Outline of a New Attempt to Classify the Senses and the Sensory InputsJ. 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 References to these essays must cite them explicitly as unpublished manuscripts.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.
The standard classifications of sensations, nerves, sense-organs, stimuli, and physical events are unsatisfactory. The classical sensations, it is now realized, are a poor guide to a description of the receptive process in general. The classical variables of stimulation, borrowed from physics, are oversimplified. Above all, the relation between stimuli and their sources in the environment, the so-called “proximal” and “distal” stimuli, has never been worked out in detail.
The channels of sense-perception cannot be represented by any fixed list of mutually exclusive “senses”. It is possible, however, to make categories and subcategories at different levels of the total receptive process. In what follows, I have tried to classify separately (a) the receptive system which can be excited, (b) the types of stimulus energy that can excite them, (c) the environmental fields of potential stimulation, (d) the environmental sources of these fields, and (e) the general sources of stimulation, including the individual himself. The effort is to get an overall picture of the receptive process as a whole with all the types of information that can be registered when the receptive channels work in cooperation.
A. The Receptive Systems
There is some vagueness in both of the terms “receptor” and “sense organ”. Instead of using them, I shall distinguish between receptive elements, sensory surfaces, and sensory systems. A receptive element is the structure or small group of structures connected with an afferent neuron. There are always bundles of these neurons and, correspondingly, the elements adjoin. But groups of terminal structures often overlap. The result is a sensory surface. The skin and the retina are notable examples of a sensory surface with overlapping receptive fields. Without overlapping neural connections the surface could appropriately be called a “receptor mosaic”.
A sensory system consists, at a minimum, of a sensory surface with its neural connections. This may be elaborated by accessory structures for selecting and filtering energy, thus producing an “organ”, like the eye or the ear. A system is also often provided with motor equipment for adjusting and exploring stimulation, that is, for optimizing in some sense the input of the receptive surface. The ocular systemhas specialized motor equipment for this purpose. The tactile system of the body extremities has only the regular performatory motor equipment for this function, but the hands for example can nevertheless adjust for this function, but the hands for example can nevertheless adjust and explore the possibilities of contact stimulation.
Following are the known sensory or receptive surfaces of the body.
1. The vestibular complex or system, and its organs.
a. The maculae (hair-cells) of the utricle and saccule (i.e., the “statocysts”)
b. The cristae (hair cells) of the semicircular canals
2. The receptive elements lining the articular surfaces (joints)
3. The receptive elements imbedded in muscles and tendons
4. The cutaneous and deep-tissue surfaces
a. The hairs of the skin and their receptive roots
b. The elements distributed in the skin proper, especially at the mobile extremities (Are the fingers sense-organs?)
c. The elements distributed in the sheathing of muscles and of bones
5. The elements of the olfactory epithelium, in connection with the nasal structure and the breathing mechanism.
6. The elements of the gustatory surfaces (taste-buds), in connection with the mouth and the eating mechanisms.
7. The bilateral auditory complex. Two organs of Corti (hair cells), each with elaborate accessory structures (cochlea, windows, middle ear, eardrum, and pinna). The system includes adjustments (ear turning or head turning) for “localizing”.
8. The bilateral visual complex. Two retinas (photosensitive elements), each with accessory structures (eye, lens cornea, etc.) and elaborate adjustive mechanisms (of the lens, pupil, eyelid, and the reflexes for ocular stabilization) and also elaborate exploratory mechanisms (for convergence, pursuit, and saccadic eye movements).
Numbers 1, 2, and 3 above are relatively passive or purely receptive systems. Number 4, “touch”, can be either passive or active. Numbers 5, 6, 7, and especially 8 are active and exploratory receptive systems. It may be noted that the first three systems are receptive to the bodily movements of the individual and are what Sherrington called “proprioceptors”, whereas the last four systems are receptive to external events are called “exteroceptors”. But, as will appear later when sources of stimulation are listed, all receptors can be proprioceptive in the above meaning of the term. Not all receptors, however, can selectively explore the possibilities of stimulation.
Sherrington’s third class, the interoceptors, have been omitted from the above list. If they exist in physiological fact, they are not yet understood.
B. The Types of Energy Which Can Affect the Receptive Systems
It has often been assumed that the sense organs are specialized for the reception of the different forms of physical energy, as the eye for light, the ear for sound, the nose and mouth for chemicals, and the skin for mechanical touch. But this seriously distorts the facts. It is necessary to distinguish between physical energy and stimulus energy before the activity of receptive systems can be properly understood. They are specialized for the latter, not the former.
Physical energy and its various transformations, as studied by physics and as applied to biophysical and biochemical energy exchanges between organisms and their environment, is one thing. Energy, as such, is manifested in metabolism; it can nourish, damage, or destroy an organism, but stimulus energy consists only of that which reaches a sensory surface, and the amount of energy required to excite a receptor is extremely small. Whether a single receptive element is triggered or many elements in combination, the amount of work performed is negligible. The forms of physical energy can be listed as follows:
1. Electrical energy (current flow)
2. Chemical energy (chemical action)
3. Molar kinetic energy, or heat
4. Molecular kinetic energy, or heat
5. Propagated kinetic energy, that is, longitudinal pressure-waves in a material medium
6. Radiated electromagnetic energy, or radiation, that is transverse waves traveling in a straight line through empty space or clear air.
By no means all of the manifestations of these forms of energy will affect receptors. The last two, consisting of “waves”, exist over wide ranges of wavelength (frequency), but only a limited band of “sound” frequency is picked up by a given ear, and only narrow band of electromagnetic frequencies called “light” is picked up by a given eye, with an adjacent band called “radiant heat” registered by the skin.
With respect to the first four types of energy, they are not transmitted through an environmental medium and can only affect a sensory surface when in contact with it.
In addition to energy, physics defines fields of force. There are three types, electrical, magnetic, and gravitational. The first two have no effect whatever on receptors. The last, however, is detected indirectly by ways of mechanical energy or “pressure” on the lining of a statocyst, or on the skin.
The types of stimulus energy, as contrasted with physical energy, can be listed as follows, in parallel with the list above.
1. Transitions of current flow at any receptive element or receptive surface. An electrical impulse is a universal stimulator of receptors, but an artificial one, and is so used by sensory physiologists for experimental purposes. the firing of a nerve cell being itself an electrochemical process, this fact is not surprising. But as a stimulus, it is called inadequate, meaning inappropriate or unsuitable. Electric shocks are rare in the terrestrial environment, and animals have not developed special receptors to register them (although some fish may have done so).
2. Chemical reactions, mostly with organic substances, at a surface of chemoreceptors
a. For soluble substances at the gustatory surfaces
b. For volatile substances at the olfactory epithelium
3. Mechanical deformation of any receptive tissue or sensory surface in the vestibule, joints, muscles, cutaneous hairs, outer skin, or deeper tissue. A variety of mechanical actions are possible; bending, tension, compression, pressure, vibration, scraping, impact, or mere “touching”, but the receptive elements are all presumably excited by some manifestation of kinetic energy. Although variable in structure, the microscopic elements are all termed mechanoreceptors.
4. Heat flow at the skin, due to
a. Conduction of heat between the skin and another body
b. Convection of heat, by currents of air (e.g., wind)
c. Radiation of heat.
5. Sound vibrations of the eardrum and thence, or otherwise, transmitted to the cochleas.
6. Light reaching a surface of photoreceptors, such as the retina.
The first three types of stimulus energy, and number 4a, are transmitted to the receptive surface by local action. However, the last two types, and number 4c, consisting of wave motion, are energies normally transmitted to the receptor from a distant source. The latter makes possible “distance reception”, that is, the capacity to react specifically to far-off things. There can also be noted two special cases which are intermediate between exclusively local stimulating action and action at a distance. These are 2b and 4b. Volatile substances in the air which is breathed, odors, have a source. The odor may be carried to a considerable distance by molecular diffusion or by wind. Also warm or cold drafts of air may travel to considerable distances from their sources.
The six types of stimulus energy listed above may be thought of either as applications of energy to sensory surfaces or as variations of energy at sensory surfaces. In the latter meaning, the various forms of energy are carriers of stimulation rather than stimuli. This distinction will be developed later.
C. The Environmental “Fields” of Stimulation
The abstract physical environment contains matter and energy. The terrestrial environment, however, although reducible to the former, may be said to contain first, earth, air, water, other animals, and all sorts of objects, and second, heat, light, sound, odor, and the possibility of contact with the various substances , animals and objects. In other words the ordinary environment includes not only substantial things but stimuli for the sense organs. Can the possibilities – of – being – stimulated be classified and listed? Following is an attempt to do so, subject to correction and modification.
Fields of Continuous Contact. There seem to be three types of continuous or incessant exposure to stimulation:
1. Contact with the medium (air) having a definitive temperature, pressure, oxygen content, and other chemical composition;
2. Contact with the ground, or surface of support (upward pressure on the body); and
3. Weight of limbs and head, and of statolith (downward force of gravity).
Fields of Potential Contact Stimulation Depending on Proximity. There are a great many fields of stimulation attached to the material bodies of the environment which depend on close proximity or touch. Actual energy-exchange depends on the Location of the individual relative to the object question. The question is therefore a locus of potential stimulation. Since a living individual moves about in his environment, he is intermittently stimulated by (or stimulates himself with) these fields. There are two general, cases, touching, and being touched.
1. Contact or coalition with a fixed environmental object
a. Accompanied by heat-flow with a hot (or cold) substance
b. Accompanied by chemical action, with a reactive substance
2. Contact with a moving or an animate object
a. Accompanied by heat-flow
b. Accompanied by chemical action
Fields of Potential Stimulation not Depending on Contact. There are also exists in the environment fields of potential stimulation which are propagated from a fixed a source or center. They are fields in a more exact, not a figurative, sense of that term, since they decreased in concentration or intensity outward from the source. Some obey the inverse square law. There is thus a certain stimulating potential at every point in the field. Five types can be listed.
1. The odor field diffusing from a source of scent (modified by air movement)
2. The field of sound waves propagated in air from a mechanical disturbance (modified by air movement)
3. The field of heat radiated form a source
a. From the sun
b. From other heat source
4. The field of light radiated from a source
a. From the sun
b. From other heat sources
5. The gravitational field of forces toward the center of the earth (not a propagated field)
The fields of stimulation are not dependent on the immediate proximity of a receptive individual, like those of contact stimulation. Fields of vapor, sound, heat, and light existed on the face of the earth before animals were there to sense them, and so did the field of gravity. With reference to the world they are physical facts; with reference to an individual they are potential stimuli. They express what we mean by saying that the world is “full of” smells, sounds and lights. The terrestrial sources of odor, noise, heat and light (even from the sun) may wax and wane by the temporary islands of energy they produce are important components of the world.
Some objects and events propagate several types of potential stimulation at the same time. For example, a fire gives off odor, sound, heat, and light so that it can be simultaneously smelled, heard, felt, and seen.
The Field of Potential Luminous Stimulation at a Point in Space. The Optic Array. Odors, sounds, and radiant heat generally arrive at the station point of an observer from one direction, and the source can be localized, more or less, by appropriate sensory exploration. But light has the unique characteristic of arriving at a station point from all directions in straight lines. This is ambient light, not radiant light, and it depends on the diffuse multiple reflection of light from textured surfaces, that is, the reverberation of light waves which we call “illuminationÓ. The open spaces of a terrestrial environment are filled with a dense network of pencils of rays. Hence at any point in such a space there is a spherical pencil of rays consisting of different intensities of light in different directions. This I call an optic array. The optic array at a station point is an entirely different kind of field than any yet considered. The structured of texture of this array is the means by which the eyes are sensitive to the surface reflecting light, not just the light as such. The “source” of an optic array is the whole environment, not the emitter of energy.
An ambient field of stimulus energy accompanies the individual wherever he goes or, better, exists at whatever position in the environment he may take. A propagated field of stimulus energy is anchored to its source in the environment. Both are objective, but they are quite different sorts of potential stimuli.
It should be noted that the ambient optic array is a stimulus for the whole visual sensory system and its exploratory activity. The sector of an optic array entering an eye is the stimulus for accommodation and pupillary adjustment of the eye. The retinal image is the stimulus for the sensory surface of that eye, the retina. And a spot of the retinal image, considered as an application of energy, is the stimulus for one receptive element of that retina.
D. The Sources of Environmental Fields of Stimulation
The sources of the fields described above, the objects and events that emit potential stimuli, comprise the environment, or more precisely, the environment that can be registered by the senses. It consists of things intermediate in size between the atom and the universe. Its general components might be classified as follows:
1. The air, the surfaces of support, gravity
2. The fixed rigid environment (permanent places and objects)
3. The moving changing environment (events, processes, non-rigid surfaces).
The next two classes are partly “things” and partly “events”:
4. The animate environment (animals, people, their actions and expressions)
5. The cultural environment
a. Tools and artifacts
b. Surrogates for human communication
(1) Iconic surrogates (pictures and images)
(2) Coded surrogates (signs, signals, symbols, spoken or written)
E. The Inner and Outer Sources of Variation of Proximal Stimulation
It was suggested earlier that a very important kind of stimulus was a variant of energy at, as distinguished from an application of energy to, the sensory surface. Stimulus variations are important because they are presumably more or less specific to their causes. If differences in stimulation correspond to the differences between sources of stimulation then, figuratively speaking, they “carry information” about the sources. The information registered by the various types of sensory surface such as the statocyst, the skin, the joints, the retina, and the basilar membrane is in part different for each type of sensory surface and in part common to several of them. There is no clean-cut division between “inner” and “outer” sources of stimulation as we shall see.
With respect to the types of stimulus variation they are being actively investigated by experimenters. Abstractly, one kind is pattern variation at the sensory surface and another is temporal variation. The former is supposedly exemplified by the retinal image and the latter by the events in the cochlea, but actually pattern variation and temporal variation are combined, in different ways, at all sensory surfaces. The pattern of stimulation of hair-cells over the basilar membrane is crucial for hearing, and transformation of the pattern of the retinal image occurs in seeing. Pattern and sequence specify stimulation jointly, but they are not necessarily represented as such in the sensory experience. The temporal and spatial patterns of excitation at a bank of receptive elements have to be determined empirically in order to understand how the sense works. One thing is certain, namely, that spatiotemporal variations of receptor excitation are as important for joint sensitivity, taste and smell, or temperature sensitivity as they are for the eye and the ear.
The sources of spatiotemporal variations of stimulation are classified in a general way, even if the variations can not. One source is supposed to be the body and the other is the world. The somatic source yields what has been called kinaesthesis, or response-produced stimulation, or “feedback”. The external source yields what might be called environment-produced stimulation. But there is still another source of variation which has not been recognized. The individual can enhance the external stimulation by sensory adjustment, or by sensory exploration, or by locomotion approach to the center of a stimulus-field. Moreover, during locomotion the individual will transform optical stimulation by changing his station point, thus altering the perspective shape of everything visible.
In short self-produced stimulation is not of one type but of several. The individual cannot only register his motor actions (somatic source) but can also maximize or select external stimulation and can transform external stimulation. In these cases, the source is both somatic and external.
There are this at least four sources of stimulus variation. The following classification is tentative and subject to correction.
1. Change in the environment (external or exterospecific variation)
a. Change of illumination
b. Change of air temperature
c. Waxing and waning of stimulus-fields, from odorous, noisy, hot, or luminous events, and from light reflecting objects.
(1) Flow, deformation, growth, or decay of objects
(2) Transportations and rotations of objects
2. Muscular action of the body (classical kinaesthesis)
a. Motor behavior
b. Locomotor behavior
3. Sensory adjustment and exploration (stimulus-enhancement)
a. Getting closer to an object
b. Ocular responses and eye movement
c. Exploratory touching
d. Sniffing and tasting
e. Head turning (auditory localization)
4. Moving about in the world (stimulus transformation)
a. Approach (optical expansion)
b. Withdrawal (optical contraction)
c. Lateral movement (slant transformation)
Note that numbers 2, 3, and 4 might all be thought of as kinds of self-stimulation of proprioception, not just number 2. Although number 2 “controls” behavior in a certain sense, number 4 also does so, and number 1 may do so in another sense. Although number 1 is the obvious source of external stimulation, the actual stimuli depend also on number 3 and number 4.
Nothing has been said so far about nonvariation of stimulus energy at sensory surfaces, that is, constant stimulation. For many receptors, the steady application of a stimulus results in adaptation. For many receptors, the steady application of a stimulus results in adaptation. (The statocyst and the linings of the joints and perhaps the cochlea seem to be continuously excitable, however, and the retina normally averts adaptation by moving). When we consider not application but complex variation of stimulation, however, it becomes evident that there may be a constant or “invariant” stimulus underlying the variation. In nearly all the types of change listed above, something remains constant. The nonchange of stimulation results from a nonchanging feature of the environment. The registering of the constant components of stimulation is obviously just as important a function of the receptive systems as is the registering of the variant components.
Two examples may be given of a stimulus invariant which is specific to a permanent feature of the environment. First, the locus of pressure of a statolith on the surface of the statocyst, the locus of pressure of the ground on the skin, and the orientation of parallel lines on the retina are all normally linked to give a single invariant. They specify the direction “downward” during all changes of bodily posture. Their source is the earth. If they are experimentally unlinked, the subject becomes confused. Second, the textured or untextured character of an area of the optic array specifies a surface or a hole in the world, that is, an obstacle or a space, a thing or nothing. This difference, together with the quality of the texture which specifies the substance of which the physical surface is composed, is invariant under all transformations of the area. A solid object is specified at changing distances or points of view.
SummaryAn overall picture of the operation of the receptive mechanisms has been sketched. The sensory system as a whole does not consist of a number of mutually exclusive “senses”. Hence the classical modalities or “departments of sense” are in part fallacious, and the classes of “sense data” are questionable.
The simple distinction between exteroception and proprioception is insufficient. Response-produced stimuli are of several distinct types, not one.
The assumption that the “basic” variables of stimulation are intensity, frequency, duration, location, and extension is inadequate.
The distinction between “proximal” and “distal” stimuli is insufficient. There are sources of stimulation at more than one level of analysis. The specifying capacity of proximal stimuli, the “information” content, can be determined.