Concerning Animal Perception
I wish to call attention to a phase of animal psychology which has received, it seems to me, but inadequate treatment. This inadequacy is evident not only in the general psychologies, but also in special experimental investigations of animal intelligence. The difficulty gathers about the doctrine of perception, and is due in part to the incomplete character of the theory of perception in human psychology, and in part to a failure to analyze sufficiently the conditions of possible perception in lower animal forms.
Can we draw a line between perception and higher cognitive processes, leaving below the line a cognition which is not rational though intelligent, such as characterizes the adaptations of a crab or a rat, and placing above the line all the consciousness of relation which makes human intelligence rational? Do our own predominately perceptive processes, such as those of rapidly climbing a steep, rocky cliff, or playing a game of tennis, where we are seemingly unconscious of anything except the physical environment and our reactions thereto, differ qualitatively from the more abstract processes in which we consciously deal in symbols and isolate the relations of things?
If these discursive processes are mere developments of contents which are implicitly present in perceptual consciousness, is there any definite line which can be drawn between the intelligence of man and that of the lower forms, unless we deny them the form of consciousness which we call perceptual in ourselves? Hobhouse,  for example, assumes that the cat, the dog and the monkey, which he observed, apprehend perceptual relations, which enabled them to learn by experience, without the ability to isolate the relations as elements in thought.
Stout  would grant to the chick that learns to reject a cin-
(384)-nabar caterpillar, an 'apprehension of meaning or significance, which would come to the same thing. On the other hand, Thorndyke  explains such learning by experience on the part of lower animals through the association of an ' impulse' with a stimulus, which seems to imply a qualitatively different state of consciousness from that which would ordinarily be called perceptual in human experience. He undertakes to illustrate this by phases of human consciousness in which even perception would be reduced to a minimum. This latter illustration indicates a possibility of discrimination which seems to me to have been but inadequately recognized. In learning to play billiards or tennis, we are moving in a perceptual world, but the process of improvement takes place largely below even the perceptual level. We make certain movements which are more successful than others, and these persist. We are largely conscious only of the selection which has already begun. We emphasize this and control to some extent the conditions under which the selection takes place, but the actual assumption of the better attitude, the actual selection of the stroke, lies below even this level of consciousness. Thorndyke calls this selection a process of stamping in by the pleasure coming with success. This explanation, however, calls for its own explanation and ascribes active control to states of pleasure and pain, which is by no means proved and opens up another field of dubious animal psychology. Thorndyke calls the process of improvement an association of an impulse and a stimulus, which lies quite outside of associations of ideas. The phrase is perhaps a vague one, that calls for further specification, but it answers to a large number of instances which are commonly conceived of as perceptions by the animal psychologists, although it is to be presumed that Thorndyke himself assumes that these animals move in a perceptual world. The instances to which I refer may be well illustrated by the action of the chick in rejecting the cinnabar caterpillar or the orange-peel. Is there a revival of the past experiences which leads the chick to reject these disagreeable objects; or may we assume that the impulse to reject has become associated with this particular stimulus, without any intervening redintegrated psychoses ?
This question is closely allied to that which arises with reference to the plasticity of the young form and the manner in which it acquires the specific habits which are not found performed in its nervous system. A chick learns to make use of the impulse to hide when a hawk sails overhead. A young fox learns to run away from the odor of man. The process of hiding and running away are indeed performed in these young animals. It is the association of the instinctive action with determinate stimuli which is acquired. What seems to take place is this: The animal tastes a disagreeable morsel when it instinctively strikes at a moving object before it. The action of the flavor of the morsel upon the organs of taste sets free an equally instinctive reaction of rejecting the morsel. At the same time, the chick eyes the caterpillar under the excitement of the disagreeable experience. Now the caterpillar hereafter to be avoided must be different from a mere moving object such as would have called forth the reaction of pecking. It is fair to assume that the condition for this discrimination made by the chick lies in the different reaction which it has called forth. The mere redintegration of the experience would not protect the chick. Either the chick would peck again, since presumably the same bad taste and same rejection would follow, simply reinforced by the revival of the past experience, and this would bring about no improvement in adaptation; or else the past experience would be revived with the appearance of the old stimulus. This stimulus was not a caterpillar with certain markings, but a moving object within reach. The revival of the experience with this generalized stimulus to which, as Lloyd Morgan's experiments show, the chick reacts, would lead to the rejection, not of cinnabar caterpillars alone, but of all moving objects within reach. The ability to distinguish between stimuli which had been identical in their value before, arises together with the new reaction, that of rejection. The meaning of the plasticity of the young form seems to be that there exist in the form instinctive reactions which have not as yet determined external stimuli. Through the experience of the animal the appropriate stimuli are determined. One condition, at any rate, is found in the new visual or olfactory expe-
(386)-rience which arises when, for any reason, this new reaction takes place. A dog's shrinking from the sight of the whip involves not simply the revival of the painful experience of the flogging; it involves his reacting to characteristicts(sic) in the sight of the whip which led to no reaction at first. It is not then so much the association of an old visual or olfactory experience with the impulse, as the arising of a new visual or olfactory experience which now becomes the stimulus for the particular impulse or reaction. If there be association of ideal contents, it is between this new visual or olfactory experience and the old experience which had not as yet been discriminated; of this association, Mr. Thorndyke remarks, we have little or no evidence. What we must assume, in what is implied above, is that the animal gets the new visual or olfactory experience because it is carrying out a new reaction; that the ground for discrimination in sensation lies in the difference of reaction to that which is sensed, an assumption that is reinforced by the recognition that the process of sensing is controlled and directed by the reaction to the stimulus.
Now what is implied in perception is the association of the new sensory experience with the old. If the chick perceives a caterpillar as a 'thing,' he may associate the former experience of pecking at a thing with the new experience of rejecting the peculiarly marked thing. But evidence for such an association in the case of the chick certainly is lacking. What has appeared in its conduct is a new stimulus of a visual character for a performed reaction, which up to this and other like experiences had no determined visual stimulus.
The question then arises, what are the conditions for the appearance of this permanent core to which varying sensory elements may be associated? It is impossible to appeal directly to the introspective analysis of human perception. We cannot get inside the consciousness of the lower forms. It is, however, possible to find in our own experience of physical objects what constitutes this core which endows it with its Thinghood, and investigate the conduct and sensory equipment of these forms, with a view to determining whether their experience can also contain this identical core to which varying phases of the same
(387) object can be referred. Stout  finds this core in what he terms ' manipulation,' understanding by this any contact experiences which arise as the result of visual stimuli, such as the hearing, scratching, pulling, shoving, as well as our actual handling of what we ace. This he illustrates by the visual experience of a hole to which an animal is fleeing and which answers to an experience of contact, that enables the animal to determine whether the opening is passable.
If this distinction be carried out somewhat further, we find that the sensory experiences of animal life may be divided into two categories: those that come through what may be called the distance sense organs, the visual, olfactory and auditory senses, and those that come through the contact sensations. The distinction suggested by Stout's use of the term 'manipulation' is that intelligent conduct, when it reaches the stage of perception, implies a reference of what comes through the distance sensations to contact sensation. There is perhaps nothing inherent in contact experiences which accounts for their being the substantial element in perception -that to which, so far as physical, i. e., perceptual, experience goes, all other experience is referred. Visual discriminations are much finer and more accurate than those of manipulation. The auditory and olfactory experience are richer in emotional valuations. But it remains true that our perception of physical objects always refers color, sound, odor, to a possibly handled substrate, a fact which was of course long ago recognized in the distinction between the so-called primary and secondary senses.
The ground of this is readily found in the nature of animal conduct, which, in so far as it is overt can be resolved into movements, stimulated by the distance senses, ending up in the attainment or avoidance of certain contacts. Overt food, protective, reproductive, fighting processes, all are made up of such movements toward or away from possible contacts, and the success of the conduct depends upon the accuracy with which the distance stimulation leads up to appropriate contacts. Consciously intelligent conduct within the perceptual field lies in the estimate of the sort of contact to which distance sensory stimulates the animal form, that is the conscious reference of experience result-
(388)-ing from the stimulation of the eye, the ear, the olfactory tracts, even the skin, by the movement of the air, etc., to the contacts which this stimulation tends to bring about.
The vast importance of the human hand for perception becomes evident when we recognize how it answers to the eye, especially among the distance senses. The development of space perception follow in normal individuals upon the interaction of the eye and the hand, and this interaction works a continual meeting of the discriminations of the eye by those of the skin, mediated through the manipulating hand. It is this contact experience which gives the identical core to which the contents coming from the distance senses are referred in the so-called process of complication. It is this core which answers to varying experiences while it remains the same. It is this core which is a condition sine qua non of our perception of physical objects. Of course this content of contact experience is supplied by the process of association or complication out of past experience in most of our perceptions. The objects about us look hard or soft, large or small. But the reference is always there.
There are two respects in which the contact experiences of lower animal forms are inferior to those of man for the purposes of perception. The organs of manipulation are not as well adapted in form and function for manipulation itself, and, in the second place, the contact experiences of lower animals are, to a large extent, determined, not by the process of manipulation, but are so immediately a part of eating, fighting, repose, etc., that it is hard to believe that a consciousness of a 'thing' can be segregated from these instinctive activities.
To develop this second point a little further, we need only to recall what has been brought out by Dewey  and Stout  that perception involves a continued control of such an organ as that of vision by such an organ as that of the hand, and vice versa. We look because we handle, and we are able to handle because we look. Attention consists in this mutual relationship of control between the processes of stimulation and response, each directing the other. But while this control is essential to perception, perception itself is neither eating, fighting, nor any
(389) other of the organic activities which commence overtly with stimulation and end with the response. On the contrary, perception lies within these activities, and represents a part of the mechanism by which these activities are carried out in highly organized forms. Perception is a process of mediation within the act; and that form of mediation by which the possible contact value of the distance stimulation appears with that stimulation, in other words, a mediation by which we are conscious of physical things. The actual eating, fighting or resting, etc., are not mediations within the act, but the culminations of the acts themselves. We could not perceive bread as a physical thing if that cognitive state grew out of the presentation of the mastication and taste which constitute eating. We perceive what we masticate, what we taste, etc., except in so far as we may perceive, through their movements, our various organs, as things.
The great importance of the human hand for perception lies in the fact that it is essentially mediatory within the organic acts out of which the physiological process of life is made up. The presentation of a physical thing which must be made up out of the contacts necessary to the actual processes of eating or those of locomotion cannot offer as fruitful a field for the growth of perception as those which are based upon the mediations of the hand within the act. And the contents of contact experience which a mouth or the paws can present must be very inadequate, for just that function of correspondence between the elements of the retinal and the tactual experience out of which the physical world of normal perception arises.
To assume that a chick can find in the contact of its bill together with those of its feet the materials that answer to the perception of a physical thing is almost inconceivable. Even the cat and the dog must find in their paws or mouths, fashioned seemingly for the purposes, not of 'feeling things,' but of locomotion or tearing and masticating, but a minimum of that material which goes into the structure of our perceptions. In the case of the monkey the question arises whether the function of locomotion is so dominant in use of the so-called hands that that of 'feeling' can be isolated out of the monkey's contact experiences to build up perception.
Finally, to recur to the difficulties inherent in the doctrine of perception referred to at the opening of this paper, the assumption of a perception of things, that is, of what is mediatory in experience, carries with it the essence certainly of reasoning, i.e., the conscious use of something - a certain type of experience -for something else, another type of experience. Every perceived thing is in so far as perceived a recognized means to possible ends, and there can be no hard and fast line drawn between such perceptual consciousness and the more abstracted processes of so-called reasoning. Any form that perceives is in so far carrying on a process of conscious mediation within its act and conscious mediation is ratiocination.