rules, and this canenrich their creative curriculum.

Besides having creative consequences, re-duced contact with natural form could also sub-tly influence the way in which people thinkabout the world. Inquiries into semantic mem-ory indicate that processing conceptual infor-mation about living things mainly relies on per-ceptual information (e.g., the concept “zebra”activates perceptual information, such as“stripes”), whereas processing nonliving thingsor artifacts depends on functional information(e.g., the concept “knife” activates functionalinformation, such as “cutting”; e.g., Crutch &Warrington, 2003; Farah & McClelland, 1991).These findings could have important implica-tions. The presence of nonnatural things, andespecially artifacts (e.g., cell phones, comput-ers, chairs, pots, printers), is ever increasing inthe human living environment at the expense ofnatural structures or entities. A probable conse-quence is that neural areas related to an object’sfunctionality and hence functional analyses(i.e., how an object should be used or manipu-lated) are becoming increasingly more domi-nant in our thinking about the constituents of themodern living environment. As we becomemore acquainted with such thinking, it is notimplausible that it will be deployed in otherdomains as well (e.g., to generate explanations).This could especially occur when knowledgeabout phenomena in a certain domain, such asthe natural world, becomes increasingly morescarce or more underdeveloped.

Although functional postures are importantand necessary in certain fields, transferring

them to other domains or contexts can proveproblematic and harmful. Today, we can wit-ness how thinking about nature and natural re-sources in terms of things that can be manipu-lated has devastating effects. The upshot is thatthis shifts the balance even further toward func-tional thinking, because nature is replaced byentities that predominantly require functionalanalyses. Probably, this process can be coun-tered by extensive contact with the naturalworld, and developing a rich conceptual frame-work about it (e.g., by nature education). Thiscould help people realize that functional think-ing is not always desirable with regard to na-ture. While being more speculative, it can behypothesized that integrating naturalistic ele-ments in architecture can counteract the increas-ing dominance of functional semantic networksand the associated epistemological attitude. Ad-mittedly, people will not consider biophilic ar-chitecture or design as actual nature. However,such architecture shares some essential formalfeatures with living things, and research indi-cates that perceptual features are important forrecognizing living things. Biophilic designcould lead to more attentiveness to an object’sperceptual qualities, thereby leading attentionaway from its possible functions and the asso-ciated functionalist postures. Furthermore, be-cause of the (hardwired) emotional affiliationwith certain natural elements, nature-based ar-chitecture can awaken fascination for naturalforms. Such an attitude could be ecologicallyrelevant, because it is found that proenvironmentalbehavior is positively influenced by emotionalstates toward nature (Kals, Schumacher, &Montada, 1999).

Without a doubt, people can get used to lessformal diversity in the built environment. How-ever, such a situation is not desirable because anincreasing dominance of uniform (modernist)environments will probably have a number ofpsychological and physiological costs. Recallhow, under evolutionary pressures, naturalforms and environments became associatedwith a broad range of emotions, ranging fromfear to excitement. In the human ancestralworld, such associations promoted fitness be-cause they motivated the organism to undertakeadaptive reactions (e.g., flight). Today, thereseems to be a discrepancy between the habitatshumans have evolved in and modern urban set-tings. For example, it was already noted that the

310 JOYE

former was characterized by, among others, amix of complexity and order (S. Kaplan, 1987,1988; Ulrich, 1983). Yet current architecturalsettings do not appeal to this ordered complex-ity. Modernist architecture mainly consists ofsimple volumetric forms and thus deprives thesenses in their constant search for meaningfulinformation. On the other hand, postmodern anddeconstructive architecture deliberately destroyarchitectural coherence, either by jumblingtogether disparate stylistic and formal elementsor by placing the destruction of coherenceand structure at the heart of the tradition(Salingaros, 2004). Furthermore, modern build-ing is often dictated by efficiency and economicmotives, barely leaving room for symbolic andstylistic references to natural contents (e.g., or-nament; Pinker, 2002; Salingaros, 2004). Inshort, much of the modern built environmentfundamentally lacks (references to) the contentsand structural organization that are characteris-tic of a good habitat. Exposure to such environ-ments could rapidly and automatically triggernegatively toned feelings and the associatedstress-related endocrinal reactions (Ulrich,1983). Although such responses could go bylargely unnoticed because of human habituationto this type of environment, the long-term oc-currence of such stress reactions could haveimportant health effects (Parsons, 1991).

Increasing urbanization undoubtedly has anumber of positive consequences. For example,in modern cities, people come to live closertogether, which could promote social interac-tion and the pleasure and enjoyment associatedwith this (Van den Berg, Hartig, & Staats,2007). Furthermore, there is nothing inherentlywrong or undesirable about modern buildingstyles, and there is no reason to doubt the gen-uineness of positive reactions to such buildings.What, according to the current argument, couldbecome problematic is the (growing) domi-nance of such nonnatural building styles at theexpense of settings with natural form languages(albeit natural or artificial ones). The core argu-ment of the current study is that, by includingelements of ancestral habitats in the built envi-ronment, one can counter potential deleteriouseffects, which stem from this dominance, result-ing in more positive affects and more relaxedphysiological and psychological states. In theremainder of this article, such architectural in-

terventions are denoted as nature-based, or bio-philic, architecture.

Integrating Structural Landscape Featuresin Architecture

How can the structural landscape features,discussed earlier, be meaningfully applied to thebuilt environment? This is a more difficult issuethan applying well-defined natural contents toarchitecture, because the former features are ofa more abstract nature. Furthermore, only veryfew researchers have addressed this issue andproposed clear guidelines on how to success-fully integrate these qualities in architecturalsettings.

First, turn to the type of setting that containsan ideal mix of these structural landscape fea-tures, namely the savanna. An evident strategyto imitate savannas is to integrate photographsor projections of savannas in (interior) spaces.Another, more architectural method consists ofmimicking key structural features of savannas.Possible strategies include creating wide andopen spaces; making variations in the architec-tural topography; integrating clusters of real orsymbolic trees (e.g., columns); and integrating awater feature (e.g., a fountain) or even a smallfire. Note how certain retail settings, such asshopping malls, often contain these elements.Because a major goal of the retail sector isattracting people, it should be no surprise thatorganizational features of preferred settings are(intuitively) deployed in such commercial con-texts (Heerwagen, 2003).

Because of their openness, savannas providedgood prospects on the surrounding area. Fur-thermore, trees typical of savannas (acacias)have low trunks and could, therefore, beclimbed to see across the landscape and to es-cape predators. On the other hand, the broadcanopies provided good protection against sunand rain. Grant Hildebrand (1999) uses Apple-ton’s (1975) notions of prospect and refuge asexplanatory principles for the aesthetic appealof certain buildings. Although Hildebrand doesnot provide exact guidelines, his analyses showwhich spatial organizations influence the pros-pect and refuge dimensions of buildings. Withregard to Frank Lloyd Wright’s house in Talie-sin, Wisconsin, Hildebrand notes the following:“Deep overhanging eaves, alcoves and recesses,the withdrawal of the house in the dense foliage,


and the cave-like masses of stone anchoring thehouse to the hill all convey that this is a havenwithin which one can withdraw secure. Exten-sive bands of window and the balcony reachingout over the falling landscape, moreover, indi-cate that the advantages of generous prospectare likely to be available within” (p. 28). It isclear that feelings of prospect and refuge can beevoked by specific architectural interventions.Strategies for evoking concealment are reduc-ing lighting conditions, lowering ceilings, andmaking small windowless spaces enclosed bythick walls. The prospect dimension depends onopposite characteristics: larger space dimen-sions, raised ceilings, thin transparent walls,wide views on surrounding spaces, building onan elevated site, increased lighting conditions,balconies, and so on.

Prospect and refuge can be linked to thepredictors complexity and coherence, central tothe Kaplans’ preference matrix (e.g., R. Kaplan& Kaplan, 1989). Only a setting that containsenough prominent landscape features (e.g.,trees, rocks) can provide opportunities for ref-uge. On the other hand, if a setting contains toomany elements, this makes it difficult to have aclear view over the landscape. Although com-plexity and coherence have primarily been ap-plied to landscapes, there is empirical evidencethat a balanced presence of both properties con-tributes to the aesthetic qualities of built settings(e.g., Herzog, Kaplan, & Kaplan, 1982). How-ever, how can a complex set of architecturalelements be ordered? Again, take a look at thearchitectural tradition associated with FrankLloyd Wright, namely organic architecture. Inessence, organic architecture is not restricted bystylistic conventions but is characterized by aninherent form freedom. Although not necessary,this often translates in buildings that are quiteirregular and complex both in plan and eleva-

tion. Yet organic architects often use a geomet-ric module (e.g., a triangle) as main composi-tional element (Mead, 1991). In this way, dif-ferent parts of the building are given a similarform, which results in an overall coherence(Eaton, 1998).

It could be noted that merely repeating sim-ilar elements does not guarantee an orderedcomplexity. On the contrary: It can even lead torandom structures, as in deconstructivism. Apossible solution is to organize these (similar)elements through patterns. These are often theresult of only a few simple mathematical oper-ations, such as reflectional, rotational, transla-tional, and glide symmetries. More complexpatterns are obtained when these symmetries arerepeated or when they are nested (Salingaros,2003; Figure 1A–D). Traditionally, patternstake in a prominent place within the organictradition. Historically, they can also be found in,for example, tiling, ornaments, mosaics, stainedglass windows, and (oriental) carpets.

Another structural feature that positively cor-relates with landscape selection is mystery.Some claim that this property can be conveyedby specific design elements: “When appearingaround corners, attached to walls, and hungfrom ceilings, interesting objects, architecturaldetails or motifs, graphics, video displays andartifacts can create a little mystery and surprisein the workplace” (Hase & Heerwagen, 2000, p.30). However, the most straightforward way toapply mystery to an architectural setting is bydeflected vista. This can be realized by lettingthe architectural trail (e.g., corridor) bend away,which can lead to curiosity of what might liebeyond the bend, thereby encouraging explor-ative behavior. Another mode of mystery iscalled “enticement.” Essentially, this notion re-fers to the situation in which a person is in thedark, from where it can see a partially visible

Figure 1. Patterns can be obtained by some simple mathematical transformations (Salinga-ros, 2003): (A) randomness; (B) translational symmetries; (C) reflectional symmetries; (D)rotational symmetries that are nested. (Copyright © Yannick Joye.)

312 JOYE

and enlightened area or setting. Such enlight-ened regions draw attention and trigger explor-ative behavior. Although mysterious settingscan be aesthetically appealing, too much irreg-ularity or surprise can have the result that thelayout of the building becomes confusing andnontransparent, ultimately leading to orientationand way-finding problems. Legibility can beenhanced by integrating signalizations and dis-tinctive markings, by offering views on the out-side, and by making the building shape moreregular (Evans & McCoy, 1998).

Imitating Natural Contents in Architecture

How could natural contents be integrated intothe built environment? Evidently, this can bedone by providing views on the outside envi-ronment, by integrating vegetation in built set-tings, by hanging nature pictures on the wall, bynature-oriented screensavers, and so on. Theseinterventions are what Stephen Kellert (2005)called indirect experiences of nature, and theycome quite close to the design interventionsfrom the field of evidence-based design (Ulrich& Zimring, 2004; Van den Berg & Van Winsum-Westra, 2006). It is in this sense that the modernbuilt environment sometimes imports some of theicons of habitability that are typical of ancestralhabitats. The result is that even architecture that ischaracterized by nonnatural forms can be consis-tent with the current argument. For example, somemodernist architecture (e.g., Mies’s FarnsworthHouse) is characterized by large expanses of glass,by which the building opens up to the surroundingnatural landscape, potentially causing biophilic re-sponses in the inhabitants.

Implanting a building in a natural landscapedoes not necessarily tell us something about thearchitectural form and whether it in some sensedisplays key features of our ancestral habitats.Because we factually inhabit contexts in whichbuildings are often more dominant than nature,it also becomes relevant to come to biophilicinterventions that pertain to the architecturalform. But how should such interventions beconceived? A first strategy is to architecturallyimitate preferred natural entities, such as vege-tative elements. Such imitations can take ondifferent levels of abstraction. A first option isto literally copy these elements in architecturaldesign. As already noted, there is an age-oldtradition to copy nature, especially floral and

vegetal patterns, in traditional ornament. Ad-mittedly, it could be possible that such imita-tions will not be very successful, because theassociated emotional states could quickly befollowed and suppressed by higher order orcultural beliefs. For example, the architecturalcommunity could consider such imitations askitsch. Nevertheless, it should be noted thatthere is often a discrepancy between what isfound appealing by experts and laypersons. Theprimary goal of this study is not to argue forwhat is supposedly fashionable or to defendhigh art but to indicate what could be psycho-logically appealing for the broad public.

An alternative to literal imitations is to createarchitectural designs based on schematic imita-tions of natural elements. These would nolonger be exact copies but artistic interpreta-tions that still contain some global visual simi-larities with regard to the original natural object.One of the central claims of the current study isthat such constructions will be accompanied byaffective states that are similar to those evokedby real natural contents. Orians and Heerwagen(1992, p. 572) expressed it as follows: “Anevolutionary-ecological approach to aestheticssuggests that the incorporation of trees and treeforms, actual or symbolic, into the built envi-ronment should have a strong positive impacton people. . .We predict that the presence ofthese ‘symbolic trees’ is associated with posi-tive response to built environments.”

Although the occurrence of biophilic re-sponses to symbolic representations of naturecould be prima facie plausible, it is problematicthat it is often taken for granted in the literatureon biophilic architecture. Although researchthat has directly tested this prediction is lacking,some indirect arguments can be presented thatsupport the conclusion. First, it is evident thatdomain-specific mechanisms in the brain willbe activated by the objects in which they arespecialized. For example, a face detectionmechanism will be activated by its proper input:actual human faces. Yet it seems that such do-main-specific mechanisms do not care aboutwhether the objects it analyzes are in any sensereal or symbolic. More specifically, these neuralareas also tend to become activated by elementsthat share some central geometric features withthe proper input of the domain-specific systems.This is one reason why a symbolic representa-tion of a face, such as, for example, a smiley


face ( ) or the front of a car, can be perceivedas having facelike features (Pinker, 1997; Sper-ber & Hirschfeld, 2004) and can lead to theonset of similar emotions as real eyes or faces(e.g., Aiken, 1998a). Similarly, it is probablethat the neural mechanisms specialized in pro-cessing natural elements will also be activatedby stimuli that share essential geometric fea-tures with natural elements, such as symbolic orimitative representations of nature in architec-ture. Because of the importance of quickly dis-playing adaptive behavior to natural stimuli(e.g., exploration, escape, fighting), it is proba-ble that at the early stages of processing someaffective processing or priming will alreadytake place, before any conscious recognition ofthe imitated natural elements occurs (Ulrich,1983).

Further reasons why architectural imitationsof nature could trigger biophilic responses aremore empirical in nature. First, it can be pointedout that research on environmental preferencesoften uses simulations of nature (e.g., photos,posters, videos, and even paintings). The resultsthat are obtained with these stimuli are close tothe responses associated with real nature, whichsuggests that realness does not play a decisiverole. (Yet it should be noted that in such con-texts nature is mostly depicted very realistically,and when only realistic representations of na-ture can be used in architecture, this restricts therange of possible architectural interventions al-most exclusively to ornamentation). Second,symbolic representations of nature have beenused throughout the history of art for aestheticenhancements, which suggests that these cantrigger biophilic responses. Third, research in-dicates that preferences for natural settings canbe statistically predicted by underlying geomet-ric characteristics, which lends plausibility tothe claim that geometric abstractions from na-ture can cause the associated affective effects(Hagerhall, Purcell, & Taylor, 2004; see alsoFractals and Biophilic Reactions: A CriticalEvaluation section).

There is a further important reason why sym-bolic or schematic interpretations of naturalnesscan lead to aesthetic reactions. This conclusionis based on the finding that formal abstractionsor simplifications of certain conspicuous traitsof (survival-relevant) stimuli lead to similar,or even stronger, emotional responses as theoriginal (natural) stimulus. Within the field

of ethology, such stimuli are labeled “supernor-mal stimuli” (Tinbergen & Perdeck, 1950).Ramachandran and Hirstein (1999) argue howthis phenomenon is one of the central laws thatartists (unconsciously) deploy in art. They clarifythis principle by referring to a phenomenon ob-served in the field of animal discrimination,namely the peak shift effect. A rat that is taught todiscriminate between a square and a rectangle andis rewarded for discriminating the rectangle willrespond more frequently to the rectangle. How-ever, when the original rectangle is elongated, therat will respond even stronger to this new rectan-gle than to the rectangle that it was taught todiscriminate. According to Ramachandran andHirstein (1999), artwork often taps a similar ef-fect: “What the artist tries to do (either con-sciously or unconsciously) is to not only capturethe essence of something but also to amplify it inorder to more powerfully activate the same neuralmechanisms that would be activated by the orig-inal object” (p. 17). According to Ramachandranand Hirstein, such amplifications can occur alongdifferent dimensions of the artistic work: for ex-ample, form, color, and movement (see alsoAiken, 1998b). It is clear that, in the present dis-cussion, main interest goes to amplification of thearchitectural form, which can cause a peak shifteffect with regard to real natural forms.

Maybe the most well-known examples of ar-chitecture in which schematic interpretations ofnatural forms are present have been created byAntonı Gaudı. For example, the interior col-umns of the Sagrada Familia are quite similar totreelike and flowering structures (Figure 2A–B).Indeed, one can clearly differentiate a stem,which bifurcates into further branches and sub-branches. The canopy of these treelike struc-tures consists of flowering forms, which furtherstrengthens the impression of symbolic vegeta-tion. A more modern architect whose work alsocontains schematic interpretations of natural ob-jects is Santiago Calatrava. Like in Gaudı’swork, structural forces are an important deter-minant of the shape of Calatrava’s architecture.Yet he also seems to be directly inspired by theshapes of nature. According to Von Moos, “Hisarchitecture relates to the morphologies of plantand animal life—on land, in the depth of thesea, or in imagination” (Tischhauser & VonMoos, 1998, p. 338). Particularly relevant forthe present discussion is that several buildingelements resemble vegetative structures. For in-

314 JOYE

stance, both Calatrava’s Orient Station (Lisbon)and his BCE Place (Toronto) could be interpretedas “‘forests’ of structural ‘trees’” (Tzonis, 1999, p.82; Figure 3).

Applying Nature’s Fractal Geometry toArchitecture

What is a fractal? Are there any good rea-sons to go beyond the imitation of natural con-tents? Is there something about the outlook ofnature that can be abstracted away and appliedto architecture while still giving rise to biophilicreactions? We hinted at this already and specu-lated that the underlying (fractal) geometry ofnatural scenes is perhaps a contributing factor toaesthetic and stress-reducing responses. Somescholars adhere to similar ideas. Katcher andWilkins (1993) note that it would be valuable to“search for general characteristics of the pat-terns in nature that produce relaxation. Explor-ing the ability of computer-generated fractalstructures to entrain subjects’ attention and in-duce calm could be a promising approach, aswell, since waves, flames, and clouds can beduplicated by fractals. Fractal structures couldalso relate the physiological and cognitive ef-fects of both natural phenomena such as wavesand cultural artifacts like music” (pp. 177–178).Similarly, Purcell, Peron, and Berto (2001)

speculate that the high restorativeness of naturalscenes could be due to their fractal characteris-tics, whereas built environments are low in re-storativeness because of their underlying eu-clidean geometry.

Fractal geometry has been described and ex-plored since the 1970s (Mandelbrot, 1977). Theterm fractal is derived from the Latin wordfractus, meaning broken or fractured. One of thedefining features of a fractal is that this rough-ness recurs on different scales of magnitude(Figure 4). When zooming in on a fractal, ateach magnification a structure appears that ismore or less similar to the global form of thefractal, a property labeled “self-similarity.” An-other feature that plays an important role withinthe field of fractal geometry is the concept ofdimension. In euclidean geometry, lines have adimension of one, whereas geometric objects,such as squares and triangles, have a dimensionof two, and volumes in space are three dimen-sional. In contrast, the dimension of a fractal, orthe fractal dimension, is not an integer value.For fractals in the plane, the fractal dimensionlies between the first and the second dimension,which gives a value between 1 and 2(e.g., 1.46). For fractals in space, the fractaldimension lies between 2 and 3 (Voss, 1988).Essentially, these noninteger values are due tothe fact that fractal patterns have a very wrin-

Figure 2. The interior of Gaudı’s Sagrada Familia contains schematic interpretations ofnatural contents. Left: columns as treelike structures. Right: flowerlike canopies. (FromGuillaume Paumier. Used with permission).


kled character and, therefore, occupy morespace than a simple line (first dimension) but donot fill the entire plane (second dimension). Inessence, the fractal dimension should be inter-preted as a measure of the degree in which(similar) detail recurs on different scales ofmagnitude. Hence, the concept could be under-stood as a measure of complexity.

Fractals, naturalness, and liking responses.Is there any evidence for the claim that fractalcharacteristics could as well be associated withbiophilic responses to typical natural contents,such as vegetation? In other words, why couldfractal geometry be an important beneficial in-gredient of natural scenes? First, there is a pro-found link between the shape of many naturalstructures and fractals. In essence, the formerare fractal-like in that they often display theself-similarity that is so typical of fractals. Ev-ident examples are trees, mountains, lightning,clouds, coastlines, and so on. A second linkbetween fractals and nature is the observationthat natural elements can be elegantly mimickedwith fractal geometry. For example, plants andtrees can be straightforwardly generated byfractal procedures, such as L-systems (Prusink-

iewicz & Lindenmayer, 1990). Perhaps themost well-known fractal model of vegetativeelements is the Barnsley Fern (Peitgen, Jurgens,& Saupe, 1992), which was discovered byMichael Barnsley. A third way in which fractalsand nature can be related is more psychologicalin nature. Apparently, fractal patterns evokeassociations of naturalness in human subjects(Geake, 1992). Closely related is the findingthat contact with fractals improves the ability toperceptually differentiate between highly simi-lar natural patterns (Geake, 1992).

Not only are fractals perceptually related tonaturalness, but their aesthetic value is also ob-vious to many. These rich and sometimes col-orful images often provoke awe and fascinationin viewers. It should, therefore, be no wonderthat some artists (e.g., Jackson Pollock) have(unconsciously) deployed fractal principles intheir work to reach an aesthetic effect (Mureika,2005; Taylor, 2002). Still, these connectionsbetween aesthetics and fractals are only anec-dotal and intuitive. In search for a strongerfoundation for fractal aesthetics, reference canbe made to preliminary empirical research byRichard Taylor (1998). Taylor mentions that

Figure 3. The “forest of trees” in Calatrava’s Orient Station. (From Inge Kanakaris-Wirtl;www.structurae.de. Used with permission.)

316 JOYE

more than 90% of a group of 120 studentspreferred fractal patterns over nonfractals. Yet itshould also be noted that Arthur Stamps (2002)has tested this conclusion more rigorously anddid not find that fractal patterns were aestheti-cally preferred over nonfractals. Nevertheless, itshould be pointed out that Stamps used fractalcontours, and the self-similarity was not readilyperceivable in the representations.

Other research into fractal aesthetics hasmainly focused on the relation between fractaldimension and aesthetic preference. One of thefirst empirical studies of this relation has beencarried out by Aks and Sprott (1996). The ex-periment revealed that 24 study participantspreferred fractal patterns with a fractal dimen-sion of between 1.17 and 1.38. The averagefractal dimension of the most preferred attrac-tors was 1.26 � 0.06. Spehar, Clifford, Newell,and Taylor (2003) used three different catego-ries of fractal patterns (mathematical fractals,natural fractals, and fragments of Pollock paint-ings) and found that patterns with a fractal di-mension of between 1.3 and 1.5 were preferred

most, as opposed to patterns with dimensionvalues between 1.1 and 1.2 and between 1.6and 1.9. Abraham et al.’s (2003) study revealeda nonmonotonous relation between aestheticpreference and fractal dimension. Patterns withthe highest and lowest fractal dimension wereleast preferred, whereas those with a midrangefractal dimension were liked most. More spe-cifically, attractors with a fractal dimensionranging from 1.4 to 1.6 (mean fractal dimensionof 1.54) received the highest preference ratings.

Although these results need further replica-tion and are to be treated with caution, there isa tendency to prefer patterns with an interme-diate fractal dimension, ranging from about 1.3to 1.5. Note that there are some further hints forthe special status of this range of values. For,example Rogowitz and Voss (1990) found thatrecognizing and finding new shapes in fractalpatterns is best for those patterns with a fractaldimension ranging from 1.2 to 1.4. Geake andLandini (1997) found that the variance in studyparticipants’ judgments of the complexity of

Figure 4. A typical fractal pattern. This is a detail of the Mandelbrot set.


fractal patterns exploded when pictures had di-mension values greater than 1.3.

Fractals and stress reduction. If fractalscan be meaningfully related to aesthetic reac-tions, then is there any way in which thesepatterns can be linked to the capacity of naturalcontents to reduce stress (e.g., Ulrich et al.,1991)? To answer this, it must first be noted thatthere are reasons to assume that restorativenessis the underlying factor for aesthetic reactionstoward natural settings. Van den Berg, Koole,and van der Wulp (2003) have experimentallyconfirmed this hypothesis by performing[m]ediational analyses. . .[which reveal] that af-fective restoration accounted for a substantialproportion of, the preference for the naturalover the built environments (p. 135).. How canthis finding be related to the field of fractalaesthetics? A plausible answer is that if fractalcharacteristics underlie aesthetic responses tonatural settings to a certain extent, and if theseresponses are maximal for an intermediate frac-tal dimension, then it could well be that thisrange of values will also have the highest re-storative potential.

Wise and Taylor (2002; see also Taylor et al.,2005) have carried out a preliminary study totest the relation between fractal geometry andstress reduction. They reexamined a study per-formed by Wise and Rosenberg (1986) involv-ing 24 individuals who were continuously ex-posed to four different patterns: a photograph ofa forest setting, a simplified representation (i.e.,painting) of a savanna landscape, a picture withsquares, and a white plane, which functioned asa control picture. While being exposed to theimages, participants had to undergo three stress-ful mental tasks: an arithmetic task, solvinglogical problems, and creative thinking. Be-tween every task there was a 1-min recoveryperiod. Physiological stress was determined byskin conductance, because research indicatesthat increased conductance correlates withhigher levels of stress.

It was found that the degree of physiologicalstress was dependent on the type of pattern thatwas presented to the participants. Because nat-uralness is found to be a predictor of aestheticand restorative responses, one would expect thatthe picture of the forest setting was most effec-tive in reducing stress (indeed, it looked themost like real nature). Contrary to this expecta-tion, this effect was most effectively produced

by the unrealistic painting of the savanna land-scape. The change in conductance betweenwork and rest periods was 3% lower for theforest photograph and 44% lower for the sa-vanna representation than for the control pic-ture. This means that these pictures dampened(physiological) stress associated with the tasks.

Because the researchers did not expect thisoutcome, they decided to determine the fractaldimension of each of the pictures. Only theforest and savanna pictures had fractal charac-teristics. It was found that the pattern that wasmost effective in stress reduction, the savannapicture, had a fractal dimension that fell withinthe range of dimension values that was earlierfound to correlate with highest aesthetic prefer-ence (Spehar et al., 2003). Because this pictureis only a rough and simplified representation ofa savanna, the authors speculate that the depic-tion of natural contents alone cannot be a suf-ficient condition for a restorative effect. If thiswere the case, then highest restorativenessshould be expected to come from the morerealistic and naturally looking forest setting.Instead, it seems that, besides depicting naturalelements, the scene should also have a specificfractal dimension in order to maximize stressreduction. Specifically, from this experiment, itcan be tentatively concluded that its dimensionvalue should fall within the range of 1.3 to 1.5.

Despite these remarkable results, some ques-tions still remain. First, of course, is the prelim-inary character of these experimental outcomes,which necessitates replication. Second, the lit-erature on habitat theory (e.g., Orians, 1980)claims that humans are innately predisposed toprefer savannas because it is the biome in whichthey thrived for a substantial part of evolution-ary history. Consequently, it could be arguedthat it is quite natural that the savanna pictureleads to the highest restorative responses. Thus,it remains unclear whether it is the fractal di-mension that underlies these responses or thespecific contents depicted in this image. Perhapsthe same experiment should be replicated withfractal patterns, devoid of meaningful represen-tative contents. Again, however, note thatHagerhall et al. (2004) found that preferencesfor settings could be predicted by the fractaldimension. This adds support to Taylor’s claimthat it is the fractal component that underlies therestorative responses, not only the depicted con-tents. Yet it could also be pointed out that

318 JOYE

features such as bodies of water are highlypreferred contents (Ulrich, 1983) that cannot bestraightforwardly analyzed in terms of fractalgeometry. Perhaps something similar applies tothe savanna painting. For example, the typicalshape of savanna trees (low trunk, broadcanopy) could be a highly preferred icon inlandscapes. Perhaps it is a basic “preferendum”(Ulrich, 1983), which is irreducible to fractalcharacteristics. It can also be argued that thestress-reducing character of fractals is alreadyestablished. The reason is that certain fractalsare sometimes very difficult to distinguish fromreal natural elements (e.g., the Barnsley Fern). Itis, therefore, very probable that such naturalisticfractals will lead to biophilic responses. A lot ofnatural entities are fractal, and the main differ-ence with mathematical fractals is that theirself-similarity does not extend to infinity.

The affective value of fractals. The affec-tive value of fractal patterns can be explained bythe finding that naturalness is a predictor ofbiophilic responses and by the fact that fractalgeometry eminently captures this quality.Sometimes it is even claimed that the aestheticeffect of fractals is due to the fact that suchpatterns lead to a peak shift effect (Ramachan-dran & Hirstein, 1999), because they imply anexaggeration of the dimension of recursiveness,which is a characteristic quality of natural form(Joye, 2007; Mureika, 2005). However, whatcould be the explanation for the preference forfractal patterns with low to intermediate dimen-sion values and for their restorative potential?Because Cutting and Garvin (1987) have dis-covered a correlation between the fractal dimen-sion and complexity of fractal patterns, a pos-sible answer is that the fractal dimension offersa quick cue of the complexity of a scene. Com-plexity is a predictor of habitat quality (S.Kaplan, 1987, 1988), and the preference for alow to intermediate fractal dimension could berooted in the fact that habitats of a low tointermediate complexity (e.g., savannas) of-fered the best chances for survival (Wise &Taylor, 2002). Indeed, in such settings, infor-mation can be quite easily grasped and pro-cessed, as opposed to more complex environ-ments (e.g., tropical forests). This reduces thepossibility that crucial information (e.g., preda-tors) will be missed or ignored. On the otherhand, the complexity is high enough to keep oneinterested, to awake further explorative behav-

ior, and to provide opportunities for refuge. It isquite probable that the presence of these prop-erties facilitated restoration, hence the restor-ative responses associated with patterns of anintermediate fractal dimension (Taylor et al.,2005). For example, resting from stressful ordemanding events seems more likely to occur insettings that offer retreats but that also containenough openness, which reduces the probabilitythat one will be attacked by a predator by sur-prise (Ulrich, 1993). If specialized neural mech-anisms exist that assess the quality of a habitat,then it is not too difficult to suppose that thesecompute the fractal dimension in order to havea rapid cue of habitability.

It should be noted that sometimes the aes-thetic appeal of fractal-like patterns is also ex-plained by the fact that the nervous system isgoverned by fractal-like processes. In particu-lar, Anderson and Mandell (1996, p. 114) arguethat human evolution in a fractal world hasrequired “the incorporation of fractal structuresas well as fractal processes, and these in turnwould be integrated into sensory systems, rec-ognition, memory, and adaptive behaviors.”More specifically, the authors describe how hu-man functioning is characterized by a fractalnoise signal—1/f noise—from the microscopiclevel of neural functioning to the macroscopiclevel of human behavior. For the present dis-cussion, the presence of this type of noise in thehuman mind and brain seems especially rele-vant: “In neurobiology in general, and neuro-physiology in particular, 1/f patterns in time areprofound in their recurrent appearance acrossmany levels of organization in the nervous sys-tem, from the underlying cellular dynamics ofion channels and intermittent firing patterns ofneurons to developmental phenomena occurringduring the organization of breathing to globaldynamics in the nervous system such as subcor-tical, transcortical and scalp EEG defining be-havioral states of consciousness” (Anderson &Mandell, 1996, p. 77). Some authors proposethat, because of its fractal nature, the brain isoptimized to process the statistical characteris-tics of natural scenes, which are also found to begoverned by 1/f spectra (e.g., Knill, Field, &Kersten, 1990). For instance, Gilden, Schmuck-ler, and Clayton (1993) found that discriminat-ing fractal contours was best for those sharing(statistical) properties with natural scenes. Con-sistent with this is the finding that neurons in the


V1 area of the brain show a preference for 1/fsignals (Yu, Romero, & Lee, 2005).

Some authors hypothesize that the proposedfractal nature of the human mind/brain can il-luminate the creation of fractal artwork. Essen-tially, such art should be understood as an ex-teriorization of the fractal aspects of brain func-tioning (Goldberger, 1996). However, whatdoes such an account have to say about theaesthetic value of fractals? Different authorshave described the perception of such fractal-like patterns in terms of a resonance betweenthe fractal character of basic perceptual pro-cesses and the characteristics of the perceivedpattern, or as Goldberger (1996) put it, “Theartwork externalizes and maps the internalbrain-work. . .Conversely, the interaction of theviewer with the artform may be taken as an actof self-recognition” (p. 102). Yet it is difficult tosee how a resonance between the perceiver andthe perceived can explain the aesthetic experi-ence that is often associated with these images.What can an objective description in terms ofnoise signals tell us about subjective aestheticexperiences? Another critical point is that theexplanation of biophilic responses toward frac-tals in terms of noise signals emitted by thenervous system must be situated on a differentlevel than the evolutionary account presentedhere. The former explanation refers to the work-ings of the nervous system, but it does notexplain why it is characterized by 1/f noise inthe first place. Although it is true that someresearchers believe it is the result of evolution ina natural world with similar fractal properties(Anderson & Mandell, 1996), it remains unclearhow a connection with our evolutionary frame-work should be conceived precisely. In fact, 1/fnoise occurs on all levels of human functioning,from human gait dynamics to neural function-ing, whereas in our account it is proposed thatthe affiliation with natural form has its correlatein more discrete brain mechanisms.

Fractals and biophilic reactions: a criticalevaluation. The previous review shows thatthere is convincing evidence that fractals cap-ture some of the core geometric qualities ofnatural structures. Furthermore, there is alsosome support, both intuitively and empirically,that biophilic responses are associated withsome typical fractal qualities (i.e., their degreeof recursiveness). Still, this does not necessarilyentail that it is the naturalness associated with

fractals that is the underlying cause of theseresponses. Notwithstanding that this is perhapsthe most obvious explanation, it could beequally hypothesized that fractals are quitecomplex patterns that give us the necessarydegree of arousal that our visual apparatus de-sires. Such an explanation does not rule out thefact that fractals look natural, but it makes norecourse to naturalness to explain our emotionalrelation with such patterns.

It is clear that further research is needed onthe topic of fractal aesthetics and the possiblerelations to biophilia. In this regard, it is inter-esting to note that some empirical research be-gins to unravel these issues. For example,Hagerhall et al. (2004) have performed a moresystematic inquiry into the relation betweenaesthetic preference and fractals. In the firststage of the study, 119 participants indicatedwhich silhouette outlines of 80 nature scenesthey preferred most. In a next stage, the fractaldimension of all these silhouettes was calcu-lated. It is worthwhile to note that the photos ofsettings with water features and hills were leftout of the total pool of pictures. The reason isthat these elements have a strong influence onthe visual inspection and aesthetic judgment oflandscapes (e.g., Ulrich, 1981), which coulddistract participants from concentrating on thesilhouette outlines. For the remaining 52 pic-tures, analyses showed a modest but significantcorrelation between mean preference and thefractal dimension of the silhouettes. Althoughfurther research is needed, this finding lendssupport to the claim that fractal characteristicsplay a significant role in our biophilic responsestoward vegetated/natural landscapes.

Fractal architecture. The previous discus-sion tentatively suggests that the beneficial ef-fects associated with certain natural objects(positive affect, liking reactions, stress-reduction) could be tapped with fractal charac-teristics but without the presence of actual rep-resentations or imitations of nature. Indeed,fractal structures seem to capture some essentialfeatures of naturalness, such as the recurrenceof (similar) detail on subsequent scales of mag-nitude. However, the architecture of many mod-ern environments is becoming predominantlyeuclidean. This trend is orthogonal to our (hy-pothesized) predilection for fractal structure,which could have subtle but definite psycholog-ical and physiological costs. This problematic

320 JOYE

situation can be countered (to a certain extent)by architectural work that implements some es-sential fractal characteristics. Our preference fora specific fractal dimension further indicatesthat the aesthetic effects of such buildings canbe maximized for intermediate levels of com-plexity. Perhaps this reflects our preference forintermediately complex environments, such assavannas.

If true, the previous findings underscore thevalue of integrating fractal characteristics inarchitectural design. In essence, fractal architec-ture can be realized by repeating similar detailson multiple hierarchical scales of an architec-tural design. Despite the growing attention forfractal aesthetics, which can be witnessed today,we am not aware of actual instances of fractalarchitecture explicitly rooted in the current the-oretical framework. This, however, does notpreclude that, historically, instances of fractalarchitecture have been constructed, albeit for othertheoretical or ideological reasons. According toMichael Ostwald (2001), the first examples ofarchitecture, referring to fractal theory, began toappear shortly after Benoit Mandelbrot’s seminalbook Fractals: Form, Chance and Dimension.Although there has been a rise in fractal architec-ture from 1978 to 1988, this trend did not persistduring the 1990s. Yet, in recent years there is arenewed interest in integrating the complexity sci-ences and fractals in architecture. To a certainextent, this is due to the publication of CarlBovill’s Fractals in Architecture and Design(1996), in which fractal geometry is advancedas a useful design instrument. With his booksThe New Paradigm in Architecture (2002) andThe Architecture of the Jumping Universe(1995), architectural theorist Charles Jencks hasundoubtedly an important share in bringing theissue of fractal architecture to the attention ofthe architectural community. Nevertheless,some have argued that Jencks gives a skewedaccount of fractal architecture because he essen-tially misrepresents what a fractal actually is(Joye, 2006; Salingaros, 2004). This has ledsome to conclude that his use of the notionfractal in architecture mainly serves rhetoricalpurposes (Joye, 2007).

Fractals have been implemented in architec-ture in many ways, albeit consciously or intu-itively (for a review, see Joye, 2007). Probablythe most striking examples of fractal architec-ture have been intuitively constructed, before

the systematic description of fractal geometry.For example, Leonard Eaton (1998) argues howFrank Lloyd Wright (unconsciously) emulated afractal-like geometry in his Palmer House. Nev-ertheless, it should be noted that the fractality ofthis building is predominantly situated inground plan, and this approach has been criti-cized by Joye (2007). Perhaps more convincingexamples of three-dimensional fractal architec-ture can be found in Gothic architecture (Figure5). In this regard, Goldberger (1996, p. 101)notes that “fractals capture several key featuresof Gothic architecture: its porous ‘holeyness’ orcarved-out appearance, its wrinkled crenelatedsurfaces, and its overall self-similarity. . .Froma distance, the sharp spires are the dominantfeature. Closer proximity reveals that thesespires are not smooth, but have spiny out-growths. Yet closer inspection reveals evenmore pointed detail superimposed on these or-naments. The repetition of different shapes(arches, windows, spires) on different scalesyields a combination of complexity and order.”

In agreement with the Gothic, the shape ofcertain Hindu temples also has conspicuousfractal characteristics (Figure 6). The makeup ofthese buildings cannot be viewed separatelyfrom the Hindu worldview. In particular, Hinducosmology is in a sense holographic in that allparts of the cosmological whole are the wholeitself and contain all the information about thewhole. Some schools of Hindu thought advancethe (related) view that the macrocosm is encap-sulated in the microcosm: “The entire cosmoscan be visualized to be contained in a micro-cosmic capsule, with the help of the concept ofsubtle elements called ‘tanmatras.’ The wholecosmic principle replicates itself again andagain in ever smaller scales. The human being issaid to contain within itself the entire cosmos”(Trivedi, 1989, pp. 245–246). Both these cos-mological conceptions can be straightforwardlyrelated to fractal self-similarity, where theglobal formal structure recurs, on ever finersubscales, in the microstructure.

A potential problem for the argument pre-sented here is that not much architecture dis-plays the very profound fractality of Hindu orGothic architecture, which could cast doubts onthe hypothesized universal preference for frac-tal-like patterns. However, some claim that, his-torically, architecture has always had an impor-tant fractal component (Salingaros, 2006), in


that many buildings display architectural detailon different scales of magnitude, without hav-ing a more exact sense of self-similarity. In-deed, a consideration of some of the high pointsof Western architecture, ranging from, for ex-ample, the Gothic to Baroque and Organic Ar-chitecture, quickly reveals a remarkable ten-dency for a cascade of detail, going from thelargest building structure to the smallest deco-rative element. This view is also empirically

supported for divergent building styles (e.g.,Bovill, 1996; Burkle-Elizondo & Valdez-Cepeda, 2006; Capo, 2004; Crompton, 2002).

Still, when observing modern urban settings,it is also a matter of fact that not all architectureof all times contains references to natural form,which suggests that tastes are not exclusivelydictated by biological predispositions. How canthis be reconciled with the hypothesis that weare in a sense genetically predisposed to affiliate

Figure 5. Gothic architecture often has an important fractal component because it containsa wealth of architectural elements on increasing smaller scales of magnitude. (Copyright ©Yannick Joye.)

322 JOYE

with natural forms and to express this affiliationartistically? An answer is that inborn mecha-nisms do not imply genetic determinism butinstead are in a sense open in that they caninteract with cultural or experiential parameters.Although in this study we mainly focusedon immediate affective responses to nature,Ulrich’s (1983) psychoevolutionary model alsoleaves room for more cultural modes of aes-thetic appreciation, which interact (in a complexway) with immediate affective responses. Con-sistent with this is that discussions on innatenessare sometimes framed in terms of biologicallyprepared learning rules, where experiential,and hence cultural, influences are attributed asignificant role (Cummins & Cummins, 1999;Ulrich, 1993).

Although it is plausible that changes can oc-cur in artistic taste, 1 this does not in any senseinvalidate the argument that the outlook of cer-tain habitats is more attuned to a set of specifichardwired affiliations, which were discussedhere. Again, it is unproblematic that people holdspecific tastes by which they also become sur-rounded by nonnatural architectural forms.Problems could arise when the artistic expres-sions of these (more culturally colored) tastesgain dominance, with the result that chances forexperiencing natural forms, and the associatedbiophilic responses, are significantly reduced.Of course, prevailing cultural tastes could clashwith basic-level biological tastes, which couldindicate that there is a complex interplay be-tween the two modes of appreciation. Probably,architects are sensitive to what is culturallyfashionable and are thus in a good position for

coherently fusing biological and cultural tastesinto a successful architectural expression.

Discussion

In this article, different research disciplineswere drawn together, ranging from environmen-tal psychology to architecture. This speculativeand interdisciplinary study can, therefore, beunderstood as a contribution to the integrationof (subfields of) the scientific and culturalworlds. The argument began with a survey offindings from the field of environmental psy-chology, which revealed that humans are affec-tively related to specific natural elements andsettings, being the result of human evolution ina natural environment. Today, however, modernhabitats contain increasingly less actual natureor artistic references to natural form or to thestructural organization of preferred natural set-tings. Although such a trend undoubtedly has

1 However, a question that immediately springs to mindis, who holds these tastes? Is it the general public or is it theartists themselves? If the artists, then it nevertheless remainsan open question whether these changes in taste are alsoappreciated by the broad public, which is the audience forwhom biophilic architecture is intended. In this regard, itcan be noted that there is often a discrepancy between theaesthetic tastes of laypersons and those of experts (e.g.,artists). For example, in a playful experiment, Russian art-ists Vitaly Komar and Alex Melamid found that peoplefrom different cultures preferred kitschy landscape paint-ings over nonrepresentational abstract art (see: www.di-acenter.org/km/). Although it is premature to deduce defi-nite conclusions from these observations, it neverthelessshows a more complex view than the adagio that tastes areessentially variable.

Figure 6. Three-dimensional generation of the central spire of a Hindu temple. Onearchitectural element is repeated on subsequent scales, which results in a complex aestheticobject. (Copyright © Yannick Joye.)


artistic and epistemological consequences, themore profound implication is that it can nega-tively influence psychological and physiologi-cal functioning. According to the argument ofthe current study, biophilic architecture can helpin overcoming the discrepancy between ances-tral and current habitats. More specifically, thearchitectural imitation of natural elements andhabitats that promoted fitness (e.g., vegetativestructures) can lead to the autonomous andquick onset of positive affective reactions,which can lead to positively toned feelings andstress reduction. It was argued how such imita-tions can be realized according to different lev-els of abstraction, ranging from literal imita-tions to the application of more abstract geo-metric features of natural objects (e.g., fractalgeometry) and structural features of ancestralhabitats. Applying fractal geometry to architec-ture could be a particularly successful creativestrategy, because it is not directly restricted bystylistic conventions and thus does not excludethe expression of cultural or local tastes.

Nature-based architecture implies that thebuilding enters into a dialogue with a specificset of human inborn affiliations. However, ad-herents of biophilic architecture should becomeaware that their work also has to relate to orbecome embedded in a social, historical, eco-logical, and individual context (Kellert, 2005).Indeed, the notions “should” and “has” are attheir place here. It would be contradictory that,in a social–psychological project like the onepresented here, attention is paid to a basic levelof well-being while other factors that also con-tribute to it are totally neglected. The result isthat an exclusive focus on biophilic interven-tions is not an automatic guarantee for a higherlevel of well-being. For example, it would beunthinkable for biophilic architecture to be in-terested in the short-term or immediate impactof architecture on well-being, while it remainsapathetic for ecological issues, which are rele-vant for the well-being of our future selves andfuture generations. Some could claim that thisline of thought weakens our argument for bio-philic architecture. We believe the contrary andare convinced that a careful consideration ofthese factors can maximize the biophilic re-sponses to architecture, because other interfer-ing factors are controlled for.

A growing number of academics are involvedin nature-based or biophilic design. In addition

to providing a detailed account of this researchtopic, this study hopes to awaken further inter-est in biophilic design. This is necessary be-cause the arguments presented here remainquite tentative. For example, on a theoreticallevel, it would be insightful to come to a finergrained account of how fractal forms are pro-cessed by the brain. Another more practicalissue is the question of how to create successfulbiophilic architecture. It could be true that manyarchitects have an intuitive feel for the impor-tance of nature as a source of inspiration. How-ever, looking at the modern built environment,it is also a fact that this intuition is not often putinto practice. This report, therefore, tried toprovide some practical guidelines. It should benoted, however, that it only has scratched thesurface. In a sense, only a few grammaticalrules were presented, and it is up to creativeminds to work out a formal language with theseelemental rules. Such a project can only succeedby a transdisciplinary approach, in which botharchitects and psychologists take knowledge ofthis new field of research.

References

Abraham, F. D., Sprott, J. C., Mitina, O., Osorio, M.,Dequito, E. A., & Pinili, J. M. (2003). Judgments oftime, aesthetics, and complexity as a function of thefractal dimension of images formed by chaotic at-tractors. Retrieved August 16, 2007, from http://www.blueberry-brain.org/silliman/JEM%20ms2.htm

Aiken, N. E. (1998a). Human cardiovascular re-sponse to the eye spot threat stimulus. Evolutionand Cognition, 4, 51–62.

Aiken, N. E. (1998b). The biological origins of art.Westport, CT: Praeger.

Aks, D. J., & Sprott, J. C. (1996). Quantifying aes-thetic preference for chaotic patterns. EmpiricalStudies of the Arts, 14, 1–16.

Anderson, C. M., & Mandell, A. J. (1996). Fractaltime and the foundations of consciousness: Verti-cal convergence of 1/f phenomena from ion chan-nels to behavioral states. In E. MacCormac & M. I.Stamenov (Eds.), Fractals of brain, fractals ofmind (pp. 75–128). Amsterdam, the Netherlands:John Benjamins.

Appleton, J. (1975). The experience of landscape.New York: Wiley.

Atran, S. (1995). Causal constraints on categories andcategorical constraints on biological reasoningacross cultures. In D. Sperber, D. Premack, & J.Premack (Eds.), Causal cognition. A multidisci-

324 JOYE

plinary debate (pp. 205–233). Oxford, England:Clarendon Press.

Balling, J. D., & Falk, J. H. (1982). Development ofvisual preference for natural environments. Envi-ronment and Behavior, 14, 5–28.

Bovill, C. (1996). Fractal geometry in architectureand design. Boston: Birkhauser.

Burkle-Elizondo, G., & Valdez-Cepeda, R. D.(2006). Fractal analysis of Mesoamerican pyra-mids. Nonlinear Dynamics, Psychology, and LifeSciences, 10, 105–122.

Capo, D. (2004). The fractal nature of the architec-tural orders. Nexus Network Journal, 6, 30–40.

Caramazza, A., & Shelton, J. R. (1998). Domain-specific knowledge systems in the brain: The ani-mate-inanimate distinction. Journal of CognitiveNeuroscience, 10, 1–34.

Coss, R. G. (2003). The role of evolved perceptualbiases in art and design. In E. Voland & K. Gram-mer (Eds.), Evolutionary aesthetics. Berlin:Springer-Verlag.

Crompton, A. (2002). Fractals and the picturesque.Environment and Planning B: Planning and De-sign, 29, 451–459.

Crutch, S. J., & Warrington, E. K. (2003). The se-lective impairment of fruit and vegetable knowl-edge: A multiple processing channels account offine-grain category specificity. Cognitive Neuro-psychology, 20, 355–372.

Cummins, D. D., & Cummins, R. (1999). Biologicalpreparedness and evolutionary explanation. Cog-nition, 73, 37–53.

Custers, M. (2006). Stressreductie in de volkstuin?!Experimenteel onderzoek naar het stressreducer-ende effect van tuinieren [Stress reduction in thegarden?! Experimental inquiry into the stress re-ducing effects of gardening). Leiden, the Nether-lands: University of Leiden.

Cutting, J. E., & Garvin, J. J. (1987). Fractal curvesand complexity. Perception and Psychophys-ics, 42, 365–370.

Eaton, L. (1998). Fractal geometry in the late work ofFrank Lloyd Wright. In K. Williams (Ed.), NexusII: Architecture and mathematics 1998 (pp. 23–38). Florence, Italy: Edizioni Dell’Erba.

Eibl-Eibesfeldt, I. (1989). Human ethology. NewYork: Aldine de Gruyter.

Evans, G. W., & McCoy, J. M. (1998). When build-ings don’t work: The role of architecture in humanhealth. Journal of Environmental Psychology, 18,85–94.

Farah, M. J., & McClelland, J. (1991). A computa-tional model of semantic memory impairment:Modality specificity and emergent category speci-ficity. Journal of Experimental Psychology, 120,339–357.

Feuerstein, G. (2002). Biomorphic architecture: Hu-man and animal forms in architecture. Stuttgart,Germany: Axel Menges.

Geake, J. G. (1992). Fractal computer graphics as astimulus for the enhancement of perceptual sensi-tivity to the natural environment. Australian Jour-nal of Environmental Education, 8, 1–16.

Geake, J. G., & Landini, G. (1997). Individual dif-ferences in the perception of fractal curves. Frac-tals, 5, 129–143.

Gilden, D. L., Schmuckler, M. A., & Clayton, K.(1993). The perception of natural contour. Psycho-logical Review, 100, 460–478.

Goldberger, A. L. (1996). Fractals and the birth ofGothic: Reflections on the biologic basis of cre-ativity. Molecular Psychiatry, 1, 99–104.

Hagerhall, C. M., Purcell, T., & Taylor, R. (2004).Fractal dimension of landscape silhouette outlinesas a predictor of landscape preference. Journal ofEnvironmental Psychology, 24, 247–255.

Hartig, T., Evans, G. W., Jamner, L. D., Davis, D. S.,& Garling, T. (2003). Tracking restoration in nat-ural and urban field settings. Journal of Environ-mental Psychology, 23, 109–123.

Hartig, T., Mang, M., & Evans, G. W. (1991). Re-storative effects of natural environment experi-ences. Environment and Behavior, 23, 3–26.

Hase, B., & Heerwagen, J. H. (2000). Phylogeneticdesign: A new approach for workplace environ-ments. Journal for Quality and Participation, 23,27–31.

Haviland-Jones, J., Rosario, H. H., Wilson, P., &McGuire, T. R. (2005). An environmental ap-proach to positive emotion: Flowers. EvolutionaryPsychology, 3, 104–132.

Heerwagen, J. H. (2003). Bio-inspired design: Whatcan we learn from nature? Unpublished manu-script.

Heerwagen, J. H., & Orians, G. H. (1986). Adapta-tions to windowlessness: A study of the use ofvisual decor in windowed and windowless offices.Environment and Behavior, 18, 623–639.

Heerwagen, J. H., & Orians, G. H. (1993). Humans,habitats, and aesthetics. In S. R. Kellert & E. O.Wilson (Eds.), The biophilia hypothesis (pp. 138–172). Washington, DC: Island Press.

Herzog, T. R., Kaplan, S., & Kaplan, R. (1982). Theprediction of preference for unfamiliar urbanplaces. Population and Environment, 5, 43–59.

Hietanen, J. K., & Korpela, K. M. (2004). Do bothnegative and positive environmental scenes elicitrapid affective processing? Environment and Be-havior, 36, 558–577.

Hildebrand, G. (1999). Origins of architectural plea-sure. Berkeley, CA: University of California Press.

Jencks, C. (1995). The architecture of the jumpinguniverse. London: Academy/Wiley.


Jencks, C. (2002). The new paradigm in architecture.New Haven, CT: Yale University Press.

Johnson, S. C., & Carey, S. (1998). Knowledge en-richment and conceptual change in folkbiology:Evidence from Williams syndrome. Cognitive Psy-chology, 37, 156–200.

Joye, Y. (2006). An interdisciplinary argument fornatural morphologies in architectural design. En-vironment and Planning B: Planning and De-sign, 33, 239–252.

Joye, Y. (2007). A tentative argument for the inclu-sion of nature-based forms in architecture. Unpub-lished doctoral dissertation, Ghent University, Gh-ent, Belgium.

Kahn, P. H., Jr. (1999). The human relationship withnature: Development and culture. Cambridge,MA: MIT Press.

Kals, E., Schumacher, D., & Montada, L. (1999).Emotional affinity toward nature as a motivationalbasis to protect nature. Environment and Behav-ior, 31, 178–202.

Kaplan, R., & Kaplan, S. (1989). The experience ofnature: A psychological perspective. Cambridge,England: Cambridge University Press.

Kaplan, S. (1987). Aesthetics, affect and cognition.Environment and Behavior, 19, 3–32.

Kaplan, S. (1988). Perception and landscape: Con-ceptions and misconceptions. In J. Nasar (Ed.),Environmental aesthetics: Theory, research, andapplications (pp. 45–55). Cambridge, England:Cambridge University Press.

Katcher, A., & Wilkins, G. (1993). Dialogue withanimals: Its nature and culture. In S. R. Kellert &E. O. Wilson (Eds.), The biophilia hypothesis (pp.173–197). Washington, DC: Island Press.

Kellert, S. (1997). Kinship to mastery: Biophilia inhuman evolution and development. Washington,DC: Island Press.

Kellert, S. (2005). Building for life: Understandingand designing the human-nature connection.Washington, DC: Island Press.

Kellert, S., & Wilson, E. O. (Eds.). (1993). Thebiophilia hypothesis. Washington, DC: IslandPress.

Knill, D. C., Field, D., & Kersten, D. (1990). Humandiscrimination of fractal images. Journal of theOptical Society of America, 7, 1113–1123.

Korpela, K. M., Klemettila, T., & Hietanen, J. K.(2002). Evidence for rapid affective evaluation ofenvironmental scenes. Environment and Behav-ior, 34, 634–650.

Lynn, G. (1998). Folds, bodies and blobs. Brussels,Belgium: La Lettre Volee.

Lynn, G. (1999). Animate form. New York: PrincetonArchitectural Press.

Mandelbrot, B. (1977). The fractal geometry of na-ture. New York: Freeman.

Mead, C. (1991). Houses by Bart Prince. An Amer-ican architecture for the continuous present. Al-buquerque: University of New Mexico Press.

Mineka, S., & Ohman, A. (2002). Phobias and pre-paredness: The selective, automatic, and encapsu-lated nature of fear. Biological Psychiatry, 52,927–937.

Mithen, S. (1996). The prehistory of the mind. Lon-don: Thames & Hudson.

Mureika, J. (2005). Fractal theory of aesthetic pref-erence in abstract expressionism: A connection tothe eight laws of artistic experience? Manuscriptsubmitted for publication.

Orians, G. H. (1980). Habitat selection: General the-ory and applications to human behaviour. In J. S.Lockard (Ed.), The evolution of human social be-havior (pp. 49–77). New York: Elsevier.

Orians, G. H., & Heerwagen, J. H. (1992). Evolvedresponses to landscapes. In J. H. Barkow, L. Cos-mides, & J. Tooby (Eds.), The adapted mind. Evolu-tionary psychology and the generation of culture (pp.555–579). New York: Oxford University Press.

Ostwald, M. J. (2001). “Fractal architecture”: Latetwentieth century connections between architec-ture and fractal geometry. Nexus Network Journal,Architecture and Mathematics, 3, 73–84.

Parsons, R. (1991). The potential influences of envi-ronmental perception on human health. Journal ofEnvironmental Psychology, 11, 1–23.

Parsons, R., Tassinary, L. G., Ulrich, R. S., Hebl,M. R., & Grossman-Alexander, M. (1998). Theview from the road: Implications for stress recov-ery and immunization. Journal of EnvironmentalPsychology, 18, 113–139.

Peitgen, H. O., Jurgens, H., & Saupe, D. (1992).Chaos and fractals. New frontiers of science. NewYork: Springer-Verlag.

Pinker, S. (1994). The language instinct. The newscience of language and mind. London: PenguinPress.

Pinker, S. (1997). How the mind works. London:Norton.

Pinker, S. (2002). The blank slate: The modern denialof human nature. London: Penguin Press.

Potts, R. B. (1998). Environmental hypotheses ofhominin evolution. Yearbook of Physical Anthro-pology, 41, 93–136.

Prusinkiewicz, P., & Lindenmayer, A. (1990). Thealgorithmic beauty of plants. New York: Springer-Verlag.

Purcell, T., Peron, E., & Berto, R. (2001). Why dopreferences differ between scene types? Environ-ment and Behavior, 33, 93–106.

Ramachandran, V. S., & Hirstein, W. (1999). Thescience of art: A neurological theory of aestheticexperience. Journal of Consciousness Studies, 6,15–51.

326 JOYE

Rogowitz, B. E., & Voss, R. F. (1990). Shape per-ception and low-dimension fractal boundary con-tours. In B. E. Rogowitz & J. Allenbach (Eds.),Proceedings of the Conference on Human Vision:Methods, models and applications (Vol. 1249, pp.387–394), Santa Clara, CA: Symposium on Elec-tronic Imaging.

Salingaros, N. A. (2003). The sensory value of orna-ment. Communication & Cognition, 36, 331–351.

Salingaros, N. A. (2004). Anti-architecture and de-construction. Solingen, Germany: Umbau-Verlag.

Salingaros, N. A. (2006). A theory of architecture.Solingen: Umbau-Verlag.

Smardon, R. C. (1988). Perception and aesthetics of theurban environment: Review of the role of vegetation.Landscape and Urban Planning, 15, 85–106.

Sommer, R., & Summit, J. (1995). An exploratorystudy of preferred tree form. Environment andBehavior, 27, 540–557.

Spehar, B., Clifford, C. W. G., Newell, B., & Taylor,R. P. (2003). Universal aesthetic of fractals. Com-puters & Graphics, 27, 813–820.

Sperber, D., & Hirschfeld, L. A. (2004). The cogni-tive foundations of cultural stability and diversity.Trends in Cognitive Science, 8, 40–46.

Stamps, A. E. (2002). Fractals, skylines, nature andbeauty. Landscape and Urban Planning, 60, 163–184.

Steiner, R. (1999). Architecture as a synthesis of thearts. London: Rudolf Steiner Press.

Summit, J., & Sommer, R. (1999). Further studies ofpreferred tree shapes. Environment and Behav-ior, 31, 550–576.

Sweeney, J. J., & Sert, J. L. (1960). Antoni Gaudı.London: The Architectural Press.

Synek, E., & Grammer, K. (1998). Evolutionary aes-thetics: Visual complexity and the development ofhuman landscape preferences. Manuscript. Re-trieved from http://evolution.anthro.univie.ac.at/institutes/urbanethology/projects/urbanisation/landscapes/indexland.html

Taylor, R. P. (1998). Splashdown. New Scientist,2144, 30–31.

Taylor, R. P. (2002). The construction of fractal drippaintings. Leonardo, 35, 203–207.

Taylor, R. P., Spehar, B., Wise, J. A., Clifford,C. W. G., Newell, B. R., & Martin, T. P. (2005).Perceptual and physiological responses to the vi-sual complexity of Pollock’s fractal dripped pat-terns. Journal of Non-linear Dynamics, Psychol-ogy and Life Sciences, 9, 89–114.

Thayer, R. L., Jr., & Atwood, B. G. (1978). Plants,complexity, and pleasure in urban and suburbanenvironments. Environmental Psychology andNonverbal Behavior, 3, 67–76.

Tinbergen, N., & Perdeck, A. C. (1950). On thestimulus situation releasing the begging response

in the newly hatched herring gull chick (Larusargentatus argentatus Pont). Behaviour, 3, 1–39.

Tischhauser, A., & Von Moos, S. (1998). Cala-trava—Public buildings. Basel, Switzerland:Birkhauser.

Todorova, A., Asakawa, S., & Aikoh, T. (2004).Preferences for and attitudes towards street flowersand trees in Sapporo, Japan. Landscape and UrbanPlanning, 69, 403–416.

Trivedi, K. (1989). Hindu temples: Models of a frac-tal universe. The Visual Computer, 5, 243–258.

Tyler, L. K., & Moss, H. E. (2001). Towards adistributed account of conceptual knowledge.Trends in Cognitive Sciences, 5, 244–252.

Tzonis, A. (1999). Santiago Calatrava. The poeticsof movement. London: Thames & Hudson.

Ulrich, R. S. (1981). Natural versus urban scenes—Some psychophysiological effects. Environmentand Behavior, 13, 523–556.

Ulrich, R. S. (1983). Aesthetic and affective responseto natural environment. In I. Altman & J. F. Wohl-will (Eds.), Human behavior and the environment:Volume 6 (pp. 85–125). New York: Plenum Press.

Ulrich, R. S. (1984). View through a window mayinfluence recovery from surgery. Science, 224,420–421.

Ulrich, R. S. (1986). Effects of hospital environments onpatient well-being (Research rep. 9[55]). Trondheim,Norway: Department of Psychiatry and BehaviouralMedicine, University of Trondheim.

Ulrich, R. S. (1993). Biophilia, biophobia, and natu-ral landscapes. In S. R. Kellert & E. O. Wilson(Eds.), The biophilia hypothesis (pp. 73–137).Washington, DC: Island Press.

Ulrich, R. S., Simons, R. F., Losito, B. D., Fiorito, E.,Miles, M. A., & Zelson, M. (1991). Stress recoveryduring exposure to natural and urban environ-ments. Journal of Environmental Psychology, 11,201–230.

Ulrich, R. S., & Zimring, C. (2004). The role of thephysical environment in the hospital of the 21st cen-tury: A once-in-a-lifetime opportunity. Retrieved Au-gust 18, 2007, from http://www.healthdesign.org/research/reports/pdfs/role_physical_env.pdf

Van den Berg, A. E. (2004, April). De charme van desavanne: Onderzoek naar landschapsvoorkeuren[The charm of the savanna: Inquiry into landscapepreferences]. Topos, 10–12.

Van den Berg, A. E., Hartig, T., & Staats, H. (2007).Preference for nature in urbanized societies: Stress,restoration, and the pursuit of sustainability. Jour-nal of Social Issues, 63, 79–96.

Van den Berg, A. E., Koole, S. L., & van der Wulp,N. Y. (2003). Environmental preference and res-toration: (How) are they related? Journal of Envi-ronmental Psychology, 23, 135–146.

Van den Berg, A. E., & Ter Heijne, M. (2005). Fearversus fascination: Emotional responses to natural


threats. Journal of Environmental Psychology, 25,261–272.

Van den Berg, A. E., & Van Winsum-Westra, M.(2006). Ontwerpen met groen voor gezondheid:Richtlijnen voor de toepassing van groen in ‘healingenvironments’ [Designing with green for health:Guidelines for the application of green in “healingenvironments”]. (Report 1371, reeks belevingsonder-zoek nr. 15). Wageningen, Germany: Alterra.

Voss, R. F. (1988). Fractals in nature: From charac-terization to simulation. In H. O. Peitgen & D.Saupe (Eds.), The science of fractal images (pp.21–70). New York: Springer-Verlag.

Wilson, E. O. (1984). Biophilia. Cambridge, MA:Harvard University Press.

Wilson, E. O. (1993). Biophilia and the conservationethic. In S. R. Kellert & E. O. Wilson (Eds.), Thebiophilia hypothesis (pp. 31– 41). Washington,DC: Island Press.

Wise, J. A., & Rosenberg, E. (1986). The effects ofinterior treatments on performance stress in threetypes of mental tasks (Tech. rep., Space Human Fac-tors Office). Sunnyvale, CA: NASA-ARC.

Wise, J. A., & Taylor, R. (2002). Fractal designstrategies for enhancement of knowledge work en-vironments. In Proceedings of the 46th Meeting ofthe Human Factors and Ergonomics Society (pp.854–859).Baltimore, MD: Human Factors and Er-gonomics Society.

Wolff, P., Medin, D. L., & Pankratz, C. (1999).Evolution and devolution of folkbiological knowl-edge. Cognition, 73, 177–204.

Yu, Y., Romero, R., & Lee, T. W. (2005). Preferenceof sensory neural coding for 1/f signals. PhysicalReview Letters, 94, 108103(1)–108103(4).

Received September 19, 2006Accepted July 19, 2007 �

Acknowledgment of Ad Hoc ReviewersThe Editor and Editorial Review Board take this opportunity to thank the followingpersons who reviewed submissions for volume 11, 2007.

Owen AndersonJohn ArcherPaul BainAustin BaldwinWilliam BeckerMathew BelmonteCamilla BenbowGeoff BirdSimon BoagGeorge BonannoBecky BurchNatalie CiaroccoLee CronkPaul DavidsonValerian DerlegaYvonne DelevoyeKathryn DindiaCarol DolanMichael DougherValentina D’UrsoCarol DweckAleksander Ellis

Bruce EllisRichard FlemingMelanie GreenGary HardcastleChristine HarrisWilliam HirstGeorge HoldenHarry HuntYuhong JiangPeter JudgePatrik JuslinKalman KaplanJohn KihlstromAlan KazdinDebra LiebermanScott LilienfeldDarrin McMahonJeanne MarecekDavid MilneStephen MorilloWarren MorrillShane Lopez

Robert Lynd-StevensonDavid MiallSusan MinekaJeanne MarecekIrene PepperbergSteven PlatekJ. T. PtacekDevaki RauJustin RentfrowMatt RossanoGad SaadCatherine SalmonJohn SlobodaAaron SlomanDavid Livingston SmithSusan SprecherGraham StainesGary SteinerDouglas SturmDan WegnerMichael WertheimerSarah White

328 JOYE

architectural lessons from environmental psychology… · architectural lessons from environmental...

Documents