February 1960
Proposed Set of Non-contradictory Assumptions about the Nature of Stimuli The Use of the Word “Stimulus”: Conclusions of the Survey
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 References to these essays must cite them explicitly as unpublished manuscripts.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.
If all these earlier definitions are mutually contradictory; and if the various ones are vague, or narrow, or otherwise inadequate, what should be our concept of the stimulus in psychology?
1. Stimuli are kinds of physical energy. To understand them we have to go to physical sciences like optics, acoustics, chemistry, mechanics (dynamics, kinetics), physiography, and mathematics. But we need to study these physical sciences from an ecological point of view. We require an environmental optics, acoustics, mechanics, etc.
2. Stimuli should be analyzed and quantified whenever possible as variables. But the units appropriate for analyzing physical energies are not necessarily the appropriate ones for analyzing stimulus energies. We must admit “higher order” variables, and “invariant” variables. Also, when stimulus measurement is difficult or impossible, we must use concepts from non-metric mathematics–manifold, array, flow, structure, hierarchy.
3. The physical environment of any animal provides an inexhaustible reservoir of potential stimuli.
4. Stimulus energy always involves adjacent and successive order (ordinal stimulation). That is, there is always an array, and a flow. However small the units for analysis we choose, we must considerstructure and sequence if we are treating energy as stimulus energy. (Points and instants are useful fictions for physics, but not so useful for physiology or for psychology. This applies also to frequencies and intensities.)
5. Stimuli can be arranged in hierarchies of structure and of sequence. That is, there exist patterns of pattern and sequences of sequence. (e.g., gradients is the field of view; trends in the flow of sound;directions in temporal transformations). These provide one kind of “higher order variables”, and more inclusive units are required for their specification. They might be called molar stimuli.
6. So conceived, stimulus energy at the skin of any animal carries information about the physical environment. That is the stimulus specifies its source, object or event, more or less univocally. The assumption that stimuli are necessarily and intrinsically meaningless is unwarranted.
7. In psychology, as distinguished from cellular physiology, we must consider the potential stimuli for receptive systems or sense organs, not just those for receptor cells. We may also wish to consider combinations of and relations between stimulation for the whole sensory system, i.e., stimulation for the “organism as a whole”. Multiple concurrent stimuli for several senses may carry “redundant” information.
8. Effective stimuli, as distinguished from potential, are those which arouse receptor excitation, or sense-organ adjustment, or recorded neural impulses, or overt responses, or verbal judgments whichever criterion one chooses. Potential stimuli for an animal become effective depending on anatomy, stage of maturation, sensory exploration, “attention”, and the “education” of attention. (The notion of fixed, innate thresholds of sensation, both absolute and relative, is a myth.)
9. Many potential stimuli or variables of potential stimulation are hypothetical ones in out present state of knowledge. To prove that any such can be effective, the psychologist must isolate and vary it in a psychophysical experiment (or discrimination experiment) and obtain positive results.
10. Stimulus energies arising from the natural physical environment specify their sources directly. Stimuli from what is called the “cultural” environment, however, are often surrogates or symbols which stand for something else. Hence they can specify their referents only indirectly by way of social agreement. They may properly be called coded stimuli, and the former should be called uncoded. Consequently the problem of stimuli as cues or clues (and the problem of “meaning” in perception) takes two distinct forms and will require different solutions. Learning to perceive objects is not the same as learning to apprehend objects by virtue of perceiving words (although they become mixed in the human adult).