JOURNAL OF THE EXPERIMENTAL ANALYSIS OF BEHAVIOR 1984, 41, 109-115 NUMBER I (JANUARY)

ARE THEORIES OF PERCEPTION NECESSARY?

A REVIEW OF GIBSON'S THE
ECOLOGICAL APPROACH TO VISUAL PERCEPTION
A. P. COSTALL
UNIVERSITY OF SOUTHAMPTON, U.K.

Representational theories of perception postulate an isolated and automonous "subject"

set apart from its real environment, and then go on to invoke processes of mental repre-
sentation, construction, or hypothesizing to explain how perception can nevertheless take
place. Although James Gibson's most conspicuous contribution has been to challenge rep-
resentational theory, his ultimate concern was the cognitivism which now prevails in psy-
chology. He was convinced that the so-called cognitive revolution merely perpetuates, and
even promotes, many of psychology's oldest mistakes. This review article considers Gibson's
final statement of his "ecological" alternative to cognitivism (Gibson, 1979). It is intended
not as a complete account of Gibson's alternative, however, but primarily as an apprecia-
tion of his critical contribution. Gibson's sustained attempt to counter representational
theory served not only to reveal the variety of arguments used in support of this theory,
but also to expose the questionable metaphysical assumptions upon which they rest. In
concentrating upon Gibson's criticisms of representational theory, therefore, this paper
aims to emphasize the point of his alternative scheme and to explain some of the important
concerns shared by Gibson's ecological approach and operant psychology.

My title, "Are theories of perception neces- havior is subject to lawful description in its
sary?," makes an obvious reference to Skinner's own right without appeal to "underlying"
paper on theories of learning (Skinner, 1950). structures, be they mental, neurological, or
But it is also based upon the following passage quasi-neurological (Gibson, 1966, chapter 13;
from James Gibson's (1966) book, The Senses Skinner, 1938, pp. 3-5; 1969, pp. vii-xii). Gib-
Considered as Perceptual Systems: son's recent book, The Ecological Approach to
Visual Perception, presents his final views on
When the senses are considered as percep- this matter (Gibson, 1979; see especially chap-
tual systems, all theories of perception be- ter 14).
come at one stroke unnecessary. It is no Although cognitive psychologists like to de-
longer a question of how the mind oper- fine their approach largely by contrast with
ates on the deliverances of sense, or how what they consider behaviorism, the cognitive
past experience can organize the data, or approach can nevertheless be characterized by
even how the brain can process the inputs its habitual appeal to internal, "mental" rules
of the nerves, but simply how information and representations, which it treats as exclu-
is picked up. (p. 319) sive and primitive, explanatory terms. Skinner
has presented some valuable criticism of the
The point of this double reference is that cognitivist program-for example, in his paper
not only Skinner but also Gibson rejected the in Behaviorism (Skinner, 1977a). But the effec-
kind of "theory" which is now so enthusiasti- tiveness of his challenge has been limited not
cally promoted within cognitive psychology. only by his contentious style, but also by his
The intention of their rejection was not, it stereotyped role as the villain in cognitivist
should be stressed, a denial of any role for the- melodrama. The problem is compounded by a
ory in psychology, but an insistence that be- troublesome ambiguity about much of his crit-
icism. He keeps shifting the grounds of his at-
A version of this paper was presented at the First tack so that sometimes he seems to deny the
European Meeting on the Experimental Analysis of Be- reality of the mental structure invoked by cog-
haviour, Liege, Belgium, July 26-30, 1983. Requests for nitivism, while at other times he appears
reprints should be sent to A. P. Costall, Department of
Psychology, The University, Southampton S09 5NH merely to question their heuristic value in gen-
U.K. erating research. Increasingly, his arguments
109
110 A. P. COSTALL
are pragmatic, concerned with the effectiveness trine with such evident enthusiasm that he not
of particular kinds of analysis for the purposes only lets slip its ultimate absurdities, but actu-
of control. As a result of this ambiguity, cog- ally seems to relish them:
nitive psychologists have felt free to disregard
his metaphysical criticisms and insist that Perceptions are constructed, by complex
choice of means and ends in science is merely brain processes, from fleeting fragmentary
a matter of taste. Indeed, cognitive psycholo- scraps of data signalled by the senses and
gists have felt free to enter the confines of the drawn from the brain's memory banks-
experimental analysis of behavior movement themselves snippets from the past. On this
itself, and even to own up to their alien al- view, normal everyday perceptions are not
legiances. part of-or so directly related to-the
Cognitivism is under real threat, however, world of external objects as we believe by
and from within one of its own strongholds, common sense. On this view all percep-
the theory of perception. James Gibson's chal- tions are essentially fictions; fictions based
lenge to the representational theory of percep- on past experience selected by present sen-
tion has provoked some truly fundamental de- sory data. (Gregory, 1974, p. xviii)
bate within the Establishment journals in the
last couple of years. In this paper I shall try to There is something almost disarming about
explain the nature of his challenge and to ex- confusion of this magnitude, a theory which
amine both its origins in the behaviorist tra- denies the possibility of objective knowledge
dition and some of the important concerns it and then goes on to marshall facts in its
shares with operant psychology. support. But surely sympathy cannot in itself
James Gibson engaged in a sustained attack explain why representational theory has per-
upon cognitivism over many years, from the sisted for so long. As has become increasingly
thirties until his death in 1979, and, like Skin- evident, this paradoxical theory has much
ner, his motives were frankly epistemological deeper metaphysical ramifications.
(Gibson, 1967; Skinner, 1977b, p. 380; see also While operant psychology takes as unprob-
Costall, 1981; Michaels & Carello, 1981; Reed lematic the fact that organisms can come to
& Jones, 1982). He credits the behaviorist E. B. detect and discriminate the events occurring
Holt as a major influence on his thinking, but within their surroundings, the preoccupation
after initially attempting to repair the S-R of perceptual theory, especially the theory of
(stimulus-response) formula promoted by Holt, vision, has been with how this is possible,
Gibson eventually came to recognize that per- given that the organism is in contact not with
ception must be viewed as an act rather than the events as such but rather with ambient
as a response. Perceptual information, accord- energy, such as light or sound. The classical
ing to Gibson, is obtained, not imposed (Gib- puzzle of perceptual theory is that there is
son, 1979, pp. 56-57, 149-150). But in his recog- nothing in the structure of the immediate stim-
nition that such behavior does not conform to ulus which is specific to its source; the same
the classical scheme of reflex psychology, he image on the retina, for example, would seem
shared Skinner's conviction that so-called spon- to be consistent with an infinite set of possible
taneous behavior is nonetheless related to the circumstances in the world. Internal represen-
environment in a lawful way (cf. Skinner, 1938, tations were invoked to restore in some magi-
p. 20). His encounters with the Gestalt psy- cal fashion the absence of constraint available
chologist Kurt Koffka fired his interest in the from stimulation. Gibson's most conspicuous
problems of perception and reinforced Holt's contribution has been to question this, the
earlier insight (Holt, 1914, e.g., p. 122) that the most explicit function of representational the-
representational theory of perception presents ory, as a deus ex machina resolving the sup-
a primary target for the attack on cognitivism posed ambiguity of the structures available in
(Gibson, 1967, 1971). ambient energy. By urging a molar or higher
The representational theory of perception is level description of such structures, and by
one of those strange doctrines that most psy- pointing to the constraints which obtain upon
chologists are convinced they just cannot live such structures given the actual environment
without. Richard Gregory promotes this doc- in which the organism lives, Gibson, and his
 ARE THEORIES OF PERCEPTION NECESSARY? ill

students, have begun to identify variables in form of Morse code signals, as it were, so
stimulation that are uniquely related to envi- that what we experience as the "real
ronmental properties and events (see Gibson, world", and locate outside ourselves, can-
1979, chapters 4-8). Strangely enough, however, not possibly be anything other than a rep-
many cognitivists have eventually come to con- resentation of the external world. (p. 493)
cede Gibson's point about specification-that
is, the existence of information in Gibson's (I first encountered this passage in a valuable
strict sense-and yet still persist in their ways critique of representational theory by Noel
(e.g., Palmer, 1978). Quite clearly, representa- Smith, 1983.)
tional theory is (to borrow James's comment Interestingly enough, here there is no ap-
on Wundt's psychology) like a worm: You cut peal to the usual argument for the ambiguity
it up and each fragment crawls. It was Gib- of perception. Here we find the uncritical re-
son's persistence in dissecting the cognitivist tention of another aspect of the Cartesian
system that enabled him to unmask the various scheme, the notion of a mind lurking within
enchantments of representational theory, the the body, in direct contact only with the body
hidden agenda of the current debates. and not with the environment itself. This no-
As I mentioned earlier, while developing his tion, as Reed has recently argued, derives from
information-based theory of perception, Gib- the Cartesian hypothesis of corporeal ideas
son came to reject his earlier commitment to (Reed, 1982). Gibson's own criticisms of this
S-R theory. He insisted that the organism is assumption-for example, in his discussion of
active in a very literal sense in its perceptual the visual control of manipulation-echo the
exploration of the environment. Gibson, there- important arguments Skinner has voiced over
fore, like Skinner (e.g., 1969, pp. 3-13, 175; see many years concerning the persuasive myth of
also Skinner, 1938), came to abandon the S-R the "inner man" (e.g., Skinner, 1938, chap-
formula-and has suffered the same fate. Crit- ter 1):
ics simply refuse to believe that he can be any-
thing other than a bald proponent of a mecha- The movements of the hands do not con-
nistic behaviorism if he denies that perception sist of responses to stimuli.... Is the only
is an active effort after meaning-active, that alternative to think of the hands as instru-
is, in the peculiar sense that cognitive processes ments of the mind? Piaget, for example,
must somehow intervene between the stimulus sometimes seems to imply that the hands
and the response. It is the very failure of cog- are tools of a child's intelligence. But this
nitive psychologists even to comprehend, let is like saying that the hand is a tool of an
alone answer, the arguments of the opposition inner child in more or less the same way
that indicates that more fundamental issues that an object is a tool for a child with
are at stake. So let us delve a little more deeply. hands. This is surely an error. The alter-
Consider another curious statement of the native is not a return to mentalism. We
representational theory, this time by Fred Att- should think of the hands as neither trig-
neave: gered nor commanded but controlled.
(Gibson, 1979, p. 235)
Naively, it seems to us that the outside
world, the world around us, is a given; it is Unfortunately, the resources of representa-
just there.... We all feel as if our experi- tional theory are not easily exhausted. In its
encing of the world around us were quite defense, it exposes yet further reliance upon
direct. However, the apparent immediacy the mechanistic scheme of classical physics. Its
of this experience has to be more or less next resort is to the argument that only an in-
illusory because we know that every bit of stantaneous stimulus can be said to have an
our information about external things is immediate effect, or can be considered to enter
coming through our sense organs, or has into causal or lawful relations. The influence
come in through our sense organs at some of past events can enter into an account of be-
time in the past. All of it, to the best of havior, the argument goes, only insofar as we
our knowledge, is mediated by receptor can invoke some mediating representational
activity and is relayed to the brain in the structure to fill the gap in time. This plea of
112 A; P. COSTALL
cognitivism is familiar enough, of course, been "set up" by the program of classical
though its lineage is not always appreciated. mechanics. For if, to use Locke's metaphor,
Yet, as Jack Marr (1983) has recently noted: philosophers have been keen to serve as un-
derlaborers clearing away the rubbish gener-
It is not unlike the old problem in physics ated by such master-builders as Galileo and
of action at a distance. Newton suggested Newton, Gibson was not alone in realizing that
the possibility . that an all-pervading
. . psychology had been used as an all too con-
"aether" served as the medium for such venient dumping ground. Indeed, Edwin Burtt
phenomena as gravitation and light. Psy- (1954) made this point most clearly in his im-
chology has been replete with mental portant text, The Metaphysical Foundations
aethers that mediate between stimuli and of Modern Science:
responses. Indeed, cognitive psychology
seems to be paralleling classical physics It does seem like strange perversity in
in the search for an understanding of the these Newtonian scientists to further their
structure and mechanics of the mental own conquests of external nature by load-
aether. The mental aether must have the ing on mind everything refractory to exact
property of mediating action at a tempo- mathematical handling and thus render-
ral distance. (p. 13) ing the latter still more difficult to study
scientifically than it had been before....
It is evidently not enough to insist, as Skin- Mind was to them a convenient receptacle
ner does, that appeal to cognitive structures is for the refuse, the chips and whittlings of
unnecessary, or a "diversion" (Skinner, 1977a, science, rather than a possible object of
p. 10), in the sense that we can conduct re- scientific knowledge. (p. 320; see also
search and solve problems perfectly well with- Koyre, 1965; Mead, 1938; Whitehead,
out them. The argument that such structures 1926)
can be disregarded does not in itself call into
question their very existence, and indeed Skin- The dependence of representational theor-
ner at times talks as if they might well exist ists upon the ontology of classical science is
after all. What we require is an alternative most explicit in their ultimate resort to the
scheme that does not merely question the solu- distinction between primary and secondary
tions put forward by cognitive psychologists qualities of experience. Many aspects of per-
but converts their very problems from implicit ceptual experience, they argue, must be purely
to conspicuous nonsense. It is in Gibson's final subjective, mere mental constructions, in that
work towards such an alternative scheme, a they have no counterpart in the "real" world
new ontology, that his fundamental impor- -the world, that is, described by physics. Gib-
tance for psychology really lies. son's misgivings about this distinction (Gib-
Gibson, like Skinner, viewed science not as son, 1979, p. 31) began in some early work of
some sublime logical structure but as an aspect his in the 1930s when he found that perceptual
of human practice, and he showed a similar aftereffects held to be distinctive of such sec-
respect for the reflexive status of psychology ondary qualities as color and warmth also oc-
which this view entails. Both saw that the the- curred for so-called primary qualities such as
ory (and metatheory) of psychology must at line and curvature (Gibson, 1933). By the
the very least be compatible with the fact of 1940s, he came to reject the classical, essen-
the human practice of science. Yet, as a num- tially Euclidean, notion of space as a vast,
ber of critics have remarked, Skinner himself structureless container-as an abstraction ir-
retains, and indeed sometimes recommends, relevant to the psychology of perception-in
the physicalist ontology which has proved so favor of a conception of the visual world as a
troublesome for psychology, and, despite the set of overlapping surfaces (e.g., Gibson, 1950;
dialectical status of the concept of the operant, cf. Carr, 1935, p. 1). Later still, he came to in-
tends to treat the environment as though it sist that the physicalist dimension of time was
were an autonomous cause (Kvale & Grenness, not perceived; rather we perceive ongoing
1975; Malone, 1975). events (Gibson, 1975; 1979, pp. 253-254). In
Gibson proved a good deal more alert to the his last book, in his theory of affordances, he
unfortunate sense in which psychology has went on to argue that we can properly be said
 ARE THEORIES OF PERCEPTION NECESSARY? 113
to immediately perceive the functions that ob- implication of the ecological perspective. Gib-
jects serve for our activities (1979, chapter 8). son denied perhaps the most central Cartesian
The crucial claim of Gibsonian theory is that assumption underlying cognitivism, that the
many of the so-called secondary qualities are relation between organism and environment is
indeed real properties of our environment, an essentially external one, the idea that the
and, furthermore, the structures available in organism can be construed as if it could exist
ambient energy are related to such environ- outside of any kind of coordination with an
mental properties and events in a lawful way- environment. In contrast, both ecological and
they uniquely specify them. All the organism operant psychology draw upon the important
needs to do is detect these informative struc- insight of early functionalist psychology (e.g.,
tures, and all that perceptual theory has to Dewey, 1896, 1898/1976) that it is the very
do, in turn, is explain what these structures are coordination of organism and environment
and how they are "picked-up": that must constitute the basic unit of analysis
for psychology. Both operant and ecological
The theory of psychophysical parallelism psychology are committed to the view that the
that assumes that the dimensions of con- relation between organism and environment is
sciousness are in correspondence with internal; neither term in this relation can be
the dimensions of physics and that the defined independently of the relation itself.
equations of such correspondence can be But in taking this view we should be clear
established is an expression of Cartesian about its implication, for it follows that the
dualism. Perceivers are not aware of the environment can no longer be considered, as
dimensions of physics.... They are aware it is in the Cartesian scheme, as an autonomous
of the dimensions of the information in cause (cf. Hocutt, 1967). I cannot say that
the flowing array of stimulation that are either Gibson or Skinner is altogether clear on
relevant to their lives. (Gibson, 1979, p. this point, but a most lucid statement can be
306) found in the writings of the evolutionary biol-
ogist Richard Lewontin (1982):
Gibson's effort to deny the metaphysical
basis of the distinction between primary and This view of environment as causally
secondary qualities, far from being an obsti- prior to, and ontologically independent
nate attempt to deny the very existence and of, organisms is the surfacing in evolution-
successes of science, was a considered effort ary theory of the underlying Cartesian
towards determining its proper empirical structure of our world view. The world is
basis, a task anticipated in some detail, in fact, divided into causes and effects, the exter-
in the writings of Whitehead (e.g., Whitehead, nal and the internal, environments and
1926, 1930, p. 48). the organisms they 'contain'. While this
Two points should be added about the eco- structure is fine for clocks, since main-
logical ontology which Gibson developed to springs move the hands and not vice versa,
displace Cartesian dualism. The first concerns it creates indissoluble contradictions when
the fact that the ecological laws referred to by taken as the meta-model of the living
Gibson are certainly circumscribed; as he takes world. (p. 159)
care to stress, they hold within the normal
ecology of the organism. Such restriction, how- (Lewontin moves towards overstatement, how-
ever, is not a peculiarity of ecological laws. A ever, when he continues that "organisms
crucial insight of modern physics has been that within their individual lifetimes and in the
laws specifying invariant relations need to be course of their evolution as a species do not
defined relative to an appropriate "domain of adapt to environments; they construct them"
validity" (Bohm, 1965, chapter 25). Neverthe- [p. 163; cf. Dewey, 1898/1976, pp. 279-284].)
less, many psychologists still happily pit one So far I have tried to set out the ways in
theory against another, or uncritically invoke which Gibson's attack upon representational
Popper's canon of falsification, without any re- theory has served to expose and challenge the
gard for the different sets of circumstances to deeper metaphysical assumptions of cognitive
which the theories might apply. psychology. Cognitivism is hardly about to
The second point concerns a more profound give up the ghost of Cartesian dualism, and
114 A. P. COSTALL
perhaps it will only succumb to death by a a proper psychology of cognition, a psychology
thousand qualifications. But there can be no which treats the relevant phenomena as neces-
doubt that its complacency has been disturbed, sarily grounded in social practices rather than
and so I must finally examine its usual retreat upon the essentially private and individual
from the field of theoretical wrangling to the mental structures invoked by cognitive psy-
apparently clearer ground of empirical data. chology. Much more needs to be said about
The first appeal, to laboratory experiments, their contributions in this direction, but in
seems incontrovertible enough until we real- this paper I can on-ly take the opportunity to
ize that the psychological laboratory is the very set the record straight.
microcosm of the Cartesian scheme. After all, My primary concern in this review article
our major experimental paradigms are de- has been to introduce Gibson's critique of rep-
signed explicitly to prevent the organism from resentational theory and to explain the way
transforming the experimental situation, as that it has served to expose so many vestiges
would be possible to some degree in real life of Cartesian metaphysics within contemporary
(Gadlin & Rubin, 1979). The subjects are free cognitivism. I hope I have made clear some of
only in the sense that they can react to, rather the common ground which exists between eco-
than change, the conditions which are imposed logical and operant psychology. The ecological
upon them. Furthermore, when critics, such as approach and operant psychology share a good
Fodor and Pylyshyn (1981), for example, dis- deal more than mere disenchantment with the
miss Gibson's claims for the existence of eco- status quo. Both insist that behavior presents
logical laws on the grounds that "it has been a primary datum for psychology which is not
repeatedly shown in psychological laboratories to be treated as a mere symptom of underlying
that percepts can be caused by samples of the structures of either the cognitive or physiologi-
ambient medium which demonstrably under- cal kind. They recognize that the description
determine the corresponding layout" (p. 172), of behavior is nevertheless difficult, and they
they not only choose to ignore Gibson's re- promote a molar and functional classification
quirement that our experiments should model of behavior rather than muscle-twitch psychol-
the normal ecology of the organism, but they ogy or classical reflexology. In rejecting the
also disregard the carefully defined limits Gib- S-R scheme, however, they insist that behavior
son has set to his theory. Gibson's is a theory is nonetheless subject to lawful description and
of direct perception, not a direct theory of per- that these laws refer to an irreducible organ-
ception. How people cope with the bizarre sit- ism-environment relationship. Finally, they
uations dreamt up in most psychological labo- each have special contributions to make
ratories is quite explicitly outside the scope of towards a proper psychology of cognition-a
ecological theory. psychology, that is, concerned with truly medi-
The last resort of cognitive theory is to the ated modes of behavior.
fact that people do indeed follow rules and In 1915, Gibson's mentor, Edwin Holt, at-
represent things-though, of course, they do tempted to survey the many groups seeking an
much else besides. Cognitive psychologists are alternative to the traditional cognitivist
quite wrong, however, to suppose that these scheme, and came to the following conclusion:
facts about human beings can alone support
their entire edifice of cognitive structures, and It should be obvious that a fundamental
they are just dishonest when they pretend that unity of purpose animates the investiga-
Skinner or Gibson ever wished to deny these tors of these several groups, although they
facts. Gibson (1979, pp. 258-263) was quite approach the question of cognition from
clear that the obtaining of "secondhand infor- very different directions. Will it not be a
mation" through words, pictures, and writings source of strength for all if they can man-
must be considered as truly mediated percep- age to keep a sympathetic eye on the meth-
tion, and Skinner has gone even further in ods and discoveries of one another? (Holt,
elaborating an account of how the verbal com- 1915, p. 208)
munity comes to mediate much of our behav-
ior (Burton, 1982; Skinner, 1945, 1957; see Some seventy years after Holt's suggestion,
also Tikhimorov, 1959). In fact, Skinner and this alliance is surely overdue. Operant psy-
Gibson were working towards what they saw as chologists and ecological psychologists are not,
 ARE THEORIES OF PERCEPTION NECESSARY? 115

of course, the sole opponents of the cognitivism Holt, E. B. (1914). The concept of consciousness. New
that still prevails. But, as I have tried to ex- York: Macmillan.
plain, they, at least, should keep "a sympa- Holt, E. B. (1915). The Freudian wish and its place in
ethics. New York: H. Holt.
thetic eye" on one another's progress. Koyre, A. (1965). Newtonian studies. London: Chap-
man & Hall.
Kvale, S., & Grenness, C. E. (1975). Skinner and Sartre:
REFERENCES Towards a radical phenomenology of behavior. In
A. Giorgi, C. T. Fischer, & E. L. Murray (Eds.),
Attneave, F. (1974). How do you know? American Psy- Duquesne studies in phenomenological psychology,
chologist, 29, 493-499. Vol. 2 (pp. 38-59). Pittsburgh: Duquesne University
Bohm, D. (1965). The special theory of relativity. New Press.
York: Benjamin. Lewontin, R. C. (1982). Organism and environment.
Burton, M. (1982). Theses on the verbal community. In H. C. Plotkin (Ed.), Learning, development, and
Behaviour Analysis, 3(3), 26-29. culture (pp. 151-172). Chichester: Wiley.
Burtt, E. A. (1954). The metaphysical foundations of Malone, J. C., Jr. (1975). William James and B. F.
modern physical science (rev. ed.). Garden City, NY: Skinner: Behaviorism, reinforcement, and interest.
Doubleday. (First published 1924) Behaviorism, 3, 140-151.
Carr, H. A. (1935). An introduction to space percep- Marr, J. (1983). Memory: Models and metaphors. Psy-
tion. New York: Longmans, Green. chological Record, 33, 12-19.
Costall, A. P. (1981). On how so much information Mead, G. H. (1938). The philosophy of the act. Chi-
controls so much behaviour: James Gibson's theory cago: University of Chicago Press.
of direct perception. In G. Butterworth (Ed.), In- Michaels, C., & Carello, C. (1981). Direct perception.
fancy and epistemology: An evaluation of Piaget's Englewood Cliffs, NJ: Prentice-Hall.
theory (pp. 30-51). New York: St. Martin's Press. Palmer, S. E. (1978). Fundamental aspects of cognitive
Dewey, J. (1896). The reflex arc concept in psychology. representation. In E. Rosch & B. B. Lloyd (Eds.),
Psychological Review, 3, 357-370. Cognition and categorization (pp. 259-303). Hillsdale,
Dewey, J. (1976). Lectures on psychological and politi- NJ: Erlbaum.
cal ethics, 1898. (D. F. Koch, Ed.) New York: Hafner Reed, E. S. (1982). Descartes' corporeal ideas hypothe-
Press. sis and the origins of scientific psychology. Review
Fodor, J. A., & Pylyshyn, Z. W. (1981). How direct is of Metaphysics, 35, 731-752.
visual perception?: Some reflections on Gibson's Reed, E., & Jones, R. (1982). Reasons for realism:
"ecological approach." Cognition, 9, 139-196. Selected essays of James J. Gibson. Hillsdale, NJ:
Gadlin, H., & Rubin, S. H. (1979). Interactionism: A Erlbaum.
nonsolution of the person-situation controversy. In Skinner, B. F. (1938). The behavior of organisms. New
A. R. Buss (Ed.), Psychology in social context (pp. York: Appleton-Century.
213-238). New York: Irvington. Skinner, B. F. (1945). The operational analysis of psy-
Gibson, J. J. (1933). Adaptation, after-effect and con- chological terms. Psychological Review, 52, 270-277.
trast in the perception of curved lines. Journal of Skinner, B. F. (1950). Are theories of learning neces-
Experimental Psychology, 16, 1-31. sary? Psychological Review, 57, 193-216.
Gibson, J. J. (1950). The perception of the visual Skinner, B. F. (1957). Verbal behavior. New York: Ap-
world. Boston: Houghton Mifflin. pleton-Century-Crofts.
Gibson, J. J. (1966). The senses considered as percep- Skinner, B. F. (1969). Contingencies of reinforcement:
tual systems. Boston: Houghton Mifflin. A theoretical analysis. New York: Appleton-Century-
Gibson, J. J. (1967). Autobiography. In E. G. Boring & Crofts.
G. Lindzey (Eds.), A history of psychology in auto- Skinner, B. F. (1977a). Why I am not a cognitive psy-
biography, Vol. 5 (pp. 127-143). New York: Apple- chologist. Behaviorism, 5, 1-10.
ton-Century-Crofts. Skinner, B. F. (1977b). The experimental analysis of
operant behavior. Annals of the New York Academy
Gibson, J. J. (1971). The legacies of Koffka's Princi- of Sciences, 290, 374-385.
ples. Journal of the History of the Behavioral Sci- Smith, N. W. (1983). Sensing is perceiving: An alterna-
ences, 7, 3-9. tive to the doctrine of the double world. In N. W.
Gibson, J. J. (1975). Events are perceivable but time is Smith, P. T. Mountjoy, & D. H. Ruben (Eds.), Re-
not. In J. T. Fraser & N. Lawrence (Eds.), The study assessment in psychology: The interbehavioral alter-
of time II (pp. 295-301). Berlin: Springer-Verlag. native (pp. 161-211). New York: University Press of
Gibson, J. J. (1979). The ecological approach to visual America.
perception. Boston: Houghton Mifflin. Tikhimorov, 0. K. (1959). Review of B. F. Skinner's
Gregory, R. L. (1974). Concepts and mechanisms of Verbal Behavior. Word, 15, 362-367.
perception. London: Duckworth. Whitehead, A. N. (1926). Science and the modern
Hocutt, M. (1967). On the alleged circularity of Skin- world. Cambridge: Cambridge University Press.
ner's concept of stimulus. Psychological Review, 74, Whitehead, A. N. (1930). The concept of nature (2nd
530-532. ed.). Cambridge: Cambridge University Press.