More on Affordances

March 1971

More on Affordances

J. J. Gibson, Cornell University

This memo has been published in E.S. Reed & R. Jones (Eds.) (1982). Reasons for Realism. Chapter 4.9, Parts III (pp. 406 – 407) and IV (pp. 407 – 408). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.

At one extreme stands the fact that educated adults have a conception of space, i.e., mathematical or geometrical space, Euclidean, Cartesian, non-Euclidean, etc.
Then there is what psychologists have called the perception of space. Although it is a complete muddle, and full of contradictions, depth-perception implies distance-from-here, and such perception recognizes at least the fact of a potential observer and a surrounding space.
Next there is what I call the perception of layout – the actual layouts of environmental surfaces, chiefly opaque solid surfaces, and the geometrical components of layout. Such perception depends on optical information for environmental places and objects at the set of all possible points of observation in the medium, and this takes into account both hidden an (unprojected) and unhidden (projected) surfaces at a fixed point of observation.
Finally, at the other extreme, there is the perception of the affordances of environmental surface layouts (which include objects and places and even animate objects). The activity of an observer that is afforded depends on the layout, that is, on the solid geometry of the arrangement. The same layout will have different affordances for different animals, of course, insofar as each animal has a different repertory of acts. Different animals will perceive different sets of affordances therefore. The perception is of practical layout, not theoretical layout, but it is nonetheless geometrical for all that. Animals, and children until they learn geometry, pay attention to the affordances of layout rather than the mathematics of layout. Hence, although logically one advances from space to affordance, developmentally the progress is in the opposite direction, from affordance to space. The formless invariants in the light which the eyes of the very young pick up, instead of the forms of the visual field, are just those that specify affordances.

Still More on Affordances

There has been a great gulf in psychological thought between the perception of space and objects on one hand and the perception of meaning on the other. But when space and objects are defined in terms of the opaque solid geometry of surface layout, and when meaning is defined in terms of the affordances of places, substances, surfaces and objects (hereafter termed “things”), these problems are seen to be linked. For example, what anything affords an organism depends in some degree on its shape or the features of its shape (solid shape, of course, not pictorial form). Hence it is that the shape of something is especially meaningful. The meaning or value of a thing consists of what it affords. Note the implications of this proposed definition. What a thing affords a particular observer (or species of observer) points to the organism, the subject. The shape and size and composition and rigidity of a thing, however, point to its physical existence, the object. But these determine what it affords the observer. The affordance points both ways. What a thing is and what it means are not separate, the former being physical and the latter mental as we are accustomed to believe.
The perception of what a thing is and the perception of what it means are not separate, either. To perceive that a surface is level and solid is also to perceive that it is walk-on-able. Thus we no longer have to assume that, first, there is a sensation-based perception of a thing and that, second, there is the accrual of meaning to the primary percept (the “enrichment” theory of perception, based on innate sensations and acquired images). The available information for the perception of what it affords.
The controversies over whether the values of things are “relative” or “absolute,” and whether value is a subjective phenomenon or an objective fact, should be reinterpreted in the above terms.