September 1971
The Crisis in Sensory Physiology
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 References to these essays must cite them explicitly as unpublished manuscripts.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.
In sense physiology, the excitation of a receptor by stimulus energy is nowadays said to carry information about the environment. This information is said to be conveyed to the central nervous system and then it is supposed to arouse a sensation, the sensation being specific to the receptor stimulated. But the input can only “carry information” to the extent (1) that each receptor is specialized to receive one and only one kind of energy, and (2) that each receptor transduces or translates the intensity, frequency, or other quality of the stimulus into nervous impulses in a specific way, that is, only to the extent that the is “psychophysical correspondence” between stimulation and sensation.
Even if this orthodox formula were correct, the information carried to the central nervous system would only be such as to specify the stimulus environment, not the environment of objects and events that aresources of stimuli. Only the pattern and flux of the sea on ambient energy in which the animal lives would be specified, not the pattern and change of the surrounding substances and surfaces of his environment. He would detect the proximal stimuli only, and they are notoriously poor indicators of the environment proper. The environment of stimulation, which includes self-produced or obtained stimulation as well as imposed stimulation, should not be confused with the environment that stands behind stimulation and makes it possible. But physiologists do not usually realize this fact, lumping everything together under the term physical environment.
But even the orthodox formula of sense physiology is inadequate. It is beginning to appear that receptors are not highly specialized to receive one and only one kind of energy, and that each receptor does notsimply translate the physical variables of stimulus energy into corresponding variables of nervous impulses. If so, nervous inputs to the brain cannot be said even to carry information about stimuli,” let alone information about the environment.
In the face of this situation, sensory physiologists are (reluctantly) appealing to vague concepts like patterns of stimuli and are otherwise casting around for new explanations. The question arises whether it would be more useful for them to discard the whole theory of discrete stimuli, discrete receptors, one-way sensory inputs, and central “sensations” instead of trying to salvage it and reconcile it with the facts of information-pickup. They should consider the logical possibility that information about the environment cannot possibly be transmitted along a nerve; then they can begin to think about ways in which a perceptual system might work.
The kind of information necessary for perception is simply not comparable to the kind of information that can be transmitted as messages from a sender to a receiver. It is true that nerve impulses are necessary for the activity of perception, along with adjustable organs and nerve centers, but the fibers run in both afferent and efferent directions. It is true that cutting a nerve stops nerve activity. But this does not imply the doctrine of sensory messages transmitted to the brain by way of the sensory nerves, along with its embarrassing corollaries of a sensorium, a seat of consciousness, a storehouse of memories, and so on. The activity of perception is not localized at any point in the circular process, nor is it terminated in any part of the nervous system.