Response and Cognition I: The Specific-Response Relation
Edwin B. Holt
THE novelty exhibited by things at the moment of their synthesis into an organized whole has been frequently commented on. Such moments may seem to be "critical," as when two gases condense into a liquid and the phenomena exhibited by gases are replaced by those characteristic of fluids ; or there may be less appearance of discontinuity, as when a solid is slowly dissolved in a liquid and the latter as slowly acquires new properties. In either case, separate entities have been organized into a new whole, and their former action as independent parts is now combined in the action of the whole. And while it is obvious that the whole is nothing more than the parts as thus organized, and that the properties of the whole are nothing more than the properties of the parts now acting in cooperation, it is nevertheless true that the whole now does things which the isolated parts never did or could do. New phenomena, new laws and functions have been developed.
Most of us believe that the appearance of life was such a critical moment in the evolution of the universe : that life came into existence when some, perhaps a specific, sort of chemical process was set up under such conditions as maintained it around a general-point of equilibrium. The result was undoubtedly novel ; and more novelties were to come. Living substance was to acquire a protective envelope, to become irritable, conductive and contractile, to develop specific irritability to many stimuli, to get the power of locomotion, and much else. Now in the course of this further evolution, there is a critical point which is worthy of notice. It is the point where the irritable, contractile, and conductive tissues develop systematic relations which enable them to function as an integral whole. Here, too, novelty ensues.
How "critical" this point is, how sudden and well-defined, or, on the other hand, how gradual, can not as yet be told. The integrative process in the nervous system, as Sherrington so well calls it, has not, even yet, been observed in sufficient detail. But this is of secondary importance; and the result of the process we do know definitely.
( 366) This is, that the phenomena evinced by the integrated organism are no longer merely the excitation of nerve or the twitching of muscle, nor yet the play merely of reflexes touched off by stimuli. These are all present and essential to the phenomena in question, but they are merely components now, for they have been integrated. And this integration of reflex-arcs, with all that they involve, into a state of systematic interdependence has produced something that is not merely reflex action. The biological sciences have long recognized this new and further thing, and called it "behavior."
Of recent years, many of the workers in animal psychology have been coming to call this the science of behavior, and have been dwelling less and less on the subject of animal "consciousness." They do not doubt, any of them, that at least the higher animals are "conscious"; but they find that nothing but the behavior of animals is susceptible of scientific observation. Furthermore, several students in the human field have come to the same conclusion that not the "'consciousness," but the behavior of one's fellow men, and that alone, is open to investigation. Several volumes have been put forth which even undertake to construe human psychology entirely in terms of behavior. It is obvious that this is an unstable condition in which the science now finds itself. We can not continue thus, each man proclaiming his own unquestionable gift of "consciousness," but denying that either his fellow men or the animals evince the slightest indication of such a faculty. Now I believe that a somewhat closer definition of "behavior" will show it to involve a hitherto unnoticed feature of novelty, which will throw light on this matter.
Precisely how does this new thing, "behavior," differ, after all, from mere reflex action? Can not each least quiver of each least muscle-fiber be wholly explained as a result of a stimulus impinging on some sense-organ, and setting up an impulse which travels along definite nerves with definite connections, and comes out finally at a definite muscle having a certain tonus, etc., all of which is merely reflex action? Yes, exactly; each least component can be so explained, for that is just what, and all that, it is. But it is the coordinated totality of these least components which can not be described in such terms, nor indeed in terms resembling these. For such neural and reflex terms fail to seize that integration factor which has now transformed reflex action into something else, i. e., behavior. We require, then, an exact definition of behavior.
But before proceeding to this definition we shall probably find useful an illustration from another science, which was once in the same unstable state of transition as psychology is now. In physics a theory of causation once prevailed, which tried to describe causal process in terms of successive "states,"the "state"of a body at one
( 367) moment being the cause of its "state" and position at the next, Thus the course of a falling body was described as a series of states (a, b, c, d, etc.), each one of which was the effect of the state preceding, and cause of the one next following. This may be designated as the "bead theory" of causation. Inasmuch, however, as nothing could he observed about one of these "states" which would show why the next "state" must necessarily follow, or, in other words, since the closest inspection of "states" gave no clue toward explaining the course or even the continuance of the process, an unobservable impetus (vis viva, Anstoss, "force") was postulated. This hidden impetus was said to be the ultimate secret of physical causation. But, alas, a secret ! For it remained, just as the "consciousness" of one's fellow man remains to-day in psychology,[1] utterly refractory to further investigation. Now "myth" is the accepted term to apply to an entity which is believed in, but which eludes empirical inquiry. This mythical vis viva has now, in good part, owing to the efforts of Kirchhoff and Hertz, been rejected, and, what is more important, with it has gone the bead theory itself. It is not the "previous state" of the falling body which causes it to fall, but the earth's mass. No laws, in which alone explanation resides, for falling bodies or for any other process could, on the terms of the bead theory, be extracted from the phenomena. But laws were easily found for physical processes, if the observer persuaded himself to make the simple inquiry, What are the objects doing? [2] Now the falling body is not merely moving downwards past the successive divisions of a meter-stick which I have placed beside it (which is all that the bead theory would have us consider), nor is it essentially moving toward the floor which, since a floor happens to be there, it will presently strike. The body is essentially moving toward the center of the earth, and these other objects could be removed without altering the influence of gravity. In short, the fall of a body is adequately described as a function of its mass, of the earth's mass, and of the distance between the centers of the two. And the function is constant, is that which in change remains unchanged (in the case cited, it is a constant acceleration). The physical sciences, of course, have now explicitly adopted this function theory of causation.[3] Every physical law is in the last analysis the statement of a constant function between one process or
( 368) thing and some other process or thing. This abandonment of the bead theory in favor of the function theory requires, at the first, some breadth and some bravery of vision.
Now psychology is ,at the present moment addicted to the bead theory, and I believe that this is responsible for the dispute about "consciousness" versus behavior. Our disinclination to follow the physical sciences, to adopt the functional view in place of the bead theory, has hindered us from defining accurately what behavior is, and this has prevented us from recognizing a remarkable novelty which is involved in behavior, and which is the result of reflex action becoming organized.
We are prone, even the "behaviorists" among us, to think of behavior as somehow consisting of reflex activities. Quite true so far as it goes. So, too, coral reefs in the last analysis consist of positive and negative ions, but the biologist, geographer, or sea-captain would miss his point if he conceived them in any such terms. Yet we are doing the very same thing when we conceive the behavior of a man or animal in the unintegrated terms of neural process; which means, agreeably to the bead theory, the impinging of stimulus on sense-organ, the propagation of ionization waves along a fiber, their spread among various other fibers, their combining with other similar waves, and eventually causing the lowered or heightened tonus of muscle. All this is happening. But our account has overlooked the most essential thing of all--the organization of these processes.
If now we pitch the misleading bead theory straight overboard, and put our microscope back into its case, we shall be free to look at our behaving organism (man, animal, or plant), and to propound the only pertinent, scientific question--What is this organism doing? All agree that empirical study will elicit the answer to this question, and in the end the complete answer.
What, then, is it doing? Well, the plant is being hit by the sun's rays and is turning its leaves until they all lie exactly at right-angles to the direction of these rays : the stentor, having swum into a region of CO2, is backing off, turning on its axis, and striking out in a new direction : the hen has got a retinal image of a hawk and she is clucking to her brood-shoot the hawk or remove the brood and she stops clucking, for she is reacting to neither one nor the other, but to a situation in which both are involved : the man is walking past my window ; no, I am wrong, it is not past my window that he is walking; it is to the theater; or am I wrong again? Perhaps the man is a journalist, and not the theater, nor yet the play, but the "society write-up" it is to which the creature's movements are adjusted; further investigation is needed. This last instance is important, for the man "walking past my window" is generally doing so in no more pertinent a sense
( 369) than does the dead-leaf fall to the ground "past my window." Both are doing something else. Herein the folly of the bead theory be-comes clear. This theory says that in order to understand the man's actions, as he walks by, we must consider his successive "states," for each one is the cause of each succeeding one. And if we follow the theory faithfully, it leads us back to the successive "states" of each component process, and ever back, till we arrive at the flow of ions in neuro-muscular tissue ; in which disintegrating process the man with which we started is completely dissolved and lost.[4] But now the functional view, moving in precisely the apposite direction, admonishes us to keep the man whole (if it is behavior that we are studying) and to study his movements until we have discovered exactly what he is doing, that is, until we have found that object, situation, process (or perhaps merely that relation) of which his behavior is a constant function. The analysis of this behavior, as thus exactly described, will come in later; but it in turn will be carried on in the same spirit--i. e., of discovering always and solely functions. The movements of a plant, animal, or man are always a constant function of something, or a combination of such constant functions, and these-the movements, the functions, and the things of which the movements area function-are always open to empirical investigation.
But now, it will be said, the biologists and the behaviorists are doing just this thing-discovering constant functions. They are describing the motions of plant leaves as a function of the direction of the sun's rays, and are doing the same for all the aspects of animal behavior as well. They have done this for a long time. And there is nothing "novel" in behavior as so described. To which I answer, firstly, that the behaviorists are indeed doing this, are doing just the right thing; but they do not realize the significance of that which they are doing. And this is because, secondly, they are not aware of the remarkable novelty which behavior, considered just as they are considering it, does in fact involve.
An exact definition of behavior will reveal this. Let us go about this definition. Behavior is, firstly, a process of release. The energy with which plants and animals move ("behave") is not derived from the stimulus, but is physiologically stored energy previously accumulated by processes of assimilation. The stimulus simply touches off this energy.
( 370)
Secondly, behavior is not a function of the immediate stimulus. There are cases, it is true, in which behavior is a function, though even here not a very simple function, of the stimulus. These are cases of behavior in its lower stages of development, where it is just emerging from the direct reflex process. They demonstrate the continuity of evolution at this point-a most important fact. But as behavior evolves, any correlation between it and the stimuli which are immediately affecting the organism becomes increasingly remote, so that even in fairly simple eases it can no longer be demonstrated. This fact, that the immediate stimulus recedes in importance, is the interesting point about the integration of reflexes. It has been widely recognized in psychology; perhaps most conspicuously by Spencer, who generally refers to it under the term "higher correspondence." One will see in what relatively early stages of integration the immediate stimulus is thus lost sight of, if one considers how even the "retinal image" (to say nothing of the distant object which casts that image) is not, in any exact sense, the actual physiological stimulus; yet the organism "behaves" with regard only to the distant object. Since, then, behavior is not essentially a function of immediate stimulus, this latter can not enter into a definition of behavior.
But on the other hand, thirdly, behavior remains a function of some object, process, or aspect of the objective environment (including, in rare eases, the internal vegetative organs; which are still, how-ever, "objective"). And this is our crucial point. Not quite adequately realized by the behaviorists, it is terra totaliter incognita to the subjectivists. And the proposition negates their whole gospel, including especially the notion of "consciousness." I shall revert to this. Here we need only note that the behaving organism, whether plant, fellow man, or one's own self, is always doing something, and the fairly accurate description, of this activity will invariably reveal a law (or laws) whereby this activity is shown to be a constant function of some aspect of the objective world. One has here the same task as in any other strictly physical science. In both eases some accuracy is needed, and in both alike this accuracy can generally be advanced by mare exhaustive observation, Thus it is inaccurate to say that a river flows toward the sea, since it meanders about in all directions; while it is fairly accurate to describe it as always flowing toward the next lower level of the earth's surface, and this is a law describing flow as a constant function of the earth's crust and the position of the earth's center. The test is, of course, whether this or that could be removed without changing the river's course : the "sea" could be removed, the "next lower level" could not. So in behavior, the flock of birds is not, with any accuracy, flying over the green field; it is, more essentially, flying southwards; but even
( 371) this is only a rough approximation to a law of migration. In all events the flock of birds is doing something, and the sole question which we need ever ask is, "What is it doing?" I have elsewhere explained how the same question and it alone is applicable to one's own behavior (voluntary or other).[5]
Now I believe that the foregoing three propositions yield a definition of behavior. It would run : Behavior is any process of release which is a function of factors external to the mechanism released.
But why "any" process when it is well known that behavior is a phenomenon found in none but living organisms? Precisely because behavior as thus defined is in fact a striking novelty, which does not, so far as I am able to ascertain, occur anywhere in the evolutionary series prior to the appearance of organized response. This point is somewhat later, too, than. that at which life appears. In the ordinary inorganic case of released energy, the process, once touched off, proceeds solely according to factors internal to the mechanism re-leased. When a match is touched to gunpowder the explosion is a function of nothing but the amount, quality, arrangement, etc., of the powder. The beginning of the process is a function of the moment of firing; but that is all. When, on the other hand, an organ-ism with integrated nervous system is stimulated, the organism by virtue of internal energy released, proceeds to do something, of which the strict scientific description can only be that it is a constant function of some feature of the environment; and this latter is by no means necessarily the stimulus itself. The organism responds specifically to something outside,[6] just as the falling body moves specifically toward the earth's center. This fact offers no opening for the introduction here of "subjective" categories: the investigator continues to ask, merely, What is the organism doing? The answer will be in strictly objective terms. It can not be said that the ordinary release process is a function of the temperature, moisture, etc., of the surrounding air, for it is in fact a function of these only in so far as they penetrate .and become internal to the released mechanism. In behavior, on the other hand, there is a genuine "objective reference" to the environment which is not found, so far as I can learn, in the inorganic, or the organic world prior to integrated reflex response. This is the novelty which characterizes behavior. And here, if anywhere, evolution turned a corner.
In the second place, it may he noted that the definition neither
( 372) excludes nor yet makes essential the case of the immediate stimulus being the object of which the behavior is a constant function. This often happens, and is characteristic of the simpler instances where behavior is only beginning to be differentiated from plain reflex action. Evolution is of course not discontinuous, and the development from reflex action to highly organized behavior is one in which the correlation between stimulus and organism becomes less and less direct, while that between the organism and the object of response becomes more and more prominent. Plain reflex action is a function of the stimulus and of factors internal to the neuro-muscular are. Then presently one finds reflex movements that are due, as one must (with Sherrington) agree, to "so-to-say stored stimuli"; since the immediate stimulus does not account for the reflex movement. It is here that behavior begins, and precisely here that the "bead theory" would lead us astray. The response in question is a response to a past event, it is describable only in terms of (as a function of) this past event; while the bead theory would let us look only to the present condition of neuro-muscular tissue, the "so-to-say stored stimuli." These are of course an integral part of the causal process, but not the more enlightening part. Just as the measurements of the velocity of a body at successive moments are an integral part of its fall to earth, while if we considered nothing but these, we should never arrive at the true law of fall---a constant acceleration towards the earth's center. Or it is again as if, when one had photographed the spectrum of a newly discovered earth, one were misled by the bead theory into considering the result as "merely light and dark parallel lines on a gelatine negative." It is this, indeed, but it is also an interesting combination of metallic spectra. Or, again, the camera photographs a motor-car race, and the sensitive plate is affected a millionth of a second later than that in which the phase photographed occurred. By the time the print is obtained the race is long since over. The bead theory then says : This is only a black-and-white mottled slip of paper, it is no function of the racing motors. It is in just this way that in studying behavior we think that the only scientific view of it must be in terms of ionized nerve and twitching muscle. Is it any wonder, then, that having ignored the objective functional reference of behavior, we are led into the superstition of "ideas"in the "sensorium" which have an "objective reference" to the environment?
If now the behaviorist will bear in mind that he is scientifically justified in asking broadly, What is the organism doing?, he will discover that it is set to act as a constant function of some aspect of the environment, and he will find this to be the scientific description of the phenomenon he is studying. Then with this accurate descrip-
(
373) -tion as a basis, he can proceed to analyze it into its reflex
components and the relations by which they have been organized into behavior.
EDWIN B. HOLT.
HARVARD UNIVERSITY.
