MIND

A QUARTERLY REVIEW

OF

PSYCHOLOGY AND PHILOSOPHY.

I-THE QUESTION OF VISUAL PERCEPTION IN GERMANY. (I.)

IMPARTIAL readers of recent English discussions of the spacequestion will be ready to admit that there is still ample room for more than one theory of the subject. Some years ago it was commonly thought that, thanks to the arguments of the Berkeleyans aided by the experiments of Wheatstone and others the derivative nature of visual space was amply demonstrated. Yet the skilful rehabilitation of the opposite doctrine by Bailey proved, as J. S. Mill allowed, how great the difficulties are which still beset the problem. More recently the ingenious arguments of what may perhaps be called the Dublin school, including Messrs. Abbot, Monck, and Mahaffy, have shown that the theory of visual space is even now far from being finally determined.

In Germany the same unsettled condition of the problem meets us. Indeed the division of opinion is even more strongly marked in that country than in our own. English writers on the whole have followed the direction indicated by Berkeley, who may be said indeed to have given shape to the problem in our country. In Germany, on the other hand, the discussion of the question received its initial impulse from the opposite side, namely, from the peculiar intuitional doctrine of Kant. And this fact explains why the intuitive or original view has been so ably represented in German writings. On the other hand, how

ever, the influence of Berkeley and generally of the analytic English psychology has made itself felt in the German discussions, and at present it may be said that the derivative view of space is quite abreast, if indeed not in advance of, its rival.

The field in which the space-question has been most warmly discussed is that of visual perception. This domain is clearly not one of pure psychology (in its narrow sense as a subjective science), but to some extent comes under the control of physiology. It offers ample territory for exact objective observation, and for skilfully arranged experiment. Accordingly one finds that in Germany it is the physiologists who have done most to advance the question of the nature and origin of visual space.

The immense advantage that the co-operation of these workers has secured is the accumulation of a large mass of new material which the psychologists of the future will have to work up in their theoretic constructions. Of the nature and extent of this material it is my chief object to give some account in this paper. It consists of observations and experiments which, being carried out by men trained in the conditions of accurate scientific data, is worth unspeakably more than the rough personal observations which used to be put forward as a sufficient groundwork of a psychological theory of space.

It will naturally be expected that such workers, having to deal with so complicated a set of phenomena, and not being specially trained in psychological interpretation, would, as soon as they began to theorise on their facts, reach very different results. What most strikes one, perhaps, in going over the recent literature of the subject is the number of seemingly distinct hypotheses set up in explanation of the phenomena. Closer inspection, however, shows that the diversities are often little more than verbal. Further, a free exchange of criticisms has served both to diminish the points of difference, and to reduce the number of the competitors whose claims are worth serious consideration.

I propose in the present paper to give a very brief sketch of some of the principal results of recent researches in physiological optics which bear on the nature of the visual perception of space. So far as possible I shall confine myself to facts, only giving such immediate conclusions from these as seem to be indisputable. In a second paper I hope to indicate the various ways in which the representatives of the different theories seek ultimately to interpret the facts.*

First of all, then, let us inquire what has been done of late to

I am indebted for most of these facts to Helmholtz's classical work Physiologische Optik. Next to this Wundt's elaborate treatise Physioogische Psychologie has proved most useful.

elucidate the nature of the eye's perception of space-relations in two dimensions, namely, relative direction, magnitude, and figure. We will first of all consider these properties as perceived by the single eye. The appreciation of them in binocular vision is a subject so intricate as to call for a special discussion later

on.

If we ask what are the means at the disposal of the eye in its construction of space, we find that these consist of two and only two modes of sensibility. The first of these is what is known as the discriminative local sensibility of the several nervous elements which compose the sensitive layer of the retina. It must be admitted that in the mature eye a peculiar local interpretation belongs to all impressions falling on the same retinal elements.* What the ultimate nature and the origin of this sensibility may be, is a question which must for the present be postponed. The second mode of sensibility which, as is now generally admitted, is involved in these perceptions is that which is variously known under the name of the muscular sense, feeling of innervation, and so on. There are a number of feelings attending ocular movement and the action of the ocular muscles. Of these the chief are those which accompany the actual movements of the eye, and which vary according to the direction and range of these movements. The question of the precise nature of this motor and muscular sensibility will have to be dealt with under the head of theoretic interpretations.

By help of these two orders of feeling, the eye gives local order to its impressions in respect of the relative position of points, lines, &c., their distance from one another, their magnitude, &c. In other words, by these means it is capable of conceiving the position of points, &c., in two dimensions.

Had we no other knowledge than this we should assign no particular distance to objects, nor would the surface on which we With our mature projected them have any particular shape.

ideas of space, we cannot, it is clear, conceive what our spaceintuition would be under these circumstances, though we may gain a faint imagination of it, perhaps, by thinking of the spaceideas of microscopic creatures living on the surface of a sphere, and knowing only points, lines, &c., lying on this surface.

It is sometimes said that from the first we tend to project retinal impressions on to an imaginary concave surface. Thus it is said that from the first children conceive the sky as the inner side of a hollow sphere. Yet it must not be supposed that our perceptions of the relative position and distance of points (or lines) would involve any such conception. The notion of a

Strictly speaking, this varies with the position of the eye. I assume here that the eye remains in one and the same position.

hollow sphere belongs to our complex mature space-consciousness, and our interpretation of the sky as a cupola may be explained as the resultant of many experiences.

In investigating what has been done to clear up the conditions of this side of our space-perception, we naturally begin with the discriminative sensibility of the retina. Careful observations have been made in this region by E. H. Weber and others, corresponding to the celebrated researches conducted also by Weber in the domain of tactual sensation. It appears from these that, in the case of a practised eye, in the area of perfect vision (the yellow spot) two points are distinguished when the visual angle reaches 60-90 seconds, or when the retinal image has a magnitude of 0.004-0-006 millimetres.

The discriminative sensibility is less fine as we pass from the centre to the peripheral regions of the retina; and, what is more curious, this falling-off takes place more rapidly along certain retinal meridians than along others. Thus it is less rapid in the horizontal than in the vertical direction. (Aubert & Förster.)

It seems probable that the cones, which are much more numerous in the area of the yellow spot than elsewhere, are the ultimate sensitive elements of the retina. It is a question, then, what relation exists between the minimum of local discrimination and the magnitude of the cones. This point is not yet settled, owing to the conflicting results of the measurement of these elements by different observers.*

The decrease in discriminative sensibility towards the periphery is explained by the comparative sparsity of the cones. It seems probable that over and above this circumstance, inequalities in the exercise of the different retinal regions have an influence here. It is to be supposed that just as special practice is found very considerably to increase the power of discrimination in the yellow spot, so the customary exercise of the eye would tend to render the sensibility of the central still more delicate than that of the peripheral regions.

The results of the defective observations reached on this whole subject point to the conclusion adopted by Wundt, that the local discrimination of the retina is somehow limited by the size of the ultimate nervous elements, though additional attention and practice may effect a considerable increase of sensibility within these limits.

Helmholtz, leaning on measurements of Kölliker, argues that the minimum retinal interval of distinguished points must be greater than the diameter of a cone. Wundt, following the measurements of H. Müller and Max Schultze, thinks that impressions may be distinguished which fall within the area of a single cone, and that this is effected by the help of the distinct fibrils which issue from one and the same cone.

We may now pass to the second elementary factor in the visual construction of space, namely, ocular movement.

The eye is rolled about its centre by means of six muscles. So far as a mere inspection of this mechanism would tell us, we might suppose that the eye's axis could be moved from any given point in the field to any second point by different combinations of muscular contractions. In point of fact, however, it is found that these movements are invariably carried out in one particular way. Thus it was found by Donders that to a given position of the eye's axis relatively to the head, there belongs a certain and invariable amount of rotation about this axis. In other words, whenever the eye fixates a particular point in the field (no matter from what other point it has moved), the various regions of the retina preserve the same relative arrangement.

Once more, it has been found by Listing that when the eye sets out from a certain primary position,' in which the principal axis is directed to the point of the field immediately in front of the eye, there is no rolling about the axis at all. In all such cases the movements are the same as if the eyeball rotated about an axis lying in the vertical plane which we may imagine to divide its anterior and posterior hemispheres.

There are two ways of regarding these uniformities of ocular movement. According to Wundt they answer to the least expenditure of muscular energy, and are conditioned by certain innate arrangements of the muscular mechanism.

On the other hand Helmholtz argues that these laws are to some extent the result of the individual experience. He has succeeded by the use of prisms, which impose unwonted conditions on binocular vision, in producing abnormal combinations of axial movement and rotation about the axis.

These views may be reconciled by the supposition, put forward by Wundt, that there is an innate disposition to the habitual or normal combinations, though this is itself the result of the collective experience of the race.

Some of the more obvious results of these laws in relation to visual perception are the following. First of all it follows from Donders's law that, whenever the eye returns to a particular point of the field, a fixed object in this region will be pictured on the same retinal elements. Now it is certain that the experienced eye perceives form when at rest and by help of the varying local sensibility of the retinal elements. It must follow, then, that so far as the eye appreciates form through a number of simultaneous retinal impressions, it will have an advantage in rotating as this law defines, since it will be able to return an indefinite number of times to an object at rest, and to receive from it a