February 1972
Note on The Differences between a Sensory Modality and a Perceptual System
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 References to these essays must cite them explicitly as unpublished manuscripts.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.
Five or six years ago, with the publication of The Senses Considered as Perceptual Systems, the distinction between a sensory modality and a perceptual system was made. It is not clear to everyone. The purpose here is to review and elaborate the distinction. The verb to sense can mean either to have a sensation or to detect something and the two meanings are radically different.
1. A sensory modality is a channel of input from a receptor mosaic along a nerve; a perceptual system is a circular process of input-output between the periphery and the brain.
A “sense” considered as a modality is thus defined anatomically whereas a perceptual system is defined functionally so as to include peripheral adjustments.
2. A sensory modality only allows the possibility of filtering the inputs of the nerve or channel; a perceptual system can modify the inputs of the channel.
Thus a perceptual system can obtain or prevent stimulation of a receptor mosaic, as in looking or not looking and touching or not touching.
3. A sensory modality is studied in the laboratory by imposing stimulation and preventing exploration; a perceptual system can be studied by making available an external array of potential stimulation andpermitting its normal activity.
Thus attention can only be a covert activity in the nervous system for a sense modality, whereas it can be an overt observable activity for a perceptual system. Four types of overt activity are exploration,adjustment, orientation, and optimization. But these are not forms of “behavior”.
4. A perceptual system contains components of receptors, receptive surfaces, and adjustable organs which exist in a relation of subordination and superordination; a channel of afferent input is thus only acomponent of an active perceptual system (cf. Ch. 2 of the Senses Considered).
A bank of receptors, a mosaic, or a sensory surface of energy-transductors, is of course a necessary component of a perceptual system. But the firing of photoreceptors, mechanoreceptors, or chemoreceptors is not sufficient for perception. The perceptual systems, moreover, utilize different combinations of receptor banks at different times, that is, the receptive components overlap.
The fact that a sensory surface like the retina actually contains not only a mosaic of photoreceptors but also a set of “receptive units” is consistent with the above distinction.
5. For a sensory modality a so-called “stimulus” can only consist of the separate local excitations of specific receptors in the mosaic; for a perceptual system there is no “stimulus” and there are no separate adjacent “stimuli”, but instead there is an ordered external array of ambient stimulus energy.
The existing confusion in psychology about stimuli, stimulus patterns, stimulus situations, and stimulus objects is resolved if we distinguish sharply between the local excitation of receptors and the detection of relations in ambient energy.
6. A sensory modality transmits sensory signals to the brain (“sensations” to a “sensorium”); a perceptual system does not normally transmit anything to the brain.
The theory that sensory inputs have to be processed in the brain depends on the analogy of signal-transmission. But if we assume that a perceptual system only operates in this manner, even approximately, when its overt activity is prevented then the theory of input processing (or “information” processing, to beg the question) has only a limited application. We think instead of theories of information “pickup”, or the extracting of “invariants”, or of input-output circuits that can “resonate” to the invariants.
7. The sensory modalities are supposed to have specified phenomenal qualities that corresponds to the sensory nerve and even to the sensory neurons within it, by Johannes Müller’s law; the perceptual systems are not supposed to have specific phenomenal qualities.
The attempt to classify sensations by introspections (to “inventory” them) has not been successful, and hence they cannot be assigned to nerves and neurons. The human perceptual systems, on the other hand, can be classified functionally into five modes of overt attention (looking, listening, feeling, smelling, and tasting) together with the superordinate system of orienting (Ch 3 and 4).
8. The deliverance of a sensory modality are sense data or sense impressions, which are meaningless; the outcome of the activity of a perceptual system is perception, which is meaningful.
Thus, on a sensation-based theory of perception, meaning is something that is added to experience whereas in an information-based theory of perception meaning is something that is discovered in the experience.
9. The concurrent inputs of the two sensory modalities such as sensations of vision and touch aroused by the same object have nothing in common; the concurrent products of two perceptual systems in this situation may be wholly equivalent.
Thus, in the concurrent (and covariant) activities of looking at and feeling a solid object the features of shape and texture emerge in both haptic and optic exploration. The invariants can be detected by eitherperceptual system (although the sensations vary from moment to moment). The information for shape and texture is the same; it does not have to be learned by association or mediated by words as the supposed equivalence of sensations of vision and touch would have to be (See Ch. 7). It must be admitted that the haptic system cannot detect the color of a surface and the visual system cannot detect the temperature of a surface but this does not detract from the above hypothesis.
A great many experiments on the unsolved puzzles of “touch” and “vision” have to be reinterpreted if this hypothesis is correct.
Conclusion. When a psychologist or a physiologist speaks of sense perception he usually means perception that entails either the having of sense data or the transmission of sensory signals to the brain. But he might mean perception that “senses” the information in ambient light, sound, and contact and thus detects the features of the surrounding world. This kind of perception does not entail either sense data or sensory signals, neither a modality of impressions nor a specific set of neural inputs.