The intention to use a specific affordance: A conceptual framework for psychology

Edward S. Reed

.... [p.46]

....

Gibson, Perception, and Cognition

The great potential of James Gibson's ecological psychology for improving our understanding of cognition has been obscured by the rather vitriolic debate surrounding his ecological theory of perception (Cutting, 1986; Fodor & Pylyshyn, 198; Hagen, 1985; Reed, 1988a; Turvey, Shaw, Reed, & Mace, 1981; Ullman, 1980). The greater proportion of this debate has been fueled by a misapprehension: Proponents and opponents of Gibson's views alike have assumed that perception, at least in Gibson's sense of the term, should be contrasted with cognition. But this is a false contrast. According to Gibson, perception is one kind -the most basic kind- of cognition (J. Gibson, 1966, p. 91; Reed, 1987, 1988a, p. 298f.; Shaw & Bransford, 1977).

Throughout this chapter, cognition is used to refer to any and all psychological processes that function to give an organism knowledge about its environment, and its situation within the environment. By knowledge I mean any functional awareness of aspects of the environment as they exist (perception), or as they have existed (memory), or as they should come to [p. 47] exist (insight and anticipation), or even as they ought to exist (planning). By functional awareness, I mean any capacity to guide action and thought, whether conscious or not. The distinctions among these different types of cognition are not sharp -indeed, it is notoriously difficult to demarcate certain kinds of perception, memory, and anticipation (Bartlett, 1932; J. Gibson, 1966, chapter 13)....

According to the present perspective, perception is cognitive because it yields knowledge, not because it is based on information-processing mechanisms, as much of contemporary cognitive psychology holds. Perception is a basic cognitive function because it yields knowledge of the psychologically most basic aspects of our environment -what Gibson called the affordances of the the environment. Other modes of cognition such as those involving the use of gestures, symbols (linguistic or numerical), or depictions can bring us knowledge of other aspects of the environment, such as abstract properties (color, metrical shape, cultural significances). J. Gibson (1966, p. 282; Reed, 1988a, p. 306f) rightly argued that these other modes of cognition have the very important effect of reorganizing and securing the knowledge gained via perception, and he explicitly held that most adult human apprehension involved combinations of several modes of cognition.

Once the false antagonism between perception and cognition is removed, it becomes possible to extend Gibson's description of the environment to be perceived so as to develop a general description of the environment to be cognized, acted upon, and shared. Such a project is a major undertaking..., but a necessary one. Ever since its inception, psychology has run afoul of the problem of not having an adequate account of the rich environmental structures and event in which all psychological development occurs. Behaviorists ludicrously try to reduce this environment to "stimuli" (see J. Gibson, 1960/1982). That obviously does not work. More recently, information-processing psychologists have tried to solve the problem by pretending it does not exist, and allowing all manner of items -from electric shocks to stories- to be called stimuli, focusing instead on attempts to model the knowledge structure supposedly built up from experience.

This tendency in contemporary psychology -to be concerned with the knowledge structure an individual builds up on the basis of experience with one sort of stimuli or another- is perhaps an improvement over behaviorist concerns. Nevertheless, if one is ever to evaluate the cognitive status of this [p. 48] learning, one still needs to have an independent account of what a subject has learned. Such an account can and should combine biological and cultural realism, but it must not presuppose aspects of individuals' experiences in describing what it is that people learn, at risk of circularity or vacuousness. The environment is so rich that diametrically opposed conceptions of knowledge structure -such as analog versus propositional representations- have been posited of single cognitive systems (Anderson, 1978). Worse, it is now becoming apparent that any given cognitive system can also be modeled by a number of neural networks, making comparative testing of cognitive models even less precise. Cognitive psychology must concern itself with the ecology of human behavior and knowledge, or risk being buried beneath the weight of theories that are top-heavy with untestable assumptions.

THE ECOLOGY OF PERCEIVING, ACTING, AND KNOWING

.... [p. 55]

....

INFORMATION IN THE ENVIRONMENT

.... [p. 61]

....

INTENTIONS AND THE SELECTION OF AFFORDANCES

What are Intentions?

Intention is typically used to refer to a mental state.... But, ... this mental state theory of intention has a number of problematic methodological results.... [p. 62] ....

In the present view, intentional action is a prime example of just how "spread out" cognition is. To have an intention requires objects as well as subjects (Fischer et al., chapter 4, this volume). And both objects and subjects must be situated in a setting (Bronfenbrenner, chapter 1, this volume; Heft, 1988, 1989). Intentions are thus not discrete, static, internal, mental events. They are continuous, dynamic, contextualized, public events. To identify an intention -the entire complex organization of action and knowledge- with its earliest mental components is as grave a fallacy as identifying a hurricane with its first gales.

From an ecological point of view, intentions are not causes of action, but patterns of organization of action; they are not mental as opposed to physical, but are instead embodied in the kinds of performances most likely found in cognitively capable creatures. Intention is characterized by directness, persistence, and resilience: directness toward objects, places, or events relevant to the intender's situation; persistence until the intention is met, and recognized as having been met by the agent; resilience in the face of perturbation, whether environmental or personal. The development of intention is thus the development of the ability to nest bouts of exploratory and performatory behavior so as to achieve desired outcomes. The purely cognitive ability to think of things or to plan activities is a component of intention, but only one small component of a much larger, more complex process.

Intention is a phenomenon that only emerges when a complex organism finds itself in a complex environment. Roughly speaking, intention can emerge (but may not) whenever there is a real possibility of choice among affordances.... Young children show [p. 63] evidence of intentionality: They can understand situations, plan, and act up to a point. But the resilience and persistence required for successful voluntary action in a broad range of contexts is missing in young children, and develops only slowly.

.... [p. 66]

....

The Process of Affordance Selection

Component Processes. The component processes underlying all affordance use can be be broken down into two kinds of activity, exploratory and performatory (E. Gibson, 1988; Reed, 1982). Perception in particular, and cognition in general, cannot occur except as the result of exploratory activity, in which the goal is the detection and use of available information about affordances. Even the most passive of thinkers, lost in a reverie is thinking about things at least some of which were originally discovered through exploratory activities. (The recent emphasis of cognitivists on the centrality of analogies of experience in thinking is very much in the spirit of the present theory. See especially Johnson, 1988). Performatory activity means any kind of action in which some affordance of the environment (and this includes the social environment) is altered.... These two modes of activity are complimentary, and when they work together as they normally do, I speak of a PAC [perception action cycle]. To become aware of affordances requires exploration of information, to use an affordance, requires performance, and this includes the regulation of performance via both information related to the nature of the task to be achieved, and previously acquired experience and knowledge.

The basic units of action on which psychological developmental selection can operate are... typically comprised of one or more PACs. ... [And] skilled cognition operates on several levels simultaneously....

Principle of Variation. Because the environment is extremely complex, it is inevitable that a large set of varying PACs is capable of accomplishing the same exploratory and performatory goals.... [p. 67]... That the same goals can be accomplished by diverse means is a fundamental fact of cognition and action, at any level of analysis.

Just as a given goal cannot imply a single action pattern, so a given situation is not a single, univocal context. On the contrary, one and the same situation or object may afford a considerable diversity of actions. The stick that affords poking holes in the ground for planting may also afford raking-in fruit from a high tree branch.... There is so much information available in any given situation that not all of it can be detected and used. And, similarly, there are so many possibilities for action in a single situation that not all of them can be detected and used....

Principle of Selection. With such a rich and complex environment, what is to keep anyone from attempting to use every affordance she detects? Or... from detecting a seemingly endless series of affordances, without settling [on a given course of action]? ....

The hypothesis that I have been evaluating and testing is that every intention selects information and affordances relevant to it out of the complex ... circumstances in which any animal finds itself..... Note,... that the affordances are not created by the intention, just attended to because of it. This selection process brings attention increasingly under the control of the subject's intentions, as E. Gibson and N. Rader (1979) argued. [p. 68]

A child who has little experience with a given task, or who finds him or herself interested in several tasks within a given setting, will show ... less stable action ... than a more "single-minded" child. But what allows the development of this single-mindedness? ....

.... Vygotsky and Luria argued that internalizable symbols (like words) were a necessary precursor to the development of voluntary, knowledge-based, activity. The surprisingly sophisticated cognitive and action skills of preverbal infants would seem to cast doubt on this idea....

The present argument would seem to be in line with Gopnik and Meltzoff's (1986, 1987) specificity hypothesis. These writers argue that it is not general cognitive skills that are associated with the verbal "take-off" around 18 months. Instead, they argue, the take-off emerges from a small set of specific cognitive skills that enable the child to comprehend the meanings of key words and phrases (and vice versa). For example, nouns and category terms are associated specifically with object knowledge; and certain verbs ... are associated with the child's increasingly rich understanding of events and causality. In this regard, it is significant that the kinds of utterances produced in private speech also appear to reflect some cognitive-functional specialization. Terms of self-regulation, self-description, and expressive content are more likely to appear in a 2-year-old's self-directed speech than in his other-directed utterances (Furrow, 1984).

Intentions. On the present view, intentions are spread out across mind, body, information, ecological context, and social setting. Development of [p. 69] the ability to act intentionally requires not only knowledge of how the self should perform in a situation to achieve a goal, but also the ability to regulate cognition and action so that only the intended affordances are attended to and utilized. This regulation involves organizing one's activity so that most or all of the unit PACs are "on task"....

Joint Intention. Most human learning occurs in an interactive, acculturating context. The cognition we use most in daily life thus only emerges in a system in which there exists, at the very least, a scaffolder as well as a learner....

.... Through his or her action the scaffolder establishes boundary conditions, which is to say, a set of criteria for what will count as satisfactory performance....

The scaffolder in these situations thus acts as part of the selection process affecting the learner, and the attentive scaffolder finds his or her own intentions subject to selection pressures from the learner. .... What matters is not that the learner and scaffolder necessarily [p. 70] share identical intentions, but that the two are sufficiently close, and that the direction of change is appropriate. The field of promoted action [FPA] thus feeds into cognitive development in several ways: by constraining, in increasingly specific ways, what affordances are to be attended to and used; by encouraging one set of intentionally guided PACs, and discouraging others; by creating subsidiary environments in which certain kinds of repetitions (i.e., repetitions of PACs within certain tolerances of variation) are emphasized.

Fields of Action. As a human child develops, there is an inherently social process of encouraging the formation and carrying out of certain intentions, and discouraging the formation and execution of others. This social structuring involves alterations of the child's environment, and the information available to the child within that environment, as well as alterations of the child's behavior. The field of promoted action (FPA), as described earlier, encompasses those affordances (natural or social) to which a child's attention and activities are directed by others. There is also a field of free action (FFA) (again, these are modifications of concepts of Valsiner's, 1987, chapter 4). The FFA encompasses those affordances and activities that the individual is both capable of accomplishing on his or her own, and is allowed by social circumstances to do so.

In cognitive development, there is an overall increase in the FFA, along with a progressive refinement and sophistication within the FPA. Cognitive development does not simply mean that the individual can do more things (although that is certainly an important part of cognitive growth); it also means that the individual can understand progressively subtler and more complicated instructions and ideas. In Valsiner's (1987) case studies of the development of mealtime skills, for instance, it is shown that the FFA is typically highly restricted at mealtime for middle-class children in the United States. (There is even a technology for accomplishing this restriction, the "high chair" and the newer "sassy seat.") But this restriction on the FFA would seem to be designed to facilitate a focusing of instruction, so that discrete and episodic intervention by caretakers concerning what to eat and how to do so will have a significant effect on the child. For the same child at "free play" there might be a considerably expanded FFA, but also a less articulated FPA. For example, banging is an exploratory activity favored by 6-to-12-month-old infants specifically for hard objects (Palmer, 1989). The same child might be free to bang a spoon at playtime, or even encouraged to do so, but not allowed to do so (i.e., the spoon may be taken away) at mealtime.

This subtle relationship between the FPA and the FFA determine what Vygotsky dubbed the zone of proximal development (ZPD). It would seem, however, that Vygotsky viewed the ZPD as a general phenomenon, [p. 71] reflecting the general difference between a child's individual cognitive function and that same child's ability to function with instruction. From an ecological point of view, such a notion of the ZPD as a general range is not viable. The ZPD does not even emerge except in highly channeled and specialized environments. The causes of its emergence will differ for different skills, and will be strongly affected by ecological, cultural, and individual factors. Thus, I argue that there is no general ZPD, only specific ZPDs for specific skills, reflecting particular relationships that emerge between FPAs and FFAs. The only sense in which a general ZPD can be said to exist is in terms of a set of specific, culturally determined developmental tasks which tend to be assigned in each child's FPA, and reflected in the growth of their FFAs.

To summarize, the modified concept of the ZPD, as used here, emerges from the new concepts of FFA and FPA. These are defined as Follows:

Field of promoted action (FPA): Those objects, places, events, and their affordances, which an individual is encouraged to realize. This field changes from object to object, place to place, event to even, but resistant and invariant patterns of encouragement are hypothesized to exist for given individuals, families, and other cultural groupings.

Field of free action (FFA): Those objects, places, events, and their affordances that an individual can realize on his or her own, and that the individual is allowed to use. This field expands and contracts reciprocally to the field of promoted action.

Zone of proximal (potential) development (ZPD): For any given skill, the difference between the FPA and the FFA. This represents the range of affordances and the variety of contexts within which social scaffolding facilitates affordance usage.

CONCLUSION

Cognition, like all forms of animate activity, begins with use values, with the affordances of the environment for action. At the outset, then, cognition is a specific relation between organism and environment (Holt, 1915). This relation is made possible by the existence of ecological information specifying the situation of an active observer within his or her surroundings (J. Gibson, 1966, 1979), and by each organism's capacity to detect that information, and to develop improved processes of detection (E. Gibson, 1969, 1984). The complexity of the environment is such that the specific relations of affordance use inevitably come into conflict, and successful organisms must develop processes of selection whereby the [p. 72] pattern and sequence of their affordance use is made to be appropriate for their current situation.

In social animals, all these processes, from the simplest act of affordance detection, to the most sophisticated patterning of choice, are mediated in part through interactions as well as actions. Further, social animals' actions and interactions are themselves mediated in part by information that has been selected and coded by other social animals. Not only do attention and performance become increasingly selective and attuned to the environment and information around us, they become increasingly acculturated as well. We make ourselves.... Our cognitive skills do not determine our social reality, nor does our social reality determine our cognition. Rather, our social mode of existence is a kind of life within which cognitive skills emerge, develop, and are refined. And, at bottom, all human modes of life, however cognitively or socially refined, require that nature provide us with the means of life....

The task of ecological developmental psychology is to help us understand the intricate structure of affordance use in acculturated settings that characterize all human behavior. The theory proposed here is primarily descriptive. My hope is that the concepts described here -affordances, information, the fields of promoted and free action, the concept of selective retention among general affordances to yield task specificity- are both rich enough and concrete enough to enable empirical tests of real, everyday behavior.

REFERENCES

Gibson, E. (1969). Principles of Perceptual learning and development. New York: Appleton-Century-Crofts.

Gibson, E. (1984). Perceptual development from the ecological approach. Advances in Developmental Psychology, 3, 243-286.

Gibson, J.J. (1966). The Senses Considered as Perceptual Systems. Boston: Houghton Mifflin.

Introduction (pp. 1-6); "Chapter XIII: The Theory of Information Pickup" (pp. 266-286); "Chapter XIV: The Causes of Deficient Perception" (pp. 287-318).

Gibson, J.J. (1979). The Ecological Approach to Visual Perception. Boston: Houghton Mifflin.

Gibson, J.J. (1982). Reasons for Realism: Selected essays of James J. Gibson, E. Reed & R. Jones (Eds.). Hillsdale, NJ: Lawrence Erlbaum.

Reed, E.S. (1987). James Gibson's Ecological Approach to Cognition (pp. 142-173). In A. Costall & A. Still (Eds.). Cognitive Psychology in Question. Sussex: Harvester Press.

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

Rogoff, B. & Lave, J. (1984). Everyday cognition: Its development in a social context. Cambridge, MA: Harvard University Press.

Valsiner, J. (1987). Culture and the development of cognition. New York: Academic Press.

Vygotsky, L. (1978). Mind in Society. (Trans. M. Cole). Cambridge, MA: Harvard University Press.

Vygotsky, L. (1987). The Collected works of L.S. Vygotsky (R.W. Rieber, A.S. Carton, Eds.). New York: Plenum.