Affordances can invite behavior: Reconsidering the relationship between affordances and agency

(This is a sample cover image for this issue. The actual cover is not yet available at this time.)

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

http://www.elsevier.com/copyright

Author's personal copy

Affordances can invite behavior: Reconsidering the relationship betweenaffordances and agency

Rob Withagen a,*, Harjo J. de Poel a, Duarte Araújo b, Gert-Jan Pepping a

aCenter for Human Movement Sciences, University Medical Center Groningen, University of Groningen, P.O. Box 196, 9700 AD Groningen, The Netherlandsb Faculty of Human Kinetics, Technical University of Lisbon, Portugal

Keywords:AffordancesAgencyArchitectureEcological psychologyIndustrial designPhenomenology

a b s t r a c t

The concept of agency has been central to ecological approaches to psychology. Gibson,one of the founders of this movement, made room for this concept by arguing against themechanistic conceptions in psychology. In his view, the environment is not a collection ofcauses that pushes the animal around, but consist of action possibilities, which he coinedaffordances. In making their way in the world, animals regulate their behavior with respectto these possibilities. Reed later developed this ecological conception of agency, followingGibson in conceiving of affordances as action possibilities. However, drawing uponindustrial design, architecture, and phenomenology, we argue that affordances are notmere action possibilities but that they can also invite behavior. We suggest a mutualistperspective on invitations, suggesting that they depend on the animal–environmentrelationship in multiple dimensions. The implications of this new conception of affor-dances for the ecological account of agency are explored.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Gibson’s (1966,1979/1986) ecological psychology can beunderstood as a critique on mechanistic metaphors thatdominate thinking in psychology since the 17th century(Reed, 1988a, 1996a; Withagen & Michaels, 2005). Eversince the mechanization of the worldview (Dijksterhuis,1950), philosophers and later psychologists have gener-ally tried to understand mind and behavior in mechanisticterms. In the beginning of the 20th century, Holt (1914/1973), Gibson’s graduate school mentor, claimed that“psychology is at the present moment addicted to the beadtheory” (p. 160)dit tried to understand behavior in termsof a chain of causes and effects.

Following his mentor, Gibson (1966, 1979/1986) tookaim at the mechanistic conceptions of perception andaction. He argued that animals should not be conceived of

as machines, the responses of which are caused by stimulifrom the environment. In addition, he claimed that themechanistic conception of the environment as matter inmotion is inappropriate to understand animal behavior. InGibson’s view, the animal’s environment consists of actionpossibilities, which he termed affordances. For instance, inthe human environment, a floor affords walking upon,a cup affords grasping, water affords drinking, and so on.With this new conception of the environment, Gibson putagency back on the agenda and thereby overturned themechanistic framework that underlies many psychologicalapproaches. After all, if the environment consists ofopportunities for action that do not cause behavior butsimply make it possible, animals appear as being autono-mous, making their way in the world. They are not merepuppets pushed by the environment like machines; rather,animals have agency.

The present paper explores this relationship betweenaffordances and agency. We beginwith a brief exposition ofGibson’s conception of affordances as opportunities foraction. With this new conception of the animal’s

* Corresponding author. Tel.: þ31 503638978; fax: þ31 503633150.E-mail address: [email protected] (R. Withagen).

Contents lists available at SciVerse ScienceDirect

New Ideas in Psychology

journal homepage: www.elsevier .com/locate/newideapsych

0732-118X/$ – see front matter � 2011 Elsevier Ltd. All rights reserved.doi:10.1016/j.newideapsych.2011.12.003

New Ideas in Psychology 30 (2012) 250–258


environment, Gibson placed agency at the center of hisecological approach although he did not develop a fully-fledged theory of it. We then discuss Reed’s ecologicalconception of agency that is rooted in Gibson’s concepts ofinformation and affordances. However, based on differentfields of inquiry, we argue, contrary to Gibson and Reed,that affordances are not mere opportunities for action, butthat they can also invite behaviors. Because Reed’s theory ofagency did not take this character of affordances intoaccount, the relationship between affordances and agencyhas to be reconsidered. We ultimately argue for a mutualistperspective on invitations and explore the implications ofthis perspective for the concept of agency.

2. Gibson’s conception of affordances

Gibson introduced the concept of affordances in themidsixties of the previous century (Gibson,1966; see also Reed,1988a). Although his revolutionary theory of directperception predated the concept of affordances, the latterhas become a central part of his ecological theory ofperception. There are two aspects of the concept of affor-dances that are worth considering in this paper. First of all,the concept of affordances implies that the environment inand of itself is meaningful. Ever since the mechanization ofthe worldview in the 17th century, the environment hasgenerally been conceived of as meaningless, consistingmerely of matter in motion. The founders of this mechani-zation had made a distinction between primary andsecondary qualities. The primary qualities are properties ofthe environment that exist ‘out there’ and independent ofthe observer (e.g., form, mass, speed). Secondary qualities,on the other hand, do not exist in the environment but onlyin the humanmind (e.g., color, taste, smell, meaning). Manytheories of indirect perception are based on this distinction;they assert that meaning originates in the humanmind. It isbestowed on ameaningless world (i.e. primary qualities) byan internal perceptual process (see e.g., Neisser, 1967).Gibson took aim at this mechanistic conception of theenvironment and the allied theory of perception. He arguedthat the animals’ environment is meaningful, it consists ofaffordances. “The meaning or value of a thing consists ofwhat it affords” (Gibson, 1982b, p. 407). In his last book,Gibson developed his ecological approach to understandhow these meaningful action possibilities can be directlyperceived. He argued that in the ambient energy arraysthere are patterns available that specify affordances. Hence,an agent can directly perceive the meaningful environmentby picking-up this information.

Second, and more important for our purposes, byarguing that the animal’s environment consists of oppor-tunities for action, Gibson overturned the idea that theenvironment comprises stimuli that causes the animal’sperception and action. In Gibson’s framework, the envi-ronment is conceived not as a collection of causes, but asamanifold of action possibilities. It is important to note thataffordances have a peculiar ontological status. As Gibsonput it,

An affordance is neither an objective property nora subjective property; or it is both if you like. An

affordance cuts across the dichotomy of subjective–objective and helps us to understand its inadequacy. It isequally a fact of the environment and a fact of behavior.It is both physical and psychical, yet neither. Anaffordance points both ways, to the environment and tothe observer. (Gibson, 1979/1986, p. 129)

In other words, action possibilities exist by virtue ofa relation between the properties of the environment andthe actor. As an example, it is the physical properties of thecup relative to the action capabilities of the animal’s bodythat make it graspable for that animal. This means that thesame object can afford different behavior to differentanimals, and even to the same animal at different momentsin time (e.g., during development).

Although Gibson has been criticized for being vague onthe ontology of affordances (see e.g., Chemero, 2003), hewas clear in suggesting that affordances exist indepen-dently of the animal’s perception and intentions. Thereby,Gibson took aim at the Gestalt psychologists Lewin andKoffka who had earlier suggested that the action-relevantproperties of the environment are immediately perceived.As Koffka (1935) put it, “a fruit says, ‘Eat me’; water says,‘Drink me’; thunder says, ‘Fear me’, and woman says, ‘Loveme’” (p. 7). Both Lewin and Koffka called this the ‘demandcharacter’ of the environment. However, Koffka and Lewinargued that this ‘demand character’ is part of thephenomenological world that changes as the need or theintention of the actor (in their terminology the Ego)changes. “I have a need which for the moment cannot besatisfied; then an object appears in my field which mayserve to relieve that tension, and then this object becomesendowed with a demand character” (Koffka, 1935, p. 354;emphasis added). To use one of Koffka’s (1935) examples,the postbox has a demand character whenwe want to posta letter, but loses this quality once we have done it.

Although Gibson (1979/1986, pp. 138–140) mentionedthat the concept of affordance was inspired by the ideas ofthe Gestalt psychologists, he claimed that affordances donot change as the need of the observer changes. They arenot properties of the phenomenological world that dependupon the state of the observer; rather, they are ecologicalphenomena that exist in the environment. As Gibson(1982b) put it,

The affordances of the environment are permanent,although they do refer to animals and are species-specific. The positive and negative valences of thingsthat change when the internal state of the observerchanges are temporary. The perception of what some-thing affords should not be confused with the ‘coloring’of experience by needs and motives. Tastes and prefer-ences fluctuate. Something that looks good today maylook bad tomorrow but what it actually offers theobserver will be the same. (p. 410; emphases in original)

Hence, according to Gibson, affordances are opportu-nities for action that exist in the environment and do notdepend on the animal’s mind. Moreover, being opportuni-ties for action, they do not cause behavior, but simply makeit possible. This brings us to the place of the concept ofagency in Gibson’s ecological approach.

R. Withagen et al. / New Ideas in Psychology 30 (2012) 250–258 251


2.1. Gibson on agency

Gibson has sometimes been classified as an environ-mental determinist (see e.g., Hilgard, 1987; for an excellentoverview on how Gibson is often misrepresented seeCostall & Morris, submitted for publication). However, thisportrayal is seriously mistaken. Although Gibson spentconsiderable time describing the animal’s environment, heconceived of this environment not as a collection of stimulithat pushes the animal around; instead, he conceived of theenvironment as a manifold of opportunities. “Affordancesdo not cause behavior but constrain or control it” (Gibson,1982b, p. 411). In other words, actions are not caused bythe environment or elicited by stimuli, but are the animals’means to utilize the affordances in their environment.Gibson’s texts are laced with this idea of agency. Forinstance, in his Notes on Action that Reed (1982, 1988b,1996a) later developed into a genuine theory of actionsystems, Gibson (1982a; see also Gibson, 1966, pp. 56–57)took aim at the mechanistic conception of reflexes andargued that the animal regulates its behavior by extractinginformation from the ambient arrays. This idea also figuresprominently inGibson’s (1966) book on the senses, inwhichhe developed an account of how the animal picks up themeaningful patterns in the ambient array. It is important tonote that Gibson’s ideas on the perceptual process are instark contrast with earlier accounts of perceptual processes(see Reed, 1980). These accounts focus on internal psycho-logical processes and conceive a perceiver as a passivemachine that is bombarded by stimuli and emits an output.Gibson portrayed the perceiver instead as an active animal.The detection of invariant patterns from the ambient arraysis an active process inwhichmany bodily parts are involved.Information, Gibson frequently asserted, is not imposed, butobtained.

With the concepts of affordances, information, andperceptual systems, Gibson provided a conceptual frame-work that allows for an ecological account of agency. Earliercritics, however, have argued that the Gibsonian approachneeds to be complemented with a theory of motivation andintention to offer a satisfactory theory of agency. Afterenumerating the multiple affordances that a single piece ofpaper offers a human being, Cutting (1982), for example,argued,

My behavior is virtually unconstrained by its affordan-ces. To be sure, it does not afford flying to Baghdad upon,but the exclusion of a large domain of behaviors doesnot diminish the fact that an infinity remain. To apply toadult human beings, it would seem that the theory ofaffordances needs full-blown theories of personalityand of choice. And I am sure Gibson would have agreed.(p. 216)

Indeed, if the environment consists of affordances, andaffordances are simply opportunities for action, conceptslike motivation and intention seem needed to explain whyanimals utilize certain affordances and not others ata certain moment in time. This is all the more true becausea single object generally offers multiple action possibilitiesto an individual agent (see Gibson, 1979/1986, p. 134),implying that selecting affordances is a ubiquitous and

continuous process. We agree with Cutting (1982) thatGibson did not offer a satisfactory account of this aspect ofagency. However, in the last two decades of the previouscentury, Reed (1982,1993,1996a) developed a sophisticatedecological conception of agency towhichwe shall now turn.

3. Reed’s ecological theory of agency1

Like Gibson, Reed was very critical of the mechanisticconceptions of mind and behavior (see Costall, 1999;Withagen & Michaels, 2005). In his view, psychologiesthat are grounded in the mechanistic framework oftenexplain pivotal psychological phenomena away. Amongthese phenomena is agency. “[P]sychologists have persistedin modeling animal and human behavior on mechanicalprinciples, thus neglecting perhaps the most fundamentalproblem of their fielddautonomous agency” (Reed, 1996a,p. 10). In line with the Gibsons (E.J. Gibson,1994; J.J. Gibson,1966, 1979/1986), Reed argued that agency is one of thehallmarks of animal behavior and thus should be central onthe agenda of psychologists. In his view, ecologicalpsychology offers a promising conceptual framework toelucidate this distinctive feature of the animate world. “Thegoal of ecological psychology is to explain agency scientif-ically, not to explain it away or simply offer a discourseabout it” (Reed, 1996a, p. 19; emphasis in original).

Reed’s (1982, 1985, 1993, 1996a) ecological approachincludes an account of intention and motivation that argu-ably complements the Gibsonian approach in the way thatCutting (1982) envisaged in the above quote. His account isbased on Gibson’s revolutionary concepts of affordancesand information, thereby giving rise to an alternativeconception of motivation and intention. As mentionedabove, earlier psychological approaches have generallyassumed thatmeaning is internal, it originates in the humanmind. Hence, any account ofmotivation that is rooted in thispremise tends to focus on these internal states, assumingthat a positive mental state is what animals are striving at(see Reed, 1996b, p, 119; see also Kahneman, Diener, &Schwarz, 1999). By starting with the assumption that theanimal’s environment consists of values (affordances) andmeaning (information), Reed (1993, 1996a) developed analternative, ecological account. What animals are goingafter are not private pleasures; rather, they are seekingvalues and meanings in the environment. He preferred tocall these activities “the effort after value and meaning”(Reed,1996a, chap. 7) and argued that they are ubiquitous inanimal kingdom.

Reed’s conceptualization of motivation and intention istruly ecological. He not only defined these psychologicalprocesses in terms of affordances and information, he alsoargued extensively that intentions should not be under-stood as mental states that are insulated from the agent’sbody and the environment, and cause an animal to move.

1 Over the last decades, other ecological types have developed anaccount of agency (see e.g., Iberall, 1977; Shaw, 2001). In this paper wedecided to limit ourselves to Reed because he put the concept of agencycentral to his work and developed, in our view, the most elaboratedecological account of it.

R. Withagen et al. / New Ideas in Psychology 30 (2012) 250–258252


From an ecological point of view, intentions are notcauses of action, but patterns of organization of action;they are not mental as opposed to physical, but areinstead embodied in the kinds of performances mostlikely found in cognitively capable creatures. (Reed,1993, p. 62)

Reed argued that these intentional patterns of organi-zation emerge in situations in which animals have a realchoice of behavior, that is in which different affordancescan be utilized. It is important to emphasize that althoughReed contextualized intentions and motivation, suggestingthat they “are spread out across mind, body, information,ecological context, and social setting” (Reed,1993, p. 68), hefollowed Gibson in his conception of affordances as actionpossibilities. Indeed, Reed stressed that animals are incharge of their own activity – “they make their way in theworld” (Reed, 1996a, p. 19; emphases in original) – puttingthe ideas of environmental indeterminism and self-controlcentral to his ecological approach.

In explaining how agents choose between affordances,Reed (1993, 1996a) drew upon the selective retentiontheory (e.g., Ghiselin, 1969) that figures prominently in hisecological approach (see e.g., Costall, 1999; Withagen & vanWermeskerken, 2010). According to Reed, intentions, likeother biological phenomena, emerge out of a process ofvariation and selection. In the case of intentions, thecompetition is between basic units of action that Reed(1993, p. 65) referred to as perception actions cycles(PACs). Each PAC is capable of utilizing a certain affordanceby attuning to information that can guide the realization ofthe action possibility. Reed asserted that in an environmentthat offers multiple affordances, PACs compete and inten-tions emerge out of this competition. He drew a parallelherewith Darwin’s theory of evolution by natural selection.“Intention are thus the ‘species’ that emerge out ofcompetition among perceptual and action processes forutilizing affordances” (Reed, 1993, p. 65). And this emer-gence of intentions result in the goal-directed behavior thatanimals exhibit; the intention will select informationand affordances in complex environments, leading topurposeful behavior.2

4. Can affordances invite behavior?

We have seen that Gibson placed agency central in hisecological approach. By arguing that the animal’s envi-ronment consists of affordances, Gibson overturned themechanistic worldview and thereby made room for theidea that animals make their way in the world. However,Gibson’s approach has been criticized for not offeringa fully-fledged theory of agency. Indeed, we have seen thatCutting (1982) argued that the concept of affordancesneeds to be complemented with a theory of motivation andintention to explain the actual behavior animals perform atdifferent moments in time. Reed developed such an

account of agency, following Gibson’s conception of theenvironment as a manifold of opportunities for action. Inthe next section, however, we will question the idea thataffordances are mere opportunities for action and suggestthat they can also invite actions. Because Reed did notrecognize this character of affordances in his approach, therelationship between affordances and agency has to bereconsidered.

4.1. Lessons from industrial design

The idea that the environment can make certainactions more likely to occur is easily demonstrated byexperiments in industrial design. Especially over thelast ten years, researchers in this field have emphasizedthat objects are not merely functional but always affectthe agent emotionally, making certain behaviors morelikely to occur (see e.g., Desmet & Hekkert, 2007, 2009).For example, Ju and Takayama (2009) examined howpeople visually perceive automatic door movements.Each of the contrived automatic door movements intheir study afforded entering. However, when the doorssimply opened or opened with a pause, these move-ments were experienced as inviting the participants tocome in or even urged them to do so. When, on theother hand, the door opened and then closed, theparticipants perceived the door movements as reluctantto let them enter.

It is interesting to note that the concept of affordances isfrequently used in the literature on industrial design. Themost notable example is probably Norman’s (1988/2002)popular book The design of everyday things (but see also,Flach, Hancock, Caird, & Vicente, 1995). After severaldiscussions with Gibson, Norman came to appreciate theconcept of affordances but was reluctant to accept Gibson’sideas of direct perception. Inclined to a cognitive approach,Norman claimed instead, “I believe that affordances resultfrom the mental interpretation of things, based on our pastknowledge and experience applied to our perception of thethings about us” (Norman, 1988/2002, p. 219, emphasesadded). In addition, he argued, again contrary to Gibson,that affordances are not mere opportunities for action, butare perceived action possibilities that suggest actions to ananimal. Although Norman (1999) later rectified his initialmisconstrual of the Gibsonian concept, his idea of affor-dances as perceived suggestions for action was adopted bymany industrial designers. In the aforementioned study byJu and Takayama (2009), for example, it is claimed that “Toimprove an affordance to enter a building, a designer wouldmake the passability of the doorway more obvious” (p. 2).This is an unfortunate misconstrual of the concept ofaffordances. After all, in Gibson’s terms, to improve anaffordance is to make the environment more compatiblewith the action capabilities of the human body, not tomakeit more prominent. Yet, Ju and Takayama’s study nicelyillustrates the fact that by manipulating a design one canmake certain actions more likely to occur. Apparently, theenvironment is not a neutral manifold of action possibilitiesthe agent simply chooses from; rather, the environmentcan invite a certain action or even urge a person to dosomething.

2 It is interesting to note that Cisek and Kalaska (2010) recentlydeveloped a theory of action selection that is similar to Reed ideas butthat is far more sophisticated and grounded in contemporaryneuroscience.



4.2. Lessons from the built environment

The idea that theenvironment canprompt certain actionsis also highlighted in architecture. At a symposium onperception in architecture, Gibson (1982b) once made thebold statement that “architecture and design do not havea satisfactory theoretical basis” (p. 413) and suggested thatecological psychology in general and the concept of affor-dances in particular can provide such a basis. In designingbuildings, architects should indeedbe aware of how to createaffordances and how they are perceived (Warren, 1995).However, designing a building is not merely designinga layout of action possibilities or creating an aestheticexperience. Architects can also contrive places that invitecertain behaviors. In his book Lessons for Students in Archi-tecture, theDutchstructuralistHertzberger (1991) illustratedhow forms can dictate certain actions and how architectscould design buildings with more functional flexibility. Inline with Gibson (1979/1986, p. 134), Hertzberger empha-sized that a single object generally affords different behav-iors to a certain animal. However, an object canpresent itselfto the agent as serving one single function, giving rise to aninflexible use of the object. As he put it, “Objects thatpresent themselves explicitly and exclusively for a specificpurposede.g. for sitting ondappear to be unsuitable forother purposes” (Hertzberger,1991, p.177; emphasis added).Hence, Hertzberger encouraged architects to design more‘neutral places’ that stimulate the creativity of the agent,implying that more of its affordances are discovered andused. A famous example of such a ‘neutral place’ is themasonryplinth courses in the lobbyof theVredenburgMusicCenter in Utrecht, the Netherlands. Contrary to a chair ortable, these plinth courses are generally perceived as havingmultiple affordances (e.g., a place to sit or to put a drink on)and are used accordingly by the majority of people.

Note that many chairs and these masonry plinth coursessharemultiple affordances for human beings. They are boththings a person can sit on, stand on, put a booklet on, and soon. Yet a chair is mostly used in a single way, while these

plinth courses are generally used in multiple ways by anindividual. Hence, what designers or architects create arenot mere opportunities for action but invitations that canhave a severe influence on the behavior personswill exhibitin the building. Nice illustrations of such invitations in thebuilt environment can also be found in the work of theartist Krijn de Koning (2000). He creates mostly temporarysculptures consisting of horizontal and vertical planes.With these often site-specific pieces of art, de Koningexamined how architecture constraints and invites us.Consider, for example, the installation exhibited in 1994 atArtcite Inc., Windsor, Canada (see Fig. 1). Like nearly everyplace in a building, this specific configuration of walls andfloors affords many different types of action to a humanbeing. Indeed the number of activities that an individualcan perform at this site is infinite (e.g., sitting on theplateau, sitting on the ground, touching the walls with thehand, turning around etc.). Yet the vast majority of peoplewould be invited by this configuration to follow the createdpath and enter the door.

4.3. Lessons from phenomenology

Phenomenologists have also stressed that the environ-ment can invite certain acts. Although Gibson did not referexplicitly to phenomenologists in the development of hisecological approach, the similarities between Gibson’secological psychology and the work of Merleau-Ponty arestriking (see e.g., Glotzbach & Heft, 1982; Heft, 2001).Hence, it is of no surprise that the concept of affordancesnow features in several recent phenomenological analyses.Examining unreflective actions, Rietveld (2008), forexample, talked about the “individual’s responsiveness toaffordances” (p. 976; see also McDowell, 2007). His maininterest is in what he called the ‘phenomenology of affor-dances’, an interest that he shares with Dreyfus and Kelly(2007). In these phenomenological accounts of affordan-ces, the inviting potential of the environment is oftenstressed. Indeed, Dreyfus and Kelly claimed that when

Fig. 1. An installation by Krijn de Koning exhibited at Artcite Inc., Windsor, Canada (1994).



affordances were defined in terms of phenomenology, theyare not mere possibilities for action, but are better under-stood as ‘solicitations to act’. They drew upon the work ofGestalt psychologists in explaining this quality. “We use theGestaltist’s term ‘solicits’ to refer to a datum of phenome-nology. To say that the world solicits a certain activity is tosay that the agent feels immediately drawn to act a certainway” (Dreyfus & Kelly, 2007, p. 52, emphases in original).

Dreyfus and Kelly’s (2007) adoption of this Gestaltist’sidea in developing their concept of affordance might seemodd. Indeed, as mentioned earlier, although Gibson wasinspired by Koffka and Lewin, he criticized their idea of the‘demand character of the environment’. Dreyfus and Kelly(2007, footnote 3) were aware of this critique andstressed that Gibson indeed did not define affordances interms of phenomenology but conceived of them asecological facts that exist independently of the actor’sneeds and responses. However, when defined in terms ofphenomenology, they claimed that this solicitation char-acter of affordances is crucial. Like with Norman’s initialconstrual of affordances, one might therefore argue thatthis phenomenological account of affordances does not dojustice to Gibson’s original conceptualization. Yet what ismost interesting for the purposes of this paper is thatphenomenologists have stressed that the environmentsolicits actions and that agents respond unreflectively tosuch demands (e.g., Dreyfus & Kelly, 2007; Merleau-Ponty,1945/2002; Rietveld, 2008).

5. Reconsidering the relationship betweenaffordances and agency

In the previous sectionwe suggested, based on differentfields of inquiry, that affordances can also invite actions.This brings us to the final part of this paper. What is thenature of this inviting character of the environment? Doesit depend on the intentions of the agent as the Gestaltpsychologists had argued? Is the inviting character part ofthe phenomenological experience of affordances? Or is theinvitation an ecological phenomenon that exists by virtueof a relation between the agent and the environmentalproperties? Moreover, if the environment is not simplyamanifold of action possibilities the agent can choose from,but has the potential to invite the animal to certain actions,then what is the status of Reed’s ecological account ofagency? How are we to understand agency? In theremainder of the paper, we first discuss the nature ofinvitations. We suggest a mutualist perspective on invita-tions; they exist by virtue of a relation between the envi-ronment and the actor. We end our paper by reconsideringthe relationship between affordances and agency. It isargued that by recognizing that affordances can also invite,the concept of agency becomes more ecological. It impliesthat the animal–environment system is the proper unit ofanalysis in understanding agency.

5.1. On the nature of the inviting character

Before we proceed to the discussion on the invitingcharacter, it is important to be specific about our ideas onthe nature of affordances. As has become clear in the

previous sections, there are quite some discussions on theontology of affordances. Hence, a clarification of our posi-tion is in order.

First, we follow Gibson in conceiving of affordances asopportunities for action. Contrary to several researchers inthe field of industrial design, we do not think of affor-dances as perceived action possibilities. An affordanceexists even if it is not perceived (Chemero, 2009; Heft,2001; Michaels, 2003). For example, the floor affordswalking upon for some individual even if she is looking ather computer screen to write a paper. Moreover, devel-opmental psychologists have suggested that children haveto discover what action possibilities the environmentoffers them (e.g., Adolph, Joh, & Eppler, 2010). That is, theaffordances exist even before they are firstly perceived bythe developing child.

Second, we also agree with Gibson that affordances donot change as the intentions or needs of the actor change.As just mentioned, the Gestalt psychologists argued thatthe ‘demand character’ of the environment depends on theneeds of the agent. To reiterate Koffka’s example, a mailboxhas a demand character if and only if an agent wants to posta letter. However, as Gibson forcefully claimed, the affor-dances of the mailbox are irrespective of the intentions ofthe agentdit affords posting a letter even if one is involvedin a completely different type of activity. That is, in linewith Gibson and many other ecological psychologists (e.g.,Heft, 2001; Michaels, 2003), we think of affordances asaction-relevant properties of the environment that aredefined with respect to the animals’ action capabilities butexist independently of their needs and intentions.

However, by suggesting that affordances can also invitebehavior, we move beyond Gibson’s original conception ofaffordances as mere action possibilities. It is important toemphasize that we do not equate affordances with invita-tions. We have seen that several phenomenologicalaccounts conceive of the experience of an affordance as an“experience in which the world solicits a certain activity”(Dreyfus & Kelly, 2007, p. 52). We disagree with thisconceptualization. Although (experienced) affordances canhave the potential to invite a certain activity, the vastmajority of affordances do not. As mentioned earlier,a single object generally affords multiple behaviors to anindividual, and not all of these affordances invite. To reit-erate an example, a single piece of paper affords an infinitenumber of actions to a human (cf. Cutting, 1982), but onlysome of these affordances might invite the agent to act.Hence, we conceive of affordances not as invitations or‘solicitations to act’, but as action possibilities that caninvite. The crucial question now is: How are we to under-stand these invitations? Are invitations environmentalproperties, mental properties, or do they, like affordances,“cut across the dichotomy of subjective–objective” (Gibson,1979/1986, p. 129).

We are inclined to a mutualist perspective on invita-tionsdthey depend on the relation between the physicalproperties of the environment and the agent. Consider, forexample, the installation by Krijn de Koning that wasmentioned in Section 4.2. This installation is likely to invitethe vast majority of people to walk along the created path.However, this inviting character of the affordance depends



on the agent–environment relation, arguably in a multidi-mensional way. For example, the created path also affordswalking along for the toddler, but the developing child ismore likely to be invited by the installation to climb on theplateau, as many museum-visiting parents know. Hence,although the path affords locomotion for both the toddlerand the adult, it invites this behavior primarily for thegrown-up. This suggests that whether an affordance invitesis an animal-relative property of the environment as well.

However, it is important to stress an ontological differ-ence between possibilities and invitations. As mentionedearlier, action possibilities exist even if they are notperceived. That is, affordances do not depend on an actuallypresent observer that is in epistemic contact with them. Foran affordance to invite, on the other hand, such an observeris required. An affordance can attract or repel certainbehavior of an agent if and only if the agent perceives thisaffordance. As an example, the toddler can only be invitedto climb the above-mentioned plateau if she perceives thisaffordance. Hence, affordances can exist independently ofan actual observer, but for an affordance to invite such anobserver is indispensable.

Much work remains to be done in examining whatanimal properties invitations are relative to. Over the lastdecades, ecological psychologists have made serious prog-ress in measuring affordances. Initially, the environmentwas measured with respect to body scale (e.g., Warren,1984). Based on experimental findings (e.g., Konczak,Meeuwsen, & Cress, 1992) and theoretical considerations(e.g., Chemero, 2003), ecological psychologists came torealize that body scale only partly determines the actioncapabilities of the agent and that more organismal factors(e.g., flexibility of the action system, muscle strength) needto be included in themeasurement of affordances. It is likelythat whether an affordance invites an action to a certainanimal also depends on multiple organismal factors. Belowwe will present a list of features that are likely to beinvolved. However, we do not intend this list to beexhaustive or definite. Instead, the enumeration of prop-erties is tentative and needs to be complemented, dis-cussed, and criticized.

First, among the factors that are involved are the actioncapabilities of the agent. The relation between thesecapabilities and the environmental properties not onlydeterminewhat actions can be performed (i.e., affordances)but also what actions are invited. After all, action capabil-ities also determine the ease with which affordances can beutilized. And if the actualization of an affordance requiresgreat effort, it is not likely to invite the agent to act. In hispioneering study on stair climbing, Warren (1984) madea distinction between ‘critical points’ and ‘optimal points’.Critical points are the environmental boundaries at whichpossibilities for individual actors change. For example,when riser height increases, there will be a point at whichthe stair no longer affords bipedal climbing for an indi-vidual. Optimal points, on the other hand, are “stable,preferred regions of minimum energy expenditure, or ‘bestfit’ affordances” (Warren, 1984, p. 686). Depending on anindividual’s leg length, there will be a riser height at whichclimbing is least energy consuming. Whether an affordanceinvites behavior is likely to depend on, among other things,

the amount of effort it takes to utilize the action possibility.Stairs that are hardly climbable (i.e., close to the criticalpoint) are not likely to invite the agent to act, whereasstairs that are easy to climb (i.e., at the optimal point) mightdo so.

Second, from an evolutionary perspective, some affor-dances are more important than others (e.g., Withagen &Chemero, 2009). And affordances that are crucial forsurvival and reproduction (e.g., objects or animals thatafford danger, shelter, or nutrition) are likely to attract orrepel the agent. For example, a human that encountersa lion on a safari is likely to flee into his car as soon aspossible. Such an affordance will be acted upon immedi-ately, irrespective of the intentions of the actor at thatmoment in time (e.g. to take a picture, to pee). Hence,whether an affordance invites is not contingent on thetemporary intentions of the actor as the Gestalt psycholo-gists argued, but can be more permanent and rooted in ourevolutionary history.

Third, culture is likely to be involved as well. As we haveseen in Section 4.2, Hertzberger (1991) claimed that certainobjects present themselves to an agent as serving a singlefunction. Although offering multiple action possibilities toa human being, a chair, for example, is often perceived as anobject to sit on. The fact that this affordance stands-out isarguably a result of culture. However, this does not meanthat the inviting character is a mental product. As Ingold(2000) argued, culture is not a collection of mental repre-sentations residing in the brain that shapes our perceptionand action. Instead, cultural variations are better thought ofas variations in perceptual-motor skills. Culture grows inthe body, giving rise to a particular responsiveness tocertain affordances in the environment.

Fourth, aspects like personal history seem also relevant.Indeed, members of the same culture are often attracted todifferent objects or are invited by the same object to dodifferent things. As an example, although chocolate mayafford eating for the vast majority of people, there issubstantial variation in whether and how people areattracted to it. Some people are almost addicted and cannotwait to eat it; others might not like it and prefer to eatsomething else. It is important to note that the invitationcan also vary over time and might change on a moment-to-moment basis. For example, a person who initially likedchocolate but had suffered from gastroenteritis after eatingit is likely to be repelled by its affordance for some time.Again, the fact that personal experience partly determinesthe invitation does not mean that the inviting characteremerges in the mind. Like cultural variation, individualdifferences in perception can also be explained in terms ofvariation in what information is exploited or in the bodilyresponsiveness to such information (see also Withagen &van der Kamp, 2010).

5.2. Affordances and agency

The above list of factors is far from complete and muchwork remains to be done to specify the conditions underwhich affordances invite. However, it is beyond the scope ofthis paper to do so. Instead, wewould like to end this paperwith some implications of the idea of invitations for the



concept of agency. Indeed, as mentioned from the outset,the concept of agency is central to the ecological approachto psychology. What happens to this concept when werecognize that the environment can invite an animal to actin a certain way?

First, in our view, invitations entail agency. As previ-ously mentioned, an affordance can invite behavior if andonly if an agent perceives it. If affordances are not perceived(or even have not been discovered) they do not have thepotential to attract (or repel) the according behavior of theagent. Hence, a prerequisite for affordances to invite is anactually present observer that actively explores the affor-dances of its environment.

Second, invitations are not causes. An invitation canalways be declined. Although Krijn de Koning’s artworkinvites us to follow the path, we can decide not to do it.Hence, even if we recognize that affordances can also invitebehavior, the concept of agency retains. However, therecognition of the inviting character nurtures an account ofagency that is fully centered around the animal–environ-ment relationship. We have seen that Reed (1993, 1996a)followed Gibson in conceiving of affordances as actionpossibilities. Consequently, his conception of agency isbased on the idea of self-controld“[animals] make theirway in the world” (Reed, 1996a, p. 19; emphases in original).However, if we recognize that affordances can also invitebehavior, we are forced to a conception of agency that putsthe animal–environment relation much more central.When actively exploring the environment, the agent isattracted or repelled by some of its affordances, and theensuing behavior is partly the result of these invitations.This means that to understand how animals make theirway in the world, the inviting character of affordancesshould be taken central, implying that the animal–envi-ronment relation is the proper unit of analysis in under-standing agency.

Third, it is important to stress that animals often unre-flectively respond to the environment’s solicitations. AsDreyfus and Kelly (2007) put it,

We sense the world’s solicitations and respond to theircall all the time. In backing away from the ‘close talker,’in stepping skillfully over the obstacle, in reaching‘automatically’ for the proffered handshake, we findourselves acting in definite ways without ever havingdecided to do so. In responding to the environment thisway we feel ourselves giving in to its demands. (p. 52;emphases added)

That is, agency should not be understood as a process ofdeliberately making a decision about which affordance willbe actualized. Instead, animals unreflectively act upon theaffordances of the environment that attract and repel them(see also Rietveld, 2008). This again highlights the fact thatagency emerges in an animal–environment system (seealso, Chow, Davids, Hristovski, Araújo, & Passos, 2010).

6. Concluding remarks

In this paper we have discussed the relationshipbetween affordances and agency. By claiming that affor-dances do not cause behavior but simply make it possible,

Gibson made room for the idea of agency. In his view, ananimal regulates his encounters with the environment byselecting affordances and utilizing them. Although Gibsonhas been criticized for not offering a full-blown theory ofhow animals make their way in the world (e.g., Cutting,1982), he placed the idea of agency back on the agenda ofpsychology. Reed (1993, 1996a), one of the prevailing neo-Gibsonians, applauded Gibson’s emphasis on agency andfurthered an ecological account of this distinctive feature ofthe animate world.

These ecological accounts of agency, however, werebased on the conception of affordances as action possibil-ities. In the present paper, we have suggested, drawingupon industrial design, architecture, and phenomenology,that affordances are not mere possibilities for action butthat they can also invite behavior. Although we havesketched our ideas on the inviting character of affordances(contrasting our ideas with those of Gestalt psychologistsand phenomenologists), much work remains to be done tofurther this idea. Importantly, to develop an ecologicalaccount of invitations, we have to specify the environ-mental and organismal factors that determine whether andwhen an affordance invites. We have made a couple ofsuggestions in this paper, but a full-blown account ofinvitations awaits much theoretical and empirical work. Inour view, however, this work is essential not only tounderstand the nature of the animal’s environment butalso to come to grips with the concept of agency and theallied psychological processes as decision-making andaffordance selection. As Gibson (1966, 1979/1986) madeclear, an understanding of the perceptual process requiresa thorough examination of the information. In like fashion,an understanding of agency requires an in-depth study ofwhether, when and how affordances attract or repel ananimal.

Acknowledgments

RobWithagenwould like to thank Bart Withagen for hislessons in architecture during their joint bicycle trips. Weare grateful to Krijn de Koning for supplying and allowingus to use the photograph in Fig. 1.

References

Adolph, K. E., Joh, A. S., & Eppler, M. A. (2010). Infant’s perception ofaffordances of slopes under high- and low-friction conditions. Journalof Experimental Psychology: Human Perception and Performance, 36,797–811.

Chemero, A. (2003). An outline of a theory of affordances. EcologicalPsychology, 15, 181–195.

Chemero, A. (2009). Radical embodied cognitive science. MIT Press.Chow, J. Y., Davids, K., Hristovski, R., Araújo, D., & Passos, P. (2010).

Nonlinear pedagogy: learning design for self-organizing neurobio-logical systems. New Ideas in Psychology, 29, 189–200.

Cisek, P., & Kalaska, J. F. (2010). Neural mechanisms for interacting withaworld full of actionchoices.Annual ReviewofNeuroscience, 33, 269–298.

Costall, A. (1999). An iconoclast’s triptych: Edward Reed’s ecologicalphilosophy. Theory & Psychology, 9, 411–416.

Costall, A., & Morris, P. The ‘textbook Gibson’: a case study in theassimilation of dissidence, submitted for publication.

Cutting, J. E. (1982). Two ecological perspectives: Gibson vs. Turvey andShaw. American Journal of Psychology, 95, 199–222.

de Koning, K. (2000). Inside/outside. Rotterdam: NAi Publishers.Desmet, P. M. A., & Hekkert, P. (2007). Framework of product experience.

International Journal of Design, 1, 57–66.



Desmet, P. M. A., & Hekkert, P. (2009). Special issues editorial: design &emotion. International Journal of Design, 3, 1–6.

Dijksterhuis, E. J. (1950)). De mechanisering van het wereldbeeld. Amster-dam: Meulenhoff.

Dreyfus, H., & Kelly, S. D. (2007). Heterophenomenology: heavy-handedsleight-of-hand. Phenomenology and the Cognitive Sciences, 1, 413–425.

Flach, J., Hancock, P., Caird, J., & Vicente, K. (Eds.). (1995). Globalperspectives on the ecology of human–machine systems. Hillsdale, NJ:Lawrence Erlbaum Associates.

Ghiselin, M. T. (1969). The triumph of the Darwinian method. Berkeley:University of California Press.

Gibson, E. J. (1994). Has psychology a future? Psychological Science, 5,69–76.

Gibson, J. J. (1966). The senses considered as perceptual systems. Boston,MA: Houghton Mifflin.

Gibson, J. J. (1979/1986). The ecological approach to visual perception.Boston, MA: Houghton Mifflin.

Gibson, J. J. (1982a). Notes on action. In E. Reed, & R. Jones (Eds.), Reasonsfor realism: The selected essays of James J. Gibson (pp. 385–392). Hill-sdale, NJ: Erlbaum.

Gibson, J. J. (1982b). Notes on affordances. In E. Reed, & R. Jones (Eds.),Reasons for realism: The selected essays of James J. Gibson (pp. 401–418). Hillsdale, NJ: Erlbaum.

Glotzbach, P., & Heft, H. (1982). Ecological and phenomenologicalcontributions of the psychology of perception. Nous, 16, 108–121.

Heft, H. (2001). Ecological psychology in context: James Gibson, RogerBarker, and the legacy of William James’s radical empiricism. Mahwah,NJ: Lawrence Erlbaum Associates.

Hertzberger, H. (1991). Lessons for students in architecture. Rotterdam: 010.Hilgard, E. (1987). History of psychology in America. New York: McGraw-

Hill.Holt, E. B. (1914/1973). The Freudian wish and its place in ethics. New York:

Holt.Iberall, A. S. (1977). A field and circuit thermodynamics for integrative

physiology: I. Introduction to the general notions. American Journal ofPhysiology, 233, R171–R180.

Ingold, T. (2000). The perception of the environment: Essays in livelihood,dwelling and skill. London: Routledge.

Ju, W., & Takayama, L. (2009). Approachability: how people interpretautomatic door movement as gesture. International Journal of Design,3, 1–10.

Kahneman, D., Diener, E., & Schwarz, N. (Eds.). (1999). Well-being: Thefoundations of hedonic psychology. New York: Russell Sage Foundation.

Koffka, K. (1935). Principles of gestalt psychology. London: LundHumphries.

Konczak, J., Meeuwsen, H. J., & Cress, M. E. (1992). Changing affordancesin stair climbing: the perception of maximum climbability in youngand old adults. Journal of Experimental Psychology: Human Perceptionand Performance, 18, 691–697.

McDowell, J. (2007). What myth? Inquiry, 50, 338–351.Merleau-Ponty, M. (1945/2002). Phenomenology of perception. London:

Routledge.

Michaels, C. F. (2003). Affordances: four points of debate. EcologicalPsychology, 15, 135–148.

Neisser, U. (1967). Cognitive psychology. New York: Appleton, Century, &Crofts.

Norman, D. A. (1988/2002). The design of everyday things. New York: BasicBooks.

Norman, D. A. (1999). Affordances, conventions, and design. Interactions,6, 38–42.

Reed, E. S. (1980). Information pickup is the activity of perception.Behavioral and Brain Sciences, 3, 397–398.

Reed, E. S. (1982). An outline of a theory of action systems. Journal ofMotor Behavior, 14, 98–134.

Reed, E. S. (1985). An ecological approach to the evolution of behavior. InT. Johnston, & A. Pietrewicz (Eds.), Issues in the ecological study oflearning (pp. 357–383). Hillsdale, NJ: Erlbaum.

Reed, E. S. (1988a). James J. Gibson and the psychology of perception. NewHaven: Yale University Press.

Reed, E. S. (1988b). Applying the theory of action systems to the study ofmotor skills. In O. Meijer, & K. Roth (Eds.), Studying complex motorskills: The motor-action controversy (pp. 45–86). Amsterdam: North-Holland.

Reed, E. S. (1993). The intention to use a specific affordance: a frameworkfor psychology. In R. Wozniak, & K. Fisscher (Eds.), Development incontext: Acting and thinking in specific environments (pp. 45–75).Hillsdale, NJ: Lawrence Erlbaum Associates.

Reed, E. S. (1996a). Encountering the world: Toward an ecologicalpsychology. New York, NY: Oxford University Press.

Reed, E. S. (1996b). The necessity of experience. New York: Yale UniversityPress.

Rietveld, E. (2008). Situated normativity: the normative aspect ofembodied cognition in unreflective action. Mind, 117, 973–1001.

Shaw, R. (2001). Process, acts, and experiences: three stances on theproblem of intentionality. Ecological Psychology, 13, 275–314.

Warren, W. H. (1984). Perceiving affordances: visual guidance of stairclimbing. Journal of Experimental Psychology: Human Perception andPerformance, 10, 683–703.

Warren, W. H. (1995). Constructing an econiche. In J. Flach, P. Hancock, J.Caird, & K. Vicente (Eds.), Global perspectives on the ecology of human–machine systems (pp. 210–237). Hillsdale, NJ: Lawrence ErlbaumAssociates.

Withagen, R., & Chemero, A. (2009). Naturalizing perception: developingthe Gibsonian approach to perception along evolutionary lines.Theory & Psychology, 19, 363–389.

Withagen, R., & Michaels, C. F. (2005). On ecological conceptualizationsof perceptual systems and action systems. Theory & Psychology, 15,603–620.

Withagen, R., & van der Kamp, J. (2010). Towards a new ecologicalconception of perceptual information: lessons from a developmentalsystems perspective. Human Movement Science, 29, 149–163.

Withagen, R., & van Wermeskerken, M. (2010). The role of affordances inthe evolutionary process reconsidered: a niche constructionperspective. Theory & Psychology, 20, 489–510.


Affordances can invite behavior: Reconsidering the relationship between affordances and agency

Documents