When we say that behavior is a function of the environment, the term “environment” presumably means any event in the universe capable of affecting the organism. But part of the universe is enclosed within the organism’s own skin. Some independent variables may, therefore, be related to behavior in a unique way. The individual’s response to an inflamed tooth, for example, is unlike the response which anyone else can make to that particular tooth, since no one else can establish the same kind of contact with it. Events which take place during emotional excitement or in states of deprivation are often uniquely accessible for the same reason; in this sense our joys, sorrows, loves, and hates are peculiarly our own. With respect to each individual, in other words, a small part of the universe is private.
We need not suppose that events which take place within an organism’s skin have special properties for that reason. A private event may be distinguished by its limited accessibility but not, so far as we know, by any special structure or nature. We have no reason to suppose that the stimulating effect of an inflamed tooth is essentially different from that of, say, a hot stove. The stove, however, is capable of affecting more than one person in approximately the same way. In studying behavior we may have to deal with the stimulation from a tooth as an inference rather than as a directly observable fact. But if some of the independent variables of which behavior is a function are not directly accessible, what becomes of a functional analysis? How are such variables to be treated?
These questions may not be of interest to all readers. The issue is an ancient one, which has occupied the attention of philosophers and others for more than two thousand years. It has never been satisfactorily resolved, and perhaps the present inclination on the part of educated laymen to avoid it represents simple extinction. Fortunately, the issue is seldom crucial in the practical control of human behavior. The reader whose interests are essentially practical and who may now prefer to move on to later chapters may do so without serious trouble. Nevertheless, the issue is important and must sometime be faced. Modern science has attempted to put forth an ordered and integrated conception of nature. Some of its most distinguished men have concerned themselves with the broad implications of science with respect to the structure of the universe. The picture which emerges is almost always dualistic. The scientist humbly admits that he is describing only half the universe, and he defers to another world—a world of mind or consciousness—for which another mode of inquiry is assumed to be required. Such a point of view is by no means inevitable, but it is part of the cultural heritage from which science has emerged. It obviously stands in the way of a unified account of nature. The contribution which a science of behavior can make in suggesting an alternative point of view is perhaps one of its most important achievements. No discussion of the implications of science for an understanding of human behavior would be complete without at least a brief review of this contribution.
The verbal response “red” is established as a discriminative operant by a community which reinforces the response when it is made in the presence of red stimuli and not otherwise. This can easily be done if the community and the individual both have access to red stimuli. It cannot be done if either the individual or the community is colorblind. The latter case resembles that in which a verbal response is based upon a private event, where, by definition, common access by both parties is impossible. How does the community present or withhold reinforcement appropriately in order to bring such a response as “My tooth aches” under the control of appropriate stimulation? It may easily establish the response “My tooth is broken” because both the individual and the community have access to the stimulus for “broken,” but the community has no comparable access to the stimulus eventually controlling “aches.” Nevertheless, verbal behavior of this sort is obviously set up.
The community may resort to public accompaniments of the private event. For example, it may establish a verbal response to an aching tooth by presenting or withholding reinforcement according to a special condition of the tooth which almost certainly accompanies the private event or according to violent collateral responses such as holding the jaw or crying out. Thus we teach a child to say “That itches” or “That tickles” because we observe either public events which accompany such private stimulation (“the kinds of things which itch or tickle”) or some such identifying response as scratching or squirming. This method of circumventing the privacy of the individual is not foolproof because the public and private events may not be perfectly correlated.
There is another possibility. Verbal responses which are acquired with respect to public events may be transferred to private events on the basis of common properties. It has often been pointed out that many subjective terms are metaphorical, at least in origin. The language of emotion, for example, is almost wholly metaphorical; its terms are borrowed from descriptions of public events in which both the individual and the reinforcing community have access to the same stimuli. Here again the community cannot guarantee an accurate verbal repertoire because the response may be transferred from public event to private event on the basis of irrelevant properties.
The techniques which guarantee the reliability of a verbal report cannot be brought to bear upon a private description. The science of introspective psychology met this difficulty whenever it departed from the study of responses to controllable stimuli. The psychologist can, for example, manipulate the color, brightness, or saturation of a spot of light in order to establish a sensitive verbal repertoire in his subject with respect to these properties. Such an experimental situation does not raise the problem of privacy at all. But establishing a comparable repertoire to distinguish between various “states of emotion,” for example, is a task of a very different sort. Unless the psychologist can manipulate the events reported during emotion as he manipulates the properties of a patch of light, he must resort to imperfect public accompaniments.
The layman also finds the lack of a reliable subjective vocabulary inconvenient. Everyone mistrusts verbal responses which describe private events. Variables are often operating which tend to weaken the stimulus control of such descriptions, and the reinforcing community is usually powerless to prevent the resulting distortion. The individual who excuses himself from an unpleasant task by pleading a headache cannot be successfully challenged, even though the existence of the private event is doubtful. There is no effective answer to the student who insists, after being corrected, that that was what he “meant to say,” but the existence of this private event is not accepted with any confidence.
The individual himself also suffers from these limitations. The environment, whether public or private, appears to remain undistinguished until the organism is forced to make a distinction. Anyone who has suddenly been required to make fine color discriminations will usually agree that he now “sees” colors which he had not previously “seen.” It is hard to believe that we should not distinguish between the primary colors unless there were some reason for doing so, but we are conditioned to do this so early in our history that our experience is probably not a safe guide. Experiments in which organisms are raised in darkness tend to confirm the view that discriminative behavior waits upon the contingencies which force discriminations. Now, self-observation is also the product of discriminative contingencies, and if a discrimination cannot be forced by the community, it may never arise. Strangely enough, it is the community which teaches the individual to “know himself.”
Some contingencies involving inner stimulation do not, of course, have to be arranged by a reinforcing community. In throwing a ball we time a sequence of responses by the stimulation which our own movements generate. Here the reinforcing contingencies are determined by the mechanical and geometrical exigencies of throwing a ball, and since a reinforcing community is not involved, the question of accessibility to the behaving individual does not arise. But “knowledge,” as we saw in Chapter VIII, is particularly identified with the verbal behavior which arises from social reinforcement. Apparently, conceptual and abstract behavior are impossible without such reinforcement. The kind of self-knowledge represented by discriminative verbal behavior—the knowledge which is “expressed” when we talk about our own behavior—is strictly limited by the contingencies which the verbal community can arrange. The deficiencies which generate public mistrust lead, in the case of the individual himself, to simple ignorance. There appears to be no way in which the individual may sharpen the reference of his own verbal repertoire in this respect. This is particularly unfortunate because he probably has many reasons for distorting his own report to himself (Chapter XVIII).
It is customary to distinguish between two types of internal stimulation. Interoceptive stimuli arise mainly in the digestive, respiratory, and circulatory systems. A full or inflamed stomach, a stomach contracting in hunger, a gallstone distending the bile duct, the contractions or relaxations of small blood vessels in blushing and blanching, and a pounding heart all generate interoceptive stimuli. These are the principal stimuli to which one reacts in “feeling an emotion.” Proprioceptive stimuli, on the other hand, are generated by the position and movement of the body in space and by the position and movement of parts of the body with respect to other parts. We usually respond to stimuli of this sort in combination with exteroceptive stimulation from the surrounding environment, and we do not always correctly identify the source of stimulation. Thus when we ran our hand over a surface and judge it to be sticky, gummy, or slippery, our response is in part to the resistance encountered in moving our hand, even though we appear to be talking about the surface as a public event. The important point here, however, is not the locus of stimulation but the degree of accessibility to the community.
An important verbal repertoire describes one’s own behavior. It is generated by a community which insists upon answers to such questions as “What did you say?” “What are you doing?” “What are you going to do?” or “Why are you doing that?” Although these questions are usually practical ones, the theoretical implications are equally important. Since the individual may often observe his own behavior as a public event, the public-private distinction does not always arise. In that case the accuracy of the self-descriptive repertoire may be adequate. If a man says, “I went home at three o’clock,” there are ways in which this may be checked and his behavior reinforced to insure future accuracy. But part of the stimulation which the individual receives from his own behavior is different from that available to the community.
A description of behavior which has not been executed appears to depend upon private events only. For example, a man may say, “I was on the point of going home at three o’clock,” though he did not go. Here the controlling stimuli are not only private, they appear to have no public accompaniments. Such responses as “I’m strongly inclined to go home” or “I shall go home in half an hour” also describe states of affairs which appear to be accessible only to the speaker. How can the verbal community establish responses of this sort?
A possible explanation is that the terms are established as part of a repertoire when the individual is behaving publicly. Private stimuli, generated in addition to the public manifestations, then gain the necessary degree of control. Later when these private stimuli occur alone, the individual may respond to them. “I was on the point of going home” may be regarded as the equivalent of “I observed events in myself which characteristically precede or accompany my going home.” What these events are, such an explanation does not say. Comparable expressions may describe the momentary probability of behavior as well as its particular form.
Another possibility is that when an individual appears to describe unemitted behavior, he is actually describing a history of variables which would enable an independent observer to describe the behavior in the same way if a knowledge of the variables were available to him. The question, “Why did you do that?” is often important to the community, which establishes a repertoire of responses based upon the external events of which behavior is a function, as well as upon the functional relation itself. We are usually able to report that a particular stimulating situation, a special contingency of reinforcement, a condition of deprivation, or some emotional circumstance is responsible for our own behavior: “I often drop in on X because he serves excellent drinks,” “I spanked the brat because he had been thoroughly annoying,” “I generally take the early train because it is less crowded,” and so on. It is possible that the same data may be used to predict our own future behavior. The statement, “I shall probably go abroad next summer,” may be due to variables of a wholly public nature which make it equivalent to the statement, “Circumstances have arisen which make it highly probable that I shall go abroad.” This is not a description of behavior-to-be-emitted but of the conditions of which that behavior is a function. The individual himself is, of course, often in an advantageous position for observing his own history.
One important sort of stimulus to which the individual may possibly be responding when he describes unemitted behavior has no parallel among other forms of private stimulation. It arises from the fact that the behavior may actually occur but on such a reduced scale that it cannot be observed by others—at least without instrumentation. This is often expressed by saying that the behavior is “covert.” Sometimes it is said that the reduced form is merely the beginning of the overt form—that the private event is incipient or inchoate behavior. A verbal repertoire which has been established with respect to the overt case might be extended to covert behavior because of similar self-stimulation. The organism is generating the same effective stimuli, albeit on a much smaller scale.
The appeal to covert or incipient behavior is easily misused. If the statement, “I was on the point of going home,” is a response to stimuli generated by a covert or incipient response of actually going home, how may the response of going home be executed covertly? In such a case, one of the other interpretations may well be preferred. Verbal behavior, however, can occur at the covert level because it does not require the presence of a particular physical environment for its execution. Moreover, it may remain effective at the covert level because the speaker himself is also a listener and his verbal behavior may have private consequences. The covert form continues to be reinforced, even though it has been reduced in magnitude to the point at which it has no appreciable effect on the environment. Most people observe themselves talking privately. A characteristic report begins “I said to myself . . . ” where the stimuli which control the response “I said” are presumably similar, except in magnitude, to those which in part control the response, “I said to him . . . ”
When a man says, “There is a rainbow in the sky” or “The clock is striking twelve,” we can give a reasonable interpretation of his behavior in terms of a stimulating situation and certain characteristic conditioning procedures with which the community has set up verbal responses. But if he says, “I see a rainbow in the sky” or “I hear the clock striking twelve,” additional terms have to be taken into account. Their importance is easily demonstrated. The group usually benefits when an individual responds verbally to events with which he alone is in contact. In so doing, he broadens the environment of those who hear him. But it is also important that he report the conditions under which he is responding. In so doing he reveals, so to speak, the “source of his information.” The response, “I see a rainbow in the sky,” is of a different order of importance from the response, “They say there is a rainbow in the sky.” Other reasons why the group may be interested in the nature of the behavior of the speaker are suggested by such familiar questions as “Do you see that man over by the window?” “Can you hear me?” or “Do you smell smoke?”
When the community conditions the individual to say, “I see . . . ” “I hear . . . ” “I smell . . . ” and so on, it must have some evidence of discriminative behavior. In certain cases it may rely upon the inevitability of a response to a conspicuous stimulus—“You see, it’s raining after all.” At other times it may rely upon the orientation of receptors: we tell a child that he is seeing a dog when we are sure that his eyes are oriented toward the dog, or that he is feeling the texture of a piece of cloth when we run his fingers over the cloth. But we cannot always or safely count upon evidence that a stimulus is merely being received. We have no comparable evidence for faint odors or tastes, or for visual or auditory stimuli to which receptors need not be especially oriented. How, for example, can the community teach the individual to report correctly that he is seeing the color of a piece of cloth or hearing the oboe in a full orchestra? Here there must be clear evidence that a discriminative reaction is being made. “Do you see that bird in the bush?” “Yes.” “What kind is it?” Only when collateral information is correctly given does the community appropriately reinforce the response “Yes.”
A verbal repertoire which describes the discriminative behavior of the individual appears, then, to be established on external evidence that a discriminative response is taking place, rather than that stimuli are present or received. When the individual comes to describe his own discriminative behavior, presumably he does so, at least initially, on comparable evidence. He observes himself as he executes some identifying response. The private events correlated with the public events used by the community are also the result of discriminative behavior, not simple stimulation. The response, “I see a rainbow,” is, therefore, not equivalent to “There is a rainbow.” If it were, a single discriminative stimulus—the rainbow—would account for both forms, but “I see a rainbow” is a description of the response of seeing a rainbow. When the rainbow is actually present, the distinction may be of little moment.
But the rainbow is not always present. Perhaps the most difficult problem in the analysis of behavior is raised by responses beginning “I see . . . ,” “I hear . . . ,” and so on, when the customary stimuli are lacking. Here an accurate formulation of responses which describe one’s own discriminative behavior is essential. We may approach this problem by surveying the circumstances under which a man “sees something.” Presumably these will also be the circumstances under which he says, “I see something.” (Parallel cases for “I hear . . . ,” “I taste . . . ,” need not be explicitly discussed.) No special problem is raised when an appropriate stimulus is present. We are also prepared for instances in which the stimulus is not the customary one but has enough in common with it to control the response. The process of abstraction also provides examples in which the complete stimulus is not available but of which an adequate account may nevertheless be given. When there are no stimuli present which resemble the usual stimuli, a response beginning “I see . . . ” must be explained in terms of conditioning. There are two major possibilities corresponding to the distinction between respondent and operant conditioning.
A man may see or hear “stimuli which are not present” on the pattern of the conditioned reflex: he may see X, not only when X is present, but when any stimulus which has frequently accompanied X is present. The dinner bell not only makes our mouth water, it makes us see food. In the Pavlovian formula we simply substitute “seeing food” for “salivating.” Originally both of these responses were made to food, but through a process of conditioning they are eventually made in response to the bell. When a man reports that the dinner bell makes him see food (he is more likely to say that it “reminds him of food” or “makes him think of food”), we may suppose that he is reporting a response which is similar to the response made in the presence of food. It is only an unfortunate tradition, apparently due to the Greeks, which leads us to ask what he is seeing in such a case. When he reports that the bell makes his mouth water, we do not feel compelled to ask what he is salivating to. A stimulus function has been assumed by a different stimulus, which may control seeing food as well as salivating.
The effect of a conditioned stimulus in evoking the response of seeing something helps to explain the character of responses to stimuli which are present but which are at variance with “what is seen.” Conditioned seeing may combine with responses to unconditioned stimuli. We see familiar objects more readily and easily than unfamiliar objects; the stimuli actually present upon a given occasion may be effective both as conditioned and unconditioned stimuli at the same time. In catching only a passing glimpse of a bird, we see it distinctly if it is a familiar bird and indistinctly if it is not. The fragmentary stimuli have served to evoke conditioned seeing, which combines with the unconditioned seeing of the immediate stimulus. A poetic description of the sound of the sea is especially effective if one reads it while listening to the sea, for the verbal and nonverbal stimuli combine to produce an especially strong response. In a pack of playing cards, the shape of a heart or diamond is correlated with the color red. While playing cards one is especially likely to see a heart or diamond, rather than a spade or club, if one catches a glimpse of red. The verbal stimulus “heart” is likely to evoke seeing red as well as seeing a heart. It has been shown experimentally that if one who is familiar with playing cards is very briefly shown a heart printed in black ink, the heart is sometimes seen as red or as a mixture of red and black, perhaps reported as purple. If the card remains in view for a longer time, the current stimulus will completely mask the conditioned response of seeing red, but a brief exposure of appropriate duration leads to a fusion of conditioned and unconditioned responses.
In more general terms, conditioned seeing explains why one tends to see the world according to one’s previous history. Certain properties of the world are responded to so commonly that “laws of perception” have been drawn up to describe the behavior thus conditioned. For example, we generally see completed circles, squares, and other figures. An incomplete figure presented under deficient or ambiguous circumstances may evoke seeing a completed figure as a conditioned response. For example, a ring with a small segment missing when very briefly exposed may be seen as a completed ring. Seeing a completed ring would presumably not be inevitable in an individual whose daily life was concerned with handling incompleted rings—as might be the case in manufacturing certain types of piston rings, for example. Some of the so-called synesthesias are also examples of a fusion of conditioned and unconditioned seeing. In a common example numbers are seen as colored. Something of this sort could arise if a child first learned to respond to numbers in a book in which geometric form and color were systematically paired. The geometric form would then lead to the conditioned response of seeing the corresponding color. The spoken stimulus “Seven” would lead to two conditioned responses: seeing the form 7 and seeing the associated color.
All those circumstances under which a mature individual will exhibit the response of seeing something may be arranged in a continuum. At one extreme the momentary stimulation is optimal. If, for example, the individual is listening to a stormy sea, the sound is primarily in control. “Hearing the sea” is not a wholly unconditioned response, however, since it depends upon previous experience. If we now reduce the momentary stimulation by transporting our individual farther and farther from the sea, we increase the role played by conditioned stimuli. A faint distant roar is heard “as the sound of the sea” only because of a particular history. Any sound similar to that of the sea may have this effect—for example, that of traffic in the street. If we now begin to introduce conditioned stimuli of a clearly different form—for example, nonauditory stimuli—we may be able to show the fusion of two distinct effects. If our subject examines a picture of heavy surf, current auditory stimuli resembling the sound of surf will make the total response of seeing and hearing the sea more powerful. At the other extreme of our continuum is the purely conditioned response—hearing appropriate sounds in a quiet room while observing a painting of the sea. If such an effect occurs, it must be due to conditioning, since what is heard is an auditory stimulus but what is present is visual.
There are, of course, great differences in the extent to which individuals exhibit conditioned seeing, hearing, and so on. Francis Galton first surveyed this form of human behavior in the nineteenth century. Some of his subjects showed an exceptional ability to see things which he described to them, while others found it almost impossible. Some subjects showed special abilities in certain fields only. Congenital defects of sensory equipment are sometimes responsible—as, for example, in the color-blind or tone-deaf. Other individual differences may be traced to the histories of the individuals. One difference depends upon the extent to which the requisite conditioning has taken place. In a world in which visual stimuli are extremely important one would expect many conditioned responses of this sort to be set up. It is not surprising to find that the composer is especially likely to be able to “hear music which is not present,” while the artist is especially able to “see forms which are not present,” and so on. It is possible, of course, that a man may become an artist or musician because of special abilities of this sort, but the obvious differences in personal history are almost certainly relevant. Another difference depends upon whether the individual is able to respond to his conditioned discriminative responses, and this in turn depends upon whether the community has forced verbal responses to them. A society which breeds introspective people would probably have more data of this sort to account for, not because more private seeing occurs, but because more of it comes into the public domain through self-description. In a group which seldom insists upon such behavior, the problem might never arise at all.
When an individual reports that he sees an object which is actually before him, we can distinguish between his response to the object and his response to his response. The individual himself makes the same distinction. It is usually possible for him to say that there is or is not a rainbow present when he reports that he sees one and that this is the variable of which his behavior is primarily a function. When the stimulus only partially resembles the usual stimulus, the subject may report that it “reminds him” of it. When the “stimulus seen” is actually lacking but the subject cannot report that fact, we say that he is suffering from a hallucination. He sees something and reports that he sees something, and from these events alone he may assert that something is there. When the current situation is further clarified, he may revise this report and conclude that he “only thought” he saw it. On the other hand he may refuse to make a response to the current situation which is incompatible with his conditioned response and may insist that what he sees is “really there.” There are certain areas in which a collateral check on the presence or absence of an appropriate stimulus is not easily made. In such a case, we are much less likely to insist upon the distinction. Since we do not ordinarily confirm the presence or absence of bitter substances in the mouth, we are not likely to argue that the response, “My mouth tastes bitter,” is hallucinatory.
The practical importance of conditioned seeing. A private event is not wholly without practical importance. Stimuli which generate conditioned seeing are often reinforcing because they do so, and they extend the range of reinforcing stimuli available in the control of human behavior. The practical task of generating conditioned stimuli of special effectiveness is an important one, as the artist, writer, and composer know. If it is possible to reinforce a man with the “beauties of nature,” it is usually possible to reinforce him also with conditioned stimuli which evoke responses of seeing the beauties of nature. It is the function of the “word picture” to generate such conditioned seeing. By fusing conditioned and unconditioned seeing the artist makes the observer see the same thing in another way. Nostalgic music is effective if it “reminds one” of happier days, a return to which would also be reinforcing. The extent to which this process is used in art varies from period to period but is always considerable. It is not to be identified with realism or naturalism since the responses appropriate to the effect of pure design are also largely dependent upon experience. We shall return later to other practical applications of conditioned seeing, hearing, and so on. In evaluating the effect of a given culture it is important to note the extent to which conditioned responses of this sort are set up and the extent to which discriminative responses of self-knowledge are established with respect to them.
There are many ways of showing that the discriminative operant “seeing X” is strong. One kind of evidence is the strength of precurrent behavior which makes seeing X possible. This may be nothing more than the behavior of looking at X, which the individual may engage in at every opportunity or for long periods of time. Another sort of precurrent behavior is looking for X—looking about in ways which in the past have led to seeing X. Suppose we strongly reinforce a person when he finds a four-leaf clover. The increased strength of “seeing a four-leaf clover” will be evident in many ways. The person will be more inclined to look at four-leaf clovers than before. He will look in places where he has found four-leaf clovers. Stimuli which resemble four-leaf clovers will evoke an immediate response. Under slightly ambiguous circumstances he will mistakenly reach for a three-leaf clover. If our reinforcement is effective enough, he may even see four-leaf clovers in ambiguous patterns in textiles, wallpaper, and so on. He may also “see four-leaf clovers” when there is no similar visual stimulation—for example, when his eyes are closed or when he is in a dark room. If he has acquired an adequate vocabulary for self-description, he may report this by saying that four-leaf clovers “flash into his mind” or that he “is thinking about” four-leaf clovers.
We frequently observe strong behavior without knowing much about the circumstances which account for its strength. Consider, for example, a person who is interested in dogs. One characteristic of such a person is that the response “seeing dogs” is especially strong. He looks at dogs at every opportunity and engages in behavior which makes it possible to do so—for example, he visits kennels and dog shows. He arranges stimuli which resemble dogs—he hangs pictures of dogs on his walls, puts statues of dogs on his desk, and buys books containing pictures of dogs. If he is an artist, photographer, or sculptor, he may create similar pictures or statues himself. But the presence of a dog or of a reasonable facsimile is not essential. Conditioned stimuli which have accompanied dogs—leashes, feeding equipment, and so on—easily “remind him of dogs.” Certain verbal stimuli—for example, stories or descriptions of dogs—lead him to “picture dogs to himself,” and he may obtain or even compose such stimuli. The same strength is manifested when he sees dogs while looking at ink blots, cloud formations, or other ambiguous patterns, or when he mistakes some indistinctly seen object for a dog. The behavior of seeing dogs also takes place in the absence of any identifiable external support. He “thinks about dogs,” daydreams of dogs, and perhaps even dreams of dogs at night.
Unlike conditioned seeing in the respondent pattern, such behavior is not elicited by current stimuli and does not depend upon the previous pairing of stimuli. The primary controlling variables are operant reinforcement and deprivation. When we make a man hungry, we strengthen practical responses which have in the past been reinforced with food. We also strengthen artistic or verbal responses which produce pictures of food, or generate conditioned stimuli which are effective because they have accompanied food—the individual draws pictures of food or talks about delicious meals he has eaten. At the same time we induce him to “think of food,” to daydream of food, or to dream of food. Similarly, it is characteristic of men under strong sexual deprivation, not only that they indulge in sexual behavior as soon as an occasion presents itself or concern themselves with the production or enjoyment of sexual art or engage in sexual self-stimulation, but that they also see sexual objects or activities in the absence of relevant stimuli. That all these forms of activity are traceable to a common variable is shown by eliminating the deprivation, whereupon we eliminate all forms of the behavior.
A discriminative response which can be made when the appropriate stimulus is absent has certain advantages. It does not require the sometimes troublesome precurrent behavior which generates an external stimulus, and it can occur when such behavior is impossible—as when we daydream of a lost love or an opportunity which is wholly out of the question. Also, the private response is not punished by society, even though the overt form may be. There are, however, certain disadvantages. Such behavior does not alter the state of deprivation. The fantasies of the hungry or sexually deprived man do not alter the situation in such a way as to reduce the strength of the behavior through satiation. We often appeal to a reduction in deprivation to account for the effectiveness of a reinforcement, but, as we saw in Chapter V, the relation explains only why such stimuli are currently reinforcing in a given species. The reinforcing effect is carried by private as well as public stimuli. To one who is interested in dogs, simply seeing dogs is automatically reinforcing. The hungry or sexually deprived man is reinforced by the appearance or presence of relevant objects, as well as by seeing them when they are absent. Such reinforcement is not dependent upon an actual reduction in the state of deprivation.
Operant seeing at the private level may be reinforced in other ways. The private response may produce discriminative stimuli which prove useful in executing further behavior of either a private or public nature. In the following problem, for example, behavior is usually facilitated by private seeing. “Think of a cube, all six surfaces of which are painted red. Divide the cube into twenty-seven equal cubes by making two horizontal cuts and two sets of two vertical cuts each. How many of the resulting cubes will have three faces painted red, how many two, how many one, and how many none?” It is possible to solve this without seeing the cubes in any sense—as by saying to oneself, “A cube has eight corners. A corner is defined as the intersection of three faces of the cube. There will therefore be eight pieces with three painted faces . . . . ” And so on. But the solution is easier if one can actually see the twenty-seven small cubes and count those of each kind. This is easiest in the presence of actual cubes, of course, and even a sketchy drawing will provide useful support; but many people solve the problem visually without visual stimulation.
Private problem-solving usually consists of a mixture of discriminative and manipulative responses. In this example one may see the larger cube, cut it covertly, separate the smaller cubes covertly, see their faces, count them subvocally, and so on. In mental arithmetic one multiplies, divides, transposes, and so on, seeing the result in each case, until a solution is reached. Presumably much of this covert behavior is similar in form to the overt manipulation of pencil and paper; the rest is discriminative behavior in the form of seeing numbers, letters, signs, and so on, which is similar to the behavior which would result from overt manipulation.
There are great individual differences in the extent to which private seeing is used. Few people can equal the performance of one of Galton’s correspondents who could multiply by visualizing the appropriate section of a slide rule, setting it at the appropriate position, and reading off the answer. As in conditioned seeing, such differences may be traced either to differences in the extent to which private seeing has been established or to differences in the ability to describe the resulting self-stimulation or use it as a basis for further behavior.
There are also differences in the kind of private event preferred. In solving a chess problem, one may have an idea, in the sense of Chapter XVI, in several ways. The solution may come as the overt response of moving a piece. It may come in overt verbal form (“Move the knight to bishop seven”) or in the same form covertly. It may also come as covert nonverbal behavior, although it is admittedly hard to determine the dimensions of such a response. We commonly say, “I said to myself, ‘Move the knight,’” but we have no comparable idiom of the form, “I moved the knight to myself.” The solution may also come in the form of a discriminative reaction: we may suddenly see the knight in its new position.
Even when covert behavior is mainly verbal, other types of private responding frequently occur. Some writers report that they first hear sentences, which they then record just as they would record the speech of another person. Others execute sentences subvocally in an obviously muscular form. There are instances in which, particularly in dreams, a writer first reads a poem or story, which he then transcribes. The poet deals primarily in verbal behavior, but he may be a “seer” who resorts to words only to describe what he has seen, just as he would describe a public event.
Similar differences arise when there is some measure of external stimulation. In the Verbal Summator experiment, for example, some subjects, listening to faint speech patterns, hear the phonograph saying something. Others find themselves saying something, in which case they may also, of course, “hear” their own verbal behavior. There is commonly no parallel in the nonverbal projective tests. In the Rorschach Test, the effect of the ink blot is primarily to supplement a visual discriminative response. What is revealed is the strength of seeing X, not of saying “X.” The verbal report is usually a response to the discriminative visual response.
The verbal repertoire which describes private events may fail to distinguish between these cases. If we ask someone to think of the number seven and he reports that he has done so, he may be reporting a discriminative response in which he has seen the form 7 or the word “seven,” or some spatial arrangement of seven spots, seven subdivisions of a line, and so on. But the same report may describe the fact that he has said “seven” to himself or has drawn the form 7 covertly. In this case the report may also include the fact that he has heard himself saying “seven” or has seen the result of the nonverbal response. It is possible that more than one, or even all, of these activities may occur when one “thinks of the number seven.” The community does not insist upon a distinction among them because a distinction is usually of little importance. Usually the variables which strengthen the discriminative response of seeing an object are also those which strengthen the covert or overt responses which produce the object. If hearing X is strong, saying X will probably be strong also, since saying X is a precurrent response which makes hearing X possible. This is obvious but none the less important. It is often reinforcing to hear oneself praised. A simple expedient is, therefore, to praise oneself. Boasting is, so to speak, reinforced by the praise which one hears. Under the same conditions of motivation one may also demonstrate a high probability of hearing praise—for example, one may simply listen closely when one is being praised, or interpret an overheard flattering remark as applying to oneself, or misinterpret a neutral remark as praise.
Private discriminative responses are also reinforced by their effect in self-control. With the exception of physical restraint all the variables which one may manipulate in self-control are available at the private level (Chapter XV). One may generate an emotional response by recounting an emotional event or by simply seeing or hearing it. One may generate an aversive condition through a verbal description of punishment or through seeing or hearing the punishment again.
We account for verbal behavior which describes the discriminative response of seeing X in the following way. Such behavior is acquired when the organism is not only in the presence of X but actually making a discriminative response to X. A similar discriminative response may come to occur in the absence of X as the result of respondent or operant conditioning. The verbal response which describes the discriminative response is not inevitable, but whenever it occurs, the same variables may be assumed to be active. We have not altered the inaccessibility of the private event by this treatment, but we have succeeded in bringing the behavior which describes the event under some sort of functional control.
This is not, of course, the traditional solution to the problem of private seeing. It is usually held that one does not see the physical world at all, but only a nonphysical copy of it called “experience.” When the physical organism is in contact with reality, the experienced copy is called a “sensation,” “sense datum,” or “percept”; when there is no contact, it is called an “image,” “thought,” or “idea.” Sensations, images, and their congeries are characteristically regarded as psychic or mental events, occurring in a special world of “consciousness” where, although they occupy no space, they can nevertheless often be seen. We cannot now say with any certainty why this troublesome distinction was first made, but it may have been an attempt to solve certain problems which are worth reviewing.
There are often many ways in which a single event may stimulate an organism. Rain is something we see outside our window or hear on the roof or feel against our face. Which form of stimulation is rain? It must have been difficult to suppose that any one discriminative response could identify a physical event. Hence it may have been tempting to say that it identified a transient but unitary sensation or perception of the event. Eventually the least equivocal form—stimulation through contact with the skin—became most closely identified with reality. A form vaguely seen in a darkened room was not “really there” until one could touch it. But this was not a wholly satisfactory solution. Stimulation arising from contact may not agree perfectly with that arising visually or audibly, and we may not be willing to identify one form with reality to the exclusion of the others. There still are psychologists, however, who argue for the priority of one form of stimulation and, hence, insist upon a distinction between experience and reality. They are surprised to find that “things are not what they seem” and that a room which looks square from a given angle may be found upon tactual or visual exploration to be askew. This difficulty offers no particular problem here. It is obvious that a single event may stimulate an organism in many ways, depending upon the construction of the organism and its capacity to be stimulated by different forms of energy. We are much less inclined today to ask which form of energy is the thing itself or correctly represents it.
Another problem which the distinction between physical and non-physical worlds may have been an attempt to solve arises from the fact that more than one kind of response may be made to stimulation arising from a physical event. Rain is something you may run to escape from, catch in your hands to drink, prepare crops to receive, or call “rain.” Which response is made to “rain in itself”? The solution was to construct a passive comprehension of rain, which was supposed to have nothing to do with practical responses. So far as we are concerned here, the problem is disposed of by recognizing that many verbal and nonverbal responses may come under the control of a given form of stimulation. With the possible exception of the abstract verbal response, no behavior need be singled out as “knowing rain.”
The process of abstraction raises another difficulty from which the concept of experience may have provided escape. We saw in Chapter VIII that the referent or meaning of the response “rain” could not be identified by examining a single occasion upon which the response was made. Certain properties of a class of stimuli control such a response, and they can be revealed only by a systematic investigation of many instances. Upon any given occasion, the response appears to be relatively free of the exigencies of the physical world and to deal with a single dimension abstracted from it. The fact that the process of abstraction appears to generate a world composed of general properties, rather than of particular events, has led, however, to inconsistent interpretations. On the one hand the particular event has been regarded as immediate experience, while the process of abstraction has been said to construct a physical world which is never directly experienced. On the other hand the single occasion has been viewed as a momentary unanalyzed contact with reality, while systematic knowledge of the world has been identified with experience.
Still another difficulty which must have encouraged the distinction between two worlds was the inadequacy of early physical science. How could the individual make contact with a world which lay well beyond his skin? It was some comfort to suppose that one never knew more than one’s own experience, which could be conceived of as existing within one’s body. And if one never sees the real world but only an imaginal copy of it, then it is not difficult to account for instances in which the something seen is not there in the real world at all. We have only to assume that experience is independent of reality. To say that one sees the sensation of a thing when the thing itself is far away appears to solve the problem of the location of what is seen. To say that one sees an image of the thing when the thing itself is absent appears to solve the problem of the existence of what is seen. But the solutions are spurious. One still has to explain how the distant thing can generate the sensation or how an image can occur when the thing is not present. Modem physical science solved the first of these problems by bridging the gap between the distant object and the organism. A study of behavior solved the second by pointing to variables which lead the organism to see X in the absence of X.
Objections to the traditional view. There is scarcely any need to point out the disadvantages of terms which refer to supposed nonphysical events. Even if it were possible to define “sensation” and “image” in dimensions acceptable in a natural science, they would appear as intervening concepts comparable to “drive,” “habit,” “instinct,” and so on, and would be subject to the criticism of such concepts presented in Section II. As usual the fictional explanation has offered unwarranted consolation in the face of difficult problems. By suggesting a type of causal event the practice has discouraged the search for useful variables. Contrary to the usual view, the special contact between the individual and the events which occur within his own body does not provide him with “inside information” about the causes of his behavior. Because of his preferred position with respect to his own history, he may have special information about his readiness to respond, about the relation of his behavior to controlling variables, and about the history of these variables. Although this information is sometimes erroneous and, as we shall see in Chapter XVIII, may even be lacking, it is sometimes useful in a science of behavior. But the private event is at best no more than a link in a causal chain, and it is usually not even that. We may think before we act in the sense that we may behave covertly before we behave overtly, but our action is not an “expression” of the covert response or the consequence of it. The two are attributable to the same variables.
A recent book on abnormal behavior contains the sentence, “A system of emancipated ideas temporarily seizes control of behavior.” The facts are as well described by saying, “A system of responses is temporarily prepotent.” In either case we have still to ask, “Why?” Even though something which may properly be called an idea precedes the behavior in a causal chain, we must go back farther than the idea to find the relevant variables. If the individual himself reports, “I have had the idea for some time but have only just recently acted upon it,” he is describing a covert response which preceded the overt. Since someone who reports “having an idea” is likely to be someone who will execute the overt form, we may find the report of an idea helpful. But the report does not complete a functional account. As we saw in Chapter X, to say that a man strikes another because he feels angry still leaves the feeling of anger unexplained. When we have once identified the relevant variables, we find the feeling of anger much less important by way of explanation. Similarly, it has often been argued that the conditioned reflex is inadequate because it omits mention of a link traditionally described as the “association of ideas.” To report that a man salivates when he hears the dinner bell may be to overlook the fact that the dinner bell first “makes him think of dinner” and that he then salivates because he thinks of dinner. But there is no evidence that thinking of dinner, as that expression has been defined here, is more than a collateral effect of the bell and the conditioning process. We cannot demonstrate that thinking of dinner will lead to salivation regardless of any prior event, since a man will not think of dinner in the absence of such an event.
One is still free, of course, to assume that there are events of a nonphysical nature accessible only to the experiencing organism and therefore wholly private. Science does not always follow the principle of Occam’s razor, because the simplest explanation is in the long run not always the most expedient. But our analysis of verbal behavior which describes private events is not wholly a matter of taste or preference. We cannot avoid the responsibility of showing how a private event can ever come to be described by the individual or, in the same sense, be known to him. Our survey of the ways in which a community may impart a subjective vocabulary did not reveal any means of setting up a discriminative response to privacy as such. A world of experience which is by definition available only to the individual, wholly without public accompaniment, could never become the discriminative occasion for self-description.
Studying one’s own private world. It is sometimes suggested that the psychologist can avoid the problem of privacy by limiting his study to his own private share of the universe. It is true that psychologists sometimes use themselves as subjects successfully, but only when they manipulate external variables precisely as they would in studying the behavior of someone else. The scientist’s “observation” of a private event is a response to that event, or perhaps even a response to a response to it. In order to carry out the program of a functional analysis, he must have independent information about the event. This means he must respond to it in some other way. For a similar reason he cannot solve the problem of private events in the behavior of others by asking them to describe such events. It has often been proposed that an objective psychology may substitute the verbal report of a private event for the event itself. But a verbal report is a response of the organism; it is part of the behavior which a science must analyze. The analysis must include an independent treatment of the events of which the report is a function. The report itself is only half the story.
The physiology of sensation. The solution which follows from a functional analysis of behavior is to be distinguished from two others which are currently proposed within the framework of a natural science. One of these is closely identified with the study of the physiology of receptors in the nervous system, the other with a logical or “operational” analysis of the data of sensation and perception. Such concepts as “sensation” and “image” are designed to carry the pattern of the environment into the organism as far as possible and thus to bridge the gap between the knower and the known. The task of bringing the world to the surface of the organism is properly within the scope of physics. Beyond this point it is within the field of psychophysiology. The modem counterpart of the study of mental events in a world of consciousness is the study of the action of receptors and of the afferent and central nervous systems. The rainbow in the sky or some correlated pattern of energy is brought to the outer surface of the eye, then to the retina, then to the optic tract, and eventually to certain parts of the brain—preferably with as little distortion as possible. This makes it more plausible to say that the organism directly experiences the principal features of the rainbow. It is even tempting to suppose that at some stage (presumably the last) the pattern in the brain is the sensation or image. But seeing is a response to a stimulus rather than a mere camera-like registering. In carrying the pattern of the rainbow into the organism, almost no progress is made toward understanding the behavior of seeing the rainbow. It is of little moment whether the individual sees the actual rainbow or the sensation of a rainbow or some terminal neural pattern in the brain. At some point he must see, and this is more than recording a similar pattern. Apart from the mode of action of receptors and other organs, the physiology of sensation is concerned with the question of what is seen. The question may be a spurious one arising from an idiom or figure of speech. If we say that the rainbow (either as an objective event in the environment or as a corresponding pattern within the organism) is not “what is seen” but simply the commonest variable which controls the behavior of seeing, we are much less likely to be surprised when the behavior occurs as a function of other variables.
Operational definitions of sensation and image. Another proposed solution to the problem of privacy argues that there are public and private events and that the latter have no place in science because science requires agreement by the members of a community. Far from avoiding the traditional distinction between mind and matter, or between experience and reality, this view actually encourages it. It assumes that there is, in fact, a subjective world, which it places beyond the reach of science. On this assumption the only business of a science of sensation is to examine the public events which may be studied in lieu of the private.
The present analysis has a very different consequence. It continues to deal with the private event, even if only as an inference. It does not substitute the verbal report from which the inference is made for the event itself. The verbal report is a response to the private event and may be used as a source of information about it. A critical analysis of the validity of this practice is of first importance. But we may avoid the dubious conclusion that, so far as science is concerned, the verbal report or some other discriminative response is the sensation.
The private made public. One other way of attacking the problem within the framework of a natural science is compatible with the present analysis. The line between public and private is not fixed. The boundary shifts with every discovery of a technique for making private events public. Behavior which is of such small magnitude that it is not ordinarily observed may be amplified. Covert verbal behavior may be detected in slight movements of the speech apparatus. Deaf-mutes who speak with their fingers behave covertly with their fingers, and the movements may be suitably amplified. There is no reason why covert behavior could not be amplified so that the individual himself could make use of the additional information—for example, in creative thinking. After all, this is only what the individual does when he thinks publicly by scratching notes on paper or by manipulating an artistic medium. The problem of privacy may, therefore, eventually be solved by technical advances. But we are still faced with events which occur at the private level and which are important to the organism without instrumental amplification. How the organism reacts to these events will remain an important question, even though the events may some day be made accessible to everyone.