May 1975
What is the Relation of Concepts to Percepts?
J. J. Gibson, Cornell University
The World Wide Web distribution of James Gibson’s “Purple Perils” is for scholarly use with the understanding that Gibson did not intend them for publication. References to these essays must cite them explicitly as unpublished manuscripts. Copies may be circulated if this statement is included on each copy.
I. One concept of the concept is that it is the psychological average of a group of percepts. This may be either a numerical average or an average form, that is, a general image analogous to a composite photograph. The theory behind this seems to be that successive percepts of the same object (or percepts of similar objects belonging to a class) leave engrams in the brain which tend to form, so that the variations cancel out. Something like this was implied by Koffka’s “trace column.”
Since the objects of a class vary somewhat, every time a new number of the class is encountered the process of perception will involve an assimilation of the new form to the general image. The process of concept formation will involve the accommodating of the general image to each new form. Percepts involve sensations and memories whereas concepts involve only memories. This formula goes back further than Piaget, to G. E. Müller, for example. Note that a general image based on many engrams is stronger and more typical than one based on few engrams.
The hypothesis of average form has even been applied in the effort to explain form-constancy. The sequence of perspective forms of a solid object, the series of “views,” has been thought to average out somehow so as to yield the object itself in the sense that the true object is that seen from an “optical” point of view. But the hypothesis simply will not work in this case, as a little thought will show.
II. Another concept of the concept is that it is innate to some degree instead of being just an average of past phenomenal forms or appearances. This involves the theory of archetypes or prototypes, going back to the Scholastics and eventually to Plato. Each percept, e.g., of a dog, is assimilated to the general concept of dog. The accommodating of the concept to new percepts is not emphasized. The individual percepts of Rover, Fido, Spot, etc., are not as “real” as the ideal concept since they are merely instances of it. Concepts may be formed empirically but need not be. The discipline of taxonomy began with ideas of this sort as described by G. G. Simpson in Principles of Animal Taxonomy. But they are, as he points out, nonsense.
Is it true that an animal (or a plant) is defined by its “form”? Is it even defined by its solid anatomical shape in three dimensions? If we take a true organism to be its life-cycle, as we should, then it has many forms between the beginnings as a single cell and its death; the adult form is only one. Is it true that an inanimate object like a chair (better, a seat) is defined by its “form”? Surely an object or layout that affords sitting can have many forms (couch, bench, ledge, etc.).
III. It would seem that the hypothesis of average or ideal form is loaded with inconsistencies, although it is still influential in psychology. How do we get rid of it? An alternative to either formal or numerical norms is invariants. If we adopt this hypothesis will we not have to reject the formula of the assimilating of percepts to concepts and the accommodating of concepts to percepts. But that seems to reject the very distinction between percepts and concepts. Do the developing capacities of an observer to detect the affordances of the environment for behavior include both what we have called perceiving and conceiving in one unitary activity? If so the entities that we call percepts and concepts will disappear from psychology.