The Gibsonian movement and Koffka’s Principles of Gestalt Psychology

Abstract

In recent decades, several ecologically inclined authors have adopted the Gestalt idea of demand characters. Yet, James Gibson, the founder of the ecological approach, although being inspired by Koffka, was critical of many of his ideas, including the contention that the environment calls for certain actions. This article examines why Gibson was so reluctant to accept this concept of demand characters. To that end, the relationship between Gibson’s ecological approach and Gestalt psychology is scrutinized. After an exploration of the parallels between the frameworks of Gibson and Koffka, Gibson’s critique of Koffka’s masterpiece Principles of Gestalt Psychology is evaluated. It is argued that although Gibson’s claim about the mind–world dualism in which Koffka’s perspective is rooted is arguably valid, neither Gibson nor some of his recent devotees take Koffka’s insights into the qualities of experience sufficiently seriously in their theorizing.

What I have said here is intended to commemorate Koffka 35 years after the publication of his book. I hope my admiration is evident, for my debt to him is very great. My ideas about visual perception depend on his ideas, if not by agreement then by contrast. (Gibson, 1971, p. 9)

Over the last decades, several ecologically inclined authors have assimilated ideas of Gestalt psychology into their theorizing. This is partly the result of a renewed interest in phenomenology in the Gibsonian movement (e.g., Dings, 2018, 2021; Käufer & Chemero, 2015; Rietveld & Kiverstein, 2014; Van Dijk & Rietveld, 2017; Withagen, 2022). Although some interesting papers on the intersection of ecological psychology and phenomenology appeared in the 1980s and 1990s (e.g., Heft, 1989; Ingold, 1996/2000; Shotter, 1983), there is a recent upsurge of phenomenological considerations within ecological theorizing, stemming in large measure from Dreyfus’ writings on the invitational qualities of the environment. In his phenomenology, Dreyfus (2005/2014) adopted the Gibsonian concept of affordances—the environment is said to consist of action possibilities. Yet, Dreyfus and Kelly (2007) were equally drawing upon insights from Gestalt psychology when arguing that when experienced, affordances show up as invitations. And it is this very idea of the demand or invitational character of the environment that receives traction in current ecological thinking (e.g., Dings, 2018, 2021; Heft, 2010; Käufer & Chemero, 2015; Rietveld & Kiverstein, 2014; Withagen, 2022).

Gibson himself, however, had an ambivalent relationship with the Gestalt psychologists. Although he was generally hesitant to acknowledge the influence other thinkers had on him (Costall, 2023), in the prefaces of each of his three books, Gibson (1950, 1966, 1979/1986) expressed his debt to Kurt Koffka. The author of Principles of Gestalt Psychology (Koffka, 1935/1999) had a lasting influence on Gibson’s work. For example, when laying out the antecedents of his revolutionary concept of affordances, Gibson acknowledged the work of Koffka (Gibson, 1972/2020c, pp. 409–410, 1979/1986, pp. 138–139). Gibson (1972/2020c) asserted that his concept of affordances is “in the same line of theoretical development” (p. 410) as Koffka’s concept of demand character. Yet, Gibson was unwilling to adopt the very idea that the environment solicits action, and, at points, strongly distanced himself from Koffka.

In the present article, I will examine why Gibson was so reluctant to accept the Gestalt concept of demand character. In light of the current popularity of the idea of invitations within the ecological movement, further insight into this matter is needed. Were Gibson’s grounds for rejecting it valid? To answer this question, I will discuss Koffka’s (1935/1999) Principles in the context of Gibson’s and current ecological theorizing. The reason for centralizing this book is that it was arguably the main work in Gestalt psychology Gibson took inspiration from and issue with. I will start by describing Gibson’s early encounters with the Gestalt psychologists, their work, and how that influenced his theorizing. Then I will lay out Gibson’s (sometimes) forceful critique of some of the central pillars of Koffka’s framework. Although Gibson’s claim about the mind–world dualism in which Koffka’s thinking is rooted is arguably justified, I will claim that both Gibson and some current ecologically inclined thinkers have insufficiently recognized some essential qualities of experience that Koffka (1935/1999) had emphasized in his Principles. Hence, a further yet careful integration of Koffka’s insights into the ecological approach is called for.

Taking inspiration from Koffka

Already during his graduate studies at Princeton, Gibson was exposed to Gestalt psychology and the phenomenological tradition it was rooted in. Although Gibson was perhaps most inspired by his graduate school mentor, Holt (Heft, 2001; see also Gibson, 1967/2020a), among his other teachers who influenced him at the newly formed psychology department were Langfeld and Carmichael (see Raja, 2019). The latter had spent a year studying in Berlin, working with Köhler and Wertheimer, before he came to Princeton (Mead, 1974). And Langfeld did his PhD with Stumpf, an authority in the field of phenomenology who founded the Berlin Psychological Institute. Stumpf had supervised, among others, Koffka, Lewin, and Köhler in their doctorate research (Ash, 1995). In fact, Gestaltism was founded at Stumpf’s institute. Throughout his life, Langfeld stayed “proud of his German degree” (Welford, 1958, p. 1241) and kept close ties with Köhler. In fact, Langfeld was one of the first to take the phenomenological tradition in empirical psychology to the United States (Heft, 2001). At a keynote in Europe delivered a few years after the Second World War came to an end, Langfeld stressed that the Gestalt principles and questions had become one of the toolkits of every American psychologist (Pratt, 1969).

After taking Langfeld’s course in experimental psychology at Princeton, Gibson was offered a graduate assistantship which he considered to be the start of his academic career. “This stroke of luck gave me an identity; I was an academic; not a philosopher, but even better, a psychologist” (Gibson, 1967/2020a, p. 9). Although Gibson developed a deep respect for the phenomenological tradition Gestalt psychology was rooted in, he was already critical of several Gestalt principles. Indeed, his doctoral dissertation was devoted to proving that Wulf, one of Koffka’s students in Berlin, was wrong (Gibson, 1929; see also Lobo et al., 2018). As he wrote years later in his autobiography:

[Memories do not change] spontaneously toward better Gestalten. The drawing of my subjects differed from the originals only in accordance with laws of perceptual habit, not laws of dynamic self-distribution, I concluded with great confidence. Form perception was learned. Otherwise one fell into the arms of Immanuel Kant. (Gibson, 1967/2020a, pp. 9–10)

In 1928, after receiving his degree, Gibson moved to Smith College, at which Koffka had been appointed a year earlier. Koffka took a position in the United States some years before the Nazis came to power in Germany. After some visiting professorships, Koffka accepted the generous offer of a five-year research appointment, free from any teaching duties, at Smith College. It was at this small women’s college in Northampton, Massachusetts, that Koffka (1935/1999) wrote his landmark book Principles of Gestalt Psychology. In addition, he organized a weekly seminar (from 1928 to 1941) in which Gibson also took part. Gibson was now learning Gestalt psychology from one of its founding fathers. And although he scribbled some very critical remarks during Koffka’s seminars (see later in this article), his later ecological work followed a number of Gestalt principles.

In the first place, Gibson adopted the phenomenological attitude that Gestalt psychology was relying on (e.g., Heft, 2001). Importantly, and as Koffka (1935/1999) stressed, this method is key to psychology, both in determining the explanandum of the discipline and in revealing the explanans:

A good description of a phenomenon may by itself rule out a number of theories and indicate definite features which a true theory must possess. We call this kind of observation “phenomenology”, a word which has several other meanings which must not be confused with ours. For us phenomenology means as naïve and full a description of direct experience as possible. (p. 73)

In like fashion, Gibson stressed the importance of phenomenology. After discussing the movements of the eyes and the head in the process of visual perception, Gibson (1966) noted: “The phenomenal world [emphasis added] seems to stay put, to remain upright, and to surround one completely. This experience is what a theory of perception must explain [emphasis added]” (p. 253). ¹

Among the theories that Koffka rejected while adopting the phenomenological method was the building-block theory of consciousness. Experiences are not made of sensations or other mental elements, as many constructivist theories had asserted. Rather, experience is an integrated whole—a Gestalt—that is not built up from and cannot be analyzed into elements. After laying out the first perceptual principles, Koffka (1935/1999) concluded:

Our discussion has dealt with very elementary objects . . . . But even these humble objects reveal that our reality is not a mere collocation of elemental facts, but consist of units in which no part exists by itself, where each part points beyond itself and implies a larger whole. . . . It has been said: The whole is more than the sum of its parts. It is more correct to say that the whole is something else than the sum of its parts . . . . (pp. 175–176)

Similarly, Gibson strongly opposed the sensation-based theories. Sensations, he asserted, are not the raw material out of which perceptual experience is formed; rather they are incidental to the perceptual process.

The active observer gets invariant perceptions despite varying sensations. [They] perceive a constant object by vision despite changing sensations of light; [they] perceive a constant object by feel despite changing sensations of pressure; [they] perceive the same source of sound despite changing sensations of loudness in [their] ears. The hypothesis is that constant perception depends on the ability of the individual to detect the invariants, and that [they] ordinarily pay no attention whatever to the flux of changing sensations. (Gibson, 1966, p. 3)

Apart from this agreement on the nature of consciousness, both Koffka and Gibson stressed that the route to understand it is via behavior. As Koffka (1935/1999) put it, “if we start with behavior it is easier to find a place for consciousness and mind than it is to find a place for behavior if we start with mind or consciousness” (p. 25). Although psychology should account for mind and consciousness, it is first and foremost the study of an organism behaving in its environment. Moreover, this study provides a window into the mind. Gibson, too, centralized action in his ecological approach. Rebelling against the spectator perspective, Gibson claimed that the primary function of perception is not to gain knowledge of the environment but to guide our behavior in it. Indeed, our perceptual systems have evolved to allow us to cope adaptively with the environment. Moreover, action is necessary for perceiving. The very point of Gibson’s (1966) book on the senses was that perceiving is an activity of an animal in a richly structured environment. And it is the study of this activity, and the information that is detected in the process, that allows one to understand our perceptual experiences of the environment.

In their perspectives, Koffka and Gibson alike focused on what Koffka called the molar behavior of animals (see e.g., Kiverstein et al., 2021). In the second chapter of his Principles, Koffka followed Tolman (1932) in making a distinction between molar and molecular behavior. Examples of molar behavior are, for example, going to the grocery shop, making a coffee, and grasping a banana. It is behavior of an animal directed at an environment. Molecular behavior, on the other hand, “is something very different” (Koffka, 1935/1999, p. 25). It refers to the excitation of the nerves, the contraction of muscle. It is behavior that “takes place within the organism [emphasis added] and is only initiated by environmental factors, called the stimuli” (p. 27). Although Koffka (1935/1999) asserted that his Gestalt psychology looks “ultimately for physiological explanations” (p. 63), he criticized psychology for being mesmerized by molecular behavior:

(1) It attributes reality to parts, denying it to the wholes which these parts compose: the molar has to be resolved into the molecular; (2) as a result of this, psychology should forever remain exposed to the criticism of the Moral Sciences . . . . Meaning and significance could have no possible place in such a molecular system [emphasis added]. (Koffka, 1935/1999, p. 26)

And although Gibson discussed the implications of his ecological perspective for neuropsychology (see e.g., Gibson, 1966, pp. 4–5, p. 56), the focus in his work is equally on the behavior of the animal in its environment (see also Raja, 2019). In fact, he asserted that the earlier established facts of sciences that “zoom in” are irrelevant for understanding the perceiving and acting animal.

Anatomists knew about the eye as an organ, but not about what it can do. Physiologists knew about the nerve cells in the retina and how they work but not how the visual system works. What they knew did not seem to be relevant [emphasis added]. They could create holograms, prescribe spectacles, and cure diseases of the eye, and these are splendid accomplishments, but they could not explain vision [emphasis added]. (Gibson, 1979/1986, p. xiii)

To understand molar behavior, Koffka as well as Gibson argued that we have to get rid of the mechanistic thinking that had gripped psychology for centuries. The theory that “excitations, starting somewhere, traveling along a nerve, being transmitted to another nerve, from that to a third, until finally they gave rise to a muscle contraction or a gland secretion” (Koffka, 1935/1999, p. 54) falls short in explaining the “orderliness and meaningfulness” (p. 310) of behavior. Although dismissing mechanistic physics from the very start, Koffka (and other Gestalt psychologists) embraced the field concept in physics that was developed in the context of electricity and magnetism. Faraday had explained the movements of a body in terms of a “distributed systems of strains and stresses . . . which determines the geometry of space” (Koffka, 1935/1999, p. 42). The Gestalt psychologists applied this field theory to psychology, and this, in turn, inspired the early Gibson. Indeed, his first paper on locomotion (i.e., automobile driving), which Gibson wrote together with the engineer Crooks, explicitly followed a Lewinian field approach (Gibson & Crooks, 1938/2020, p. 119). However, although this paper anticipates many central ideas and concepts of his ecological approach, it is a bit of an “anomaly” in Gibson’s oeuvre (Heft, 2001, p. 212)—he did not adopt this field perspective in his later work. Yet, Gibson fully embraced the systems approach that Koffka was moving towards. In his Principles, Koffka (1935/1999) took aim at the reflex theory, arguing that it cannot even account for the accommodation of the lens. And in his review of this book, Gibson (1971) applauded:

The reflex theory will not work. The action of the lens is an adjustment of a system, not a response to a stimulus. Blur of the retinal image is not a stimulus but a state. What the system does is to “hunt,” seeking an equilibrium, or what Koffka called a “maximum property”. … what Koffka glimpsed was a new formula, one that did not involve any stimuli or responses at all, but a “system” that tended toward “equilibrium”. Vague as it is this notion is promising and is becoming more explicit. (pp. 5–6)

Moreover, and as alluded to in previous paragraphs, Koffka made a strong plea for a naturalistic account of meaning, trying to overcome the divide between the natural and the human sciences. Although he was hesitant to emphasize meaning in his first lectures in the United States, aware of the different “intellectual climates” there and in Germany (Koffka, 1935/1999, p. 18), Koffka (1935/1999) was convinced that psychology as a science should have a place for meaning and significance (e.g., p. 19). And as we have seen, this was one of the reasons for focusing on molar rather than molecular behavior, and for dismissing the mechanistic framework. In like fashion, Gibson put meaning center stage in his ecological perspective. In one of the final interviews he gave, Gibson concluded: “I have been moving toward a psychology of values instead of a psychology of stimulus” (quoted in Reed, 1988, p. 296; see also Hodges & Baron, 1992). Ecological psychology is indeed conceived as an important attempt at a naturalistic theoretical and empirical study of meaning (see e.g., Reed, 1996).

And, perhaps most fundamentally, Koffka convinced Gibson that perception is among the fundamental issues in psychology (Reed, 1988). Indeed, Koffka’s (1935/1999) famous question of “why do things look as they do?” (p. 75) had deeply inspired Gibson. “Merely to ask this question was a brilliant achievement, for to ask it is to realize that it needs an answer. It provides a crucial test for any theory of visual perception, and surely will continue to do so” (Gibson, 1971, p. 4). And also in his final book, Gibson (1979/1986, p. 1) included Koffka’s question—explaining why things look as they do was in his view among psychology’s main tasks.

Gibson’s critique of Gestalt psychology

Although there are important parallels between Gibson’s perspective and that of Koffka, Gibson was at points very critical of Gestalt psychology. As mentioned in the introduction, he was unwilling to accept the Gestalt idea that the environment solicits actions, an idea that gains serious traction in contemporary ecological psychology (e.g., Dings, 2018, 2021; Heft, 2010; Käufer & Chemero, 2015; Rietveld & Kiverstein, 2014; Withagen, 2022). However, to understand Gibson’s reluctance, his overall criticism of Gestalt psychology needs to be taken into account. Gibson leveled his critique mainly at his colleague at Smith College. Among other things, Gibson took issue with the theory of perception that Koffka (1935/1999) forwarded in his Principles. Like nearly all theorists of perception since Descartes, Koffka argued that the stimulus information reaching the sense organs is impoverished. As Gibson (1960/2020b) complained:

Koffka . . . was so preoccupied with the ways in which the individual structured [their] stimulus field that he scarcely considered the ways in which it might already have structure. In fact, he wrote sometimes as if it had none, as if all the structure had to be imposed on it, because the stimuli themselves were meaningless points. (p. 341)

Indeed, at points, Koffka defended the assumption of “optical pointillism” (Lombardo, 1987/2017, p. 139). In his Principles, he stated:

the immediate cause of our vision of any object is just such a mosaic of stimulation as that of the photographic plate. And that raises at once the problem: how the enormous richness and variety of our visual behavioral environment can be aroused by such a mere mosaic of light and shade and color. . . . How can such rich effects arise out of such poor causes, for clearly the “dimensions” of our environmental field are far more numerous than those of the mosaic of the stimulation? (Koffka, 1935/1999, p. 75)

After evaluating several theories, Koffka (1935/1999) formulated “the true answer” (p. 98) to this question: “Things look as they do because of the field organization to which the proximal stimulus distribution gives rise” (p. 98). And in the 200 pages that followed this conclusion, Koffka laid out the already discovered organizational forces responsible for meaningful experiences, the Gestalten.

Already in his first encounters with Gestalt psychologists, Gibson was very critical of this perception theory. While listening to one of Koffka’s addresses, he scribbled “forces, forces, forces, crazy over . . . (the song of the Gestalt)” (as cited in Reed, 1988, p. 36). And 2 years before that, while attending one of Koffka’s seminars at Smith college for the first time, Gibson was even more harsh—“the Gestalt takes the place of the pineal gland” (as cited in Reed, 1988, p. 55)—alluding to Descartes’ mysterious account of how body and soul interact. In Gibson’s view, Koffka, like nearly all other theorists of perception, insufficiently recognized that the stimulus patterns that our perceptual systems pick up are already richly structured. This means, as Gibson (1950) appears to have realized as early as the writing of his first book (p. 25), that organizational processes in the brain, which are supposed to give rise to meaningful experiences, can be dispensed with (see also Gibson, 1971, p. 9).

When it comes to understanding Gibson’s aversion to the idea of demand characters, his critique of Gestalt psychology’s metatheory is arguably most important. Indeed, Gibson was seriously worried about the ontological position Koffka’s theory of perception was, in his view, rooted in. As with many other theories of perception, the Gestalt theory seems to entail a mind–world dualism (e.g., Heft, 2001; Lombardo, 1987/2017; Reed, 1988). Gibson centered his critique of this dualism primarily around the distinction that Koffka had made between the geographical and the behavioral/phenomenal world. This distinction is often illustrated with the German legend of the person riding across the frozen Lake of Constance while experiencing traversing a snowy plain. Koffka (1935/1999) used this legend in his Principles (pp. 28–29), and also Gibson (1971) commented upon Koffka’s analysis of this legend in his critique. Yet, another illustration of these two different environments that Koffka included a few pages later in his landmark book might be more insightful.

Koffka (1935/1999) referred here to a study of Révész in which chicks were trained to peck from the smaller of two figures that were presented simultaneously. Critically, after successfully training the chicks using a variety of forms and compositions, Révész exposed the animals to the Jastrow illusion—two identical forms that looked different in size to us due to their positioning on top of one another. It appeared that chicks are susceptible to this illusion as well, as their pecking behavior illustrated. In view of these results, Koffka (1935/1999) wondered:

Why do the animals choose one of the two equal figures, when they have been trained to choose the smaller one? Described in these geographical terms their behavior seems quite unintelligible . . . . But everything becomes perfectly plain and simple if we answer our question as every unbiased [layperson] would answer it, by saying: The animals chose one of the two equal figures because it looked to them smaller; just as it looks smaller to us. Or, in our terminology, the behavioral environment in the critical experiments was similar to the behavioral environment in the training experiments inasmuch as it too contained a larger and a smaller figure, although the critical geographical environment contained two figures of equal size. (pp. 32–33)

Hence although Koffka (1935/1999) asserted that animals behave in a geographical environment, their behavior is regulated by the behavioral, phenomenal environment (p. 31). But this position was unacceptable for Gibson (1971),

To say that behavior occurs “in” a phenomenal world leads to difficulties. If this means that each animal and every [person] behaves in [their] own private world, as it can be taken to mean, it is surely mischievous. Koffka did not go so far, but he nevertheless persisted in believing that there are two distinct meanings of the basic notion of an organism-in-an-environment. With two kinds of environment, behavioral and geographical, he could hardly avoid muddle and confusion. (p. 7)

There is quite some debate about Koffka’s ideas about the behavioral environment, and where he believed it resides (e.g., Epstein & Hatfield, 1994; Heft, 2001; Kiverstein et al., 2021; see also below), but he was at points rather unequivocal about it. Indeed, he placed the behavioral environment within the real organism in his schematic overview of behavior and the environment (Koffka, 1935/1999, p. 40), a position which is in line with both his earlier discussed theory of perception and the allied ideas of isomorphism that Koffka adopted from Köhler. As he approvingly cited his German colleague, “[a]ny actual consciousness is in every case not only blindly coupled to its corresponding psychophysical processes, but is akin to it in essential structural properties” (Köhler, 1920, p. 193; quoted in Koffka, 1935/1999, p. 62). This indeed indicates that Koffka defended a mind–world dualism—the phenomenal, behavioral world resides in the head.

This dualism that Koffka’s perspective was grounded in was arguably the main reason why Gibson was so reluctant to adopt Koffka’s idea of demand character. Koffka took this idea from Lewin (1926/1999), who had earlier argued that the animal’s lived space consists of valences—they “press towards definite actions” (p. 95). The chair may invite sitting, the apple solicits eating, the book reading, and on and on. Whether these objects are endowed with a demand character depends, Koffka asserted, on the needs and motives of the agent. “I have a need which for the moment cannot be satisfied; then an object appears in my field which may serve to relieve that tension, and then this object becomes endowed with a demand character” (Koffka, 1935/1999, p. 354). And this very idea of demand character was problematic according to Gibson. As he put it:

[Koffka] maintained that the postbox “invites” the mailing of a letter; the handle “wants to be grasped”; the chocolate “wants” to be eaten; things in experience “tell us what to do with them”. But the crux of this theory is that the demand character, like the valence, was assumed to be in the phenomenal object but not in the physical object. It was in the “field” for Lewin or, for Koffka, in what he called the “behavioral” environment but not in the “geographical” environment. In short, the value of something did not have any “physical” reality. The valence of an object was bestowed upon it by a need of the observer, and a corresponding tension in the field. Koffka agreed, arguing that the postbox has a demand-character only when the observer needs to mail a letter, for only then [are they] attracted to it. (Gibson, 1972/2020c, p. 409)

In Gibson’s (1972/2020c) view, Gestalt psychology, although trying to develop a naturalistic account of meaning, “could not resolve the subjective-objective dichotomy” (p. 410). And a few years later, in his final book, Gibson (1979/1986) attempted to deal it a death blow:

What explanation could be given for these valences, the characters of objects that invited or demanded behavior? No one, not even the gestalt theorists, could think of them as physical and, indeed, they do not fall within the province of ordinary physics. They must therefore be phenomenal, given the assumption of dualism. If there were two objects, and if the valence could not belong to the physical object, it must belong to the phenomenal object. (p. 138)

And it is this pernicious mind–world dualism, this subject–object dichotomy, that Gibson tried to overcome in his ecological approach.

Gibson’s ecological alternative

One way in which Gibson tried to resolve this dualism is through the introduction of the concept of affordances. Gibson asserted that the environment human and nonhuman animals live in consists of possibilities for action. In developing the concept of affordances, he acknowledged that he was very much inspired by both Koffka and Lewin. Yet, unlike their notion of demand character, the concept of affordances, Gibson asserted, overcomes the mind–world dualism that has gripped psychology since the 17th century.

An important fact about the affordances of the environment is that they are in a sense objective, real, and physical, unlike values and meanings, which are often supposed to be subjective, phenomenal, and mental. But, actually, an affordance is neither an objective property nor a subjective property; or it is both if you like. An affordance cuts across the dichotomy of subjective-objective and helps to understand its inadequacy [emphasis added]. (Gibson, 1979/1986, p. 129)

Whether stairs afford climbing depends on the height of the stairs relative to the leg length of the agent (e.g., Warren, 1984). Stairs that are climbable for an adult might not be for a toddler. Hence, affordances are relational in nature. Although Gibson recognized a physical world, he asserted that animals perceive and regulate their behavior with respect to an ecological environment, replete with affordances. And he left no doubt that this concept was fundamentally different from the Gestalt idea of demand character:

The affordances of the environment are permanent, although they do refer to animals and are species-specific. The positive and negative valences of things that change when the internal state of the observer changes are temporary. The perception of what something affords should not be confused with the “coloring” of experiences by needs and motives. Tastes and preferences fluctuate. Something that looks good today may look bad tomorrow but what it actually offers the observer will be the same. (Gibson, 1972/2020c, p. 410)

Hence, whereas the concepts of valence, demand character, or behavioral environment were introduced to describe our experiencing of the world, the concept of affordances attempts to capture the ecological facts. And although these facts are animal-relative, they exist independent of experience. Water is drinkable for me whether I perceive it or not.

To account for our perceiving of these affordances, Gibson developed a theory of perception that is also at variance with the Gestalt theory. In fact, Gibson’s perspective was almost a direct response to it. Referring to a fight with Koffka about Mary Henle’s master thesis at Smith College, Gibson wrote in a letter: “The muddy issue of organization [versus] association . . . stuck in my mind for years and years. This old personal controversy had a lot to do with my developing the theory of the external structure of the optic array” (as cited in Reed, 1988, p. 94).

In the 1960s, Gibson (1961) developed his ecological optics, stressing that the ambient arrays are highly structured and contain patterns that specify the environment and the animal’s relation to it. This means, Gibson asserted, that organizational processes in the brain can be dispensed with. All the information needed for perceiving is already available in the ambient arrays. Animals simply have to pick up this information, allowing for a direct perceptual contact with the environment.

If the retinal image is a collection of stimuli one kind of brain process is required to explain perception, but if it is an array of contrasts quite another sort is needed. A structured array of light on the retina does not require that its elements be integrated and segregated for they already are. (Gibson, 1971, p. 9; see also Gibson, 1950, p. 25)

In recent decades, Gibson’s concepts of affordances, ambient arrays, and information as specification have given rise to an empirical research program that provides us with further insight into how animals perceive and act in their environments (see e.g., Blau & Wagman, 2023; Bruineberg et al., 2023; Chemero, 2009; Turvey, 2019; Warren, 2006). But while Gibson may have succeeded in overcoming the dualism of Koffka’s Gestalt psychology, his ecological framework leaves many essential aspects of animal life underexposed. Granted, the inherent activity and growth of an animal is emphasized. But the affective component, although acknowledged (Gibson, 1972/2020c, p. 410), is consigned to the margins. Indeed, in Gibson’s framework the animal is mainly pictured as a being surrounded by possibilities for action it can intentionally select and actualize, almost devoid of affect and emotion (Withagen, 2022). And then something essential about animal life is lost.

Lessons to take from Koffka’s Principles

Although Gibson formulated a severe critique of Koffka’s framework, several later authors have attempted to equate ecological ideas with insights from Koffka. For example, Epstein and Hatfield (1994), although aware that Gibson would “strenuously resist” (p. 174) their construal, claimed:

We cannot shake the impression that the “world of ecological reality” is largely coextensive with the world of phenomenal reality, and that the description of ecological reality, although couched in the language of “ecological physics”, nonetheless is an exercise in phenomenology . . . . Gibson’s distinction between ecological reality and physical reality parallels the Gestalt distinction between the behavioral environment and geographical environment. (Epstein & Hatfield, 1994, p. 174)

And in their paper on Koffka’s seminal book, Kiverstein et al. (2021), by contrast, map their distinction between the landscape and the field of affordances onto Koffka’s distinction between the geographical and the behavioral environment. On their reading of Koffka, the geographical environment amounts to the “publicly available shared environment” (p. 2282), which shows parallels with their notion of the landscape of affordances—the manifold of possibilities for action that are available to a “form of life” (Rietveld & Kiverstein, 2014). Koffka’s behavioral environment, on the other hand, is similar to the field of affordances: the individual’s lived environment in which several affordances stand out at a certain moment in time and invite action.

Although I am on board with several of these authors’ ideas, I believe that their claims about the similarities between Gestalt and ecological ideas are based on a misreading of Koffka and/or Gibson. As we have seen, Gibson introduced the concept of affordances as an alternative to the notion of demand character, stressing the relatively permanent ecological facts in an animal’s environment. Hence, contrary to Epstein and Hatfield (1994), Gibson’s ecological reality is not identical (or even similar) to Koffka’s behavioral environment. And contrary to Kiverstein et al. (2021), Koffka’s geographical environment is not a “publicly available shared environment” (Kiverstein et al., 2021, p. 2282). Recall the experiment by Révész that Koffka (1935/1999) used to illustrate his distinction between the behavioral and the geographical environment. Trapped by the Jastrow illusion, “the two equal figures” (Koffka, 1935/1999, p. 32) that Koffka defined as the geographical environment are in a crucial sense not “publicly available.” Hence, and in line with Epstein and Hatfield’s (1994) observation, Koffka’s geographical environment is more similar to Gibson’s physical world.

These remarks aside, I believe that these authors are right in taking Koffka’s concept of demand character seriously in their ecological theorizing (see also Dings, 2018, 2021; Heft, 2010; Käufer & Chemero, 2015; Withagen, 2022, 2023). True, and as Gibson (1979/1986) emphasized, Koffka could not think of demand characters as “physical” (p. 138). But Gibson himself could not think of them as ecological, and thus he relegated them to the fleeting realm of experience, and introduced affordances as a genuine ecological concept to describe the animal’s environment. However, as recent theoretical research has shown, demand characters or invitations can be thought of as ecological (e.g., Rietveld & Kiverstein, 2014; Withagen, 2022, 2023). They do not require an inner phenomenal theater in which they reside. Rather, like affordances, invitations are relational in nature. Whether an object attracts or repels an agent depends on the object relative to the agent who is in perceptual touch with it. And it is the very object itself, not some pale representation thereof, that is affecting the agent in question.

Incorporating the idea of demand character in the ecological movement has, in my view, significantly enriched the perspective. From its very inception, the concept of affordances has been severely criticized for insufficiently accounting for our behavior. After listing almost 30 possibilities for action that a single piece of paper offers us (e.g., making paper dolls, writing sonnets, cutting it in pieces) and arguing that the list could be easily extended, Cutting (1982) concluded:

My behavior is virtually unconstrained by its affordances. To be sure, it does not afford flying to Baghdad upon, but the exclusion of a large domain of behaviors does not diminish the fact that an infinity remain. To apply to adult human beings, it would seem that the theory of affordances needs full-blown theories of personality and of choice. (p. 216)

And more recently, Ratcliffe (2015) argued:

Things do not simply “afford” activities; they appear significant to us in all sorts of different ways. It is not helpful to say that a bull affords running away from, while a cream cake affords eating. What is needed … are distinctions between the many ways in which things appear significant to us and, in some cases, solicit activity. (p. 61)

The concept of invitations remedies these shortcomings. It captures in what way things are significant to us—it describes what the environment does to the agent. In addition, as the environment can invite behavior, “full-blown theories of personality and of choice” (Cutting, 1982, p. 216) are not needed to account for many of our actions. Indeed, the agent can simply give in to the environmental demands. As Dreyfus and Kelly (2007) put it, “[i]n backing away from the ‘close talker’, in stepping skillfully over the obstacle, in reaching ‘automatically’ for the proffered handshake, we find ourselves acting in definite ways without ever having decided to do so” (p. 52). This is not to say that all of our actions are the result of giving in to invitations—we can of course also reflect upon a situation and decide what to do. But as phenomenologists have stressed, the unreflective responding to the environmental calls is arguably our primary mode of being-in-the-world (e.g., Dreyfus, 2005/2014; Dreyfus & Kelly, 2007).

There is however another lesson from Koffka’s (1935/1999) Principles that, to my mind, has insufficiently been taken to heart in the ecological movement. Although the concept of affordances is indeed powerful, several recent ecologically inclined authors sometimes write as if the animal’s environment consists exclusively of affordances. It is important to stress that Gibson himself did not defend this position. Although he centralized affordances in his later work, arguing that we perceive the environment primarily in terms of action possibilities, he acknowledged the fact that we can also perceive objects, events, surfaces, and pictures as such (see Withagen, 2023). Yet, there is a recent tendency to couch all of our experiences and actions in terms of perceiving and actualizing affordances (e.g., García, 2023; Krueger & Colombetti, 2018; Peter, 2023; Rietveld & Kiverstein, 2014; but for critiques, see Segundo-Ortin & Heras-Escribano, 2023; Withagen, 2023; Withagen & Costall, 2022). In their landmark paper, Rietveld and Kiverstein (2014), for example, stretched the concept of affordances to accommodate color perception. “We can say that things afford perceiving what they really are. The letters appearing on our laptop screen as we prepare a PowerPoint-presentation, for instance, provide the possibility to judge correctly that their color is red” (p. 345). And more recently, they argued that the experiencing of art can be understood in terms of (inviting) affordances: “the phenomenology of an entire field of relevant affordances does have the kind of depth and complexity that can do justice to the experience of layered meaning works of art embody” (Rietveld & Kiverstein, 2022, p. 590).

Koffka, on the other hand, was not in favor of capturing all of our experiences of the environment in one single concept and actually made a strong plea for adopting a variety of notions. After describing Köhler’s experiment on how chimpanzees respond to a “Singhalese demon mask,” Koffka (1935/1999) wrote:

We conclude that objects in the field may possess characters which can be expressed in terms neither of shape and color nor of practical use, and which are apt to exert a powerful influence on our behavior. These characters are for us most pronounced in human forms, but may belong to almost any object. Our preoccupation with practical use and scientifically classifiable properties has robbed our world of a good many of them [emphasis added]. (p. 359) ²

Koffka (1935/1999) praised artists and poets for their ability to disclose these characters to us, free as they are “from the craving for efficiency” (p. 359). Hence, alongside colors, shapes, and functional properties, Koffka also talked about, for example, the aesthetic qualities of objects and their physiognomic properties (Withagen & Costall, 2022). Persons but also objects have emotional characteristics. A car can look friendly or aggressive; a person angry, sad, proud, or full of joy. And Koffka (1935/1999) was quick to point out that it was not “adequate” to “call these characters demand characters” (p. 360). Indeed, as Lambie (2020), building upon Koffka, argued, an angry face can prompt a variety of reactions in different people. That is, what it demands or invites is not equivalent to what it expresses.

This shows that arguably a variety of concepts is needed to do justice to the enormous richness of our experiencing the world. Certainly, possibilities for action matter a great deal in our everyday practices, but there is definitely more to our lives. We cannot only see that we can pick up a cup and drink from it, but can also care about which cup we drink from during our morning ritual, or appreciate a cup’s color, beauty, or historical significance (as in a museum). And rather than couching all of these experiences in terms of a single concept (e.g., affordances: the cup affords perceiving red, the cup affords aesthetic satisfaction), as some current ecological psychologists tend to do, ³ Koffka taught us that we will probably do more justice to our lived experience if we have multiple concepts, each of which captures others aspects of our experiencing the world.

Concluding remarks

In this article, I have examined why Gibson was unwilling to accept the Gestalt idea that the environment calls for certain actions. We have seen that although Gibson was deeply influenced by Koffka, he was also seriously worried about traces of a mind–world dualism he found in Koffka’s work, including in the concept of demand character. But in distancing himself from Koffka, and overturning many Gestalt ideas, Gibson arguably threw the baby out with the bath water. Granted, it is important for psychology to free itself from the shackles of the subjective–objective framework, but psychology also has the task of taking all of our experiences seriously. And the current ecological movement can learn something from Koffka here. Although some recent ecologically inclined authors have now adopted an ecological construal of demand characters (i.e., inviting affordances), several of them are reluctant to follow Koffka’s plea for using a variety of concepts to describe our lived experiences—everything is understood in terms of affordances. But this strategy runs the risk of overlooking (or even explaining away) some essential aspects of experience, a risk ecological psychology cannot take. As an approach that prides itself on studying the real organism in its meaningful environment, it has the task of doing justice to the enormous richness of our lived environment.