The Integrative Action of the Nervous System

Moreover, the supposition that the sensual reaction caused by a steady image acting at one of the pair of " corresponding" areas, is interfering with or combining with the individual phases of reaction to the intermittent image at the fellow area, is exactly the supposition that the observations dealing with symmetrical flicker show to be untenable. With intermittent lights throughout a wide range of ordinary intensities Talbot's ^* law is unimpeachable for the single eye ; and also for the two eyes if employed together under, as is usual, arrangements practically equivalent to the " simultaneous " right- left method for " symmetrical flicker." It is interesting to dis- cover how far the double retina will still observe Talbot's law when subjected to treatment such that, if the retina did then ob- serve the law, would indicate its integration to a functionally single retina. In other words, under a rapidly repeated stimu- lus, when one incidence of that stimulus has acted on a retinal point the question is: how far is it the same thing for visual brightness, whether the next incidence be upon the same retinal point or upon the twin point in the other retina? How far can

the double retina, when functioning for singleness of perception in binocular vision, be considered as functionally combined to a single retina, and how far does it then react as does a single retina, if examined for Talbot's law? The " alternate left-right arrangement" (Experiment i, LR) supplies the required method of stimulation. With speeds of revolution of the lantern too high to allow flickering, the binocular image LR (Fig. 79) is seen to appear of brightness equal to X/?, and with the uniocular images \ or p taken singly. Therefore in the above sense, Talbot's law not only does not hold for the double retina considered as functionally single, but no trace of observance of the law is detectable. The two corres- ponding points are therefore in this respect not integrated to a single retinal surface.

It was often noted that with all four lantern images of equal luminosity, using intermission frequencies too rapid to allow flicker, the brightness of the binocular combination of any two did not distinctly exceed that of the uniocular. In certain in- stances the binocular combination did appear just distinctly the brighter. This was for instance the case when of the four lan- tern images the two on the same horizontal level were combined by simple convergence. This excess of brightness is the well- known phenomenon examined by Jurin,^ Harris,^ Fechner,^^ Aubert,^^ Valerius,^^ and others. But there occurred frequent instances in which no excess was observed in the brightness of binocular combinations over that of their carefully balanced uniocular components. In these observations the brightness of the physical images is however always much above the threshold of the light adapted eye; and I have not made systematic obser- vations with the eye dark-adapted. To obtain good conditions for comparison of the brightness of the binocular and uniocular images the following arrangement can be employed.

Experiment 8. Two images LR and \p\ are placed in the visual field for mutual comparison. LR is composed of left-eye and right-eye equal and corresponding disc-shaped images as in previous experiments. X/)^ is composed of a left-eye image similar to L and R except that it lies just above or below them in the visual field. With X's right half is combined the image of the right half of a lantern image similar again [to the others, except that its left half is screened absolutely off into the )lank undetailed darkness of the general field. When this is done the two opposite visual images LR and Ap^ regarded under perfectly steady ocular fixation are stable, and no difference of brightness is discernible between them. Moreover no join is seen between the halves of Ap^ and no difference of brightness between the halves. After prolonged in- spection of them rivalry becomes troublesome ; but a judgment can be clearly arrived at before that happens.

In this experiment it might possibly be that equality of brightness between the halves of \pi is due to image pi not really being in consciousness at all during the comparison. The image might possibly lapse under competition with the partly dissimilar correspondingly placed left-eye image \. Experi- ments carried out by W. McDougall^^ give validity to such possible objection. The perceptibility of the horizontal bar in the right half of the image X/oi is guarantee however that at least part of the uniocular image pi is present. But to ascer- tain more surely whether image pi is really during the visual equation co-operating in consciousness with X the following further arrangement can be employed.

Experiment^ (Fig. 85). With the revolving lantern so arranged that images L, R, A, and pi are all of equal brightness when steady and unflickering, pi is given at a lesser frequency of intermission, so as to flicker while the others do not. A speed of revolution of lantern is then used at which just a trace of flicker is perceptible in pi when binocularly combined with A. The equation LR = Ap^ is then found to hold while flicker is still just traceable in the right half of Ap^. There is then no join seen between the halves of kp^ nor any difference between the brightness of the halves. So long as ocular fixation is steady no rivalry disturbs the observation.

In this case there can, I think, be no question but that the one half of \pi is truly binocular, for the trace of flicker is perceptible during the actual performance of the comparison. Yet no difference of brightness is perceived betv^een LR and \pi, and the two lateral halves of Xpi compared together seem of like brightness. Even when the binocular image does show the well-known slight excess of brightness over its uniocular components it, under some conditions (v. supra, " alternate " arrangement), flickers no more or even less than they.

It is doubtful therefore to me whether the slight excess in brightness of the binocular image over its two equal uniocular components is really explicable as summation of the intensities of the reactions at the corresponding spots of the two retinae. Valerius ^2 measured the increase to be one fifteenth of the brightness of the uniocular image. Aubert's^^ diagram gives it as less than one thirtieth. Aubert says it is not perceptible with brightness greater than that of white paper in diffuse day- light indoors.®^*

In certain modes of experiment a uniocular image used as a standard for comparison might itself be suffering some reduc- tion in brightness owing to slight combination with the dark field presented concurrently at the corresponding retinal area. But "rivalry" should reveal such influence. A better definition and greater vividness of detail assured by better accommodation and convergence under binocular regard, might possibly give an appearance of greater brilliance and intensity. But these are only suggestions.

I conclude that, with ordinary intensities of illumination, although a binocular image does sometimes appear of slightly greater visual brightness than either of two similar uniocular images composing it, more often it has a visual brightness not perceptibly different from that of either of its two co-equal uniocuiar com- ponents. The case then falls within a general rule regarding binoc- ular brightness attested by all observations I have on that subject. A binocular brightftess compared with its uniocuiar components is of value not greater than the greater of those ^ nor less than the lesser of them; when free from oscillations of rivalry its value is somewhat y but not far ^ above the arithmetic mean of the values of the two uniocuiar components as expressed by the measures of the physical stimuli yielding them.

The various combinations cited in Experiments 2, 3, 4, 5, 6, 7, and 8 have all, when steady and unflickering, given bright- nesses illustrating the above rule. Other illustrations are But I have not worked with combinations where the physical luminosity of one uniocuiar component has been less than y"^th the physical luminosity of the other. It was near this limit that Aubert, and just beyond it that Fechner, noted decline of the darkening effect of the darker component. In my own few observations beyond that point the oscillations of rivalry have made judgment difficult. The more manageable examples are but demonstrations of " Fechner's paradox," and fall under the above general rule. Hering^ has suggested that rivalry is really occurring even with similar and right and left uniocuiar images; he says these react according to a law of "comple- mental shares," and offers a theory, such as the name he gives implies, in explanation of the phenomenon. My Experiment 9 seems to offer difficulty to such a view.

Binocular combination of a less bright image with a more bright gives a visual image of less brightness than the latter (as stated in the rule above). But the application of the less bright physical image to the same uniocuiar area as the more bright gives a visual image of brightness greater than either. As above described, a steady image presented on an area of one retina " damps " the flicker of a flickering image concur- rently presented at the corresponding area of the other. A steady image actually physically superposed on the same retinal area as a flickering one also reduces the latter's flicker; this latter is of course in accordance with Weber's law. The modes of interference seem incomparably different in the two cases ; and experiment shows that the two interferences are often quite of different value.

Experiment lo A, Binocular fusion of R and L gives flicker ex- tinction at 65.5 phases per second. Physical fusion of R and L gives flicker extinction at 59.4 phases per second. Observer G. C. Binocular fusion of R and L gives flicker extinction at 106.6 phases per second. Physical fusion of R and L gives flicker extinction at 100.3 phases per second. R separately gives flicker extinction at 113.3 phases per second. Finally, to touch on " predominance of contours." Its facts, established by so many workers, are among the most signifi- cant concerning the difference between binocular and uniocular fusion of visual reactions. I will merely give one illustration which seems specially instructive for the point before us.

Experiment 11, A steady unflickering disc-shaped image L is pres- ent to the left eye : across the disc is a narrow dark line. An image R of similar size and shape but without the dark line is presented to the corresponding area of the right eye. If the luminosity of L is progres- sively diminished, a value of luminosity is reached at which its cross- line, though visible when L alone is observed {e. g. right eye closed) is lost or uncertain in the binocular image RL. This reduction of the luminosity of L much exceeds the reduction at which its cross-line is lost when image R is concurrently thrown on the same area of the same retina, /. e. left retina. Thus, in one experiment the diminution of luminosity of L required for loss of the cross-line under the physical superposition of R and L on the same retina was 84 per cent, while the diminution of luminosity of L required for loss (or great uncer- tainty) of the line in the binocular image was 96 per cent

Not only the ease, but the mode of disappearance of the cross-line, is significantly different in the two cases. In the " physical superposition " the dark line became gradually thinner and fainter, and finally imperceptible, as the image L is lessened in luminosity. In the case of binocular fusion the dark line oscillates out of and back into sensation more and more, the disappearances predominating more and more as the darkening of L proceeds. At a reduction of 84 per cent of the luminosity of L the cross-line was steady, dark, and sharp in the binocular image.

Our aim has been information as to the nature of the con- junction between the uniocular components in certain simple binocular sensations. The question concerns the nature of the tie between "corresponding retinal points," meaning by "retinal point" the retino-cerebral apparatus engaged in elaborating a sensation in response to excitation of a unit area of retinal surface. That a sensation initiated from corresponding retinal points is commonly referred without ambiguity to a single locus in visual space has often been regarded (Newton,^ Wollaston, ^* Rohault,^ Joh. Miiller^^) as evidence of community of the nerve apparatus belonging to the paired retinal points. Their visual image appears single. Wollaston supposed the twin points attached to one and the same nerve-fibre, which bifurcated at the chiasma. Rohault and Miiller supposed the points to be served by twin fibres " from one and the same ganglion-cell in the cerebral substance." Later (cf. Aubert^), the visual single- ness, spatial fusion of right and left impressions to a single per- ception, was taken to mean confluence of the nerve-processes, started in right and left retinae respectively, to " a single common centre or point of the sensorium." The discovery later still that the fibre-tracts from corresponding halves of the retinae both go to the occipital region of one and the same hemisphere has also been inferred to mean a spatially conjoint visual sensorium common to both retinae {e. g. Ramon-y-Cajal, Schafer). But in such questions the inferences obtainable from

mere anatomical features are equivocal and often remote in bearing. Were there to exist such a common mechanism situ- ate as a unit at conjunction of the two convergent systems and were phases of excitement timed so to arrive from one retina as exactly to fill pauses between excitations transmitted from the other, then there should be evidence of this in the time-relations of the phenomena induced. The state of excitement should tend to be maintained across periods that would otherwise checker it as pauses.

The retino-cerebral apparatus may be regarded as a structure of linked branching nerve-elements forming a system which ex- pands as traced centrally from the retinal surface. It may be figured as a tree, with its stem at the retina and an arborization spreading into the brain, its ramifications there penetrating a vast cerebral field, interlacing with others in a cerebral forest composed of nervous arborizations. The simile fails, because in the nervous forest the arborizations make functional union one with another. Is the fusion of the perceptions adjunct to paired "corresponding points" the outcome of a close concrescence of their neuronic or neuro-fibrillar arborizations, making of them practically a single upgrowth common to twin (right and left) stems rooted in the corresponding retinal units? If so, how low down, how close to their origin, are the twin systems grafted together, giving structural community to all the superstructure?

In the chain of nerve-elements attached to a sense-organ we infer in general that to the activities of the most peripheral links per se psychical events are not adjunct. Psychical processes, beginning with least complex and ascending toward develop- ment through many grades, attach to the chain in such a way that for the simplest only the more peripheral portions of the chain need be connected with the sense-organ, while for the more complex the central portions in addition become more extensively involved. But in the higher reactions of definite physical aspect, e. g. sense-perceptions, the lower apsychic and less definitely psychic activities are also implicate. Where from two sense-organs, e. g. two units of retinal surface, the two nerve-chain arborizations are mutually connected, so that

the lower activities of the one affect by low-level side connections the elements forming the other, there analysis must fail to distin- guish in the full reaction what higher components may be sepa- rately referable to one only of the two individual chains. The processes apsychic, or so indefinitely psychic as to baffle intro- spection, at root of those amenable to introspection, must by their coalescence defeat attempt to trace the final psychical product to either of its two possible sources, so long as both sources are open for its originatfon.

Were the nervous reactions initiated at twin points of the ret- inae early in its path along the retino-cerebral nerve-chain, to enter mechanisms common to both, there must, under " alter- nate " or "synchronous" right-left arrangement of stimuli (Fig. 78), be interference, algebraic summation, etc., a coalescence of events which, though apsychical in itself, would involve subse- quent confusion together of the sense-reactions of the two eyes. A state of things wholly different from this is revealed in the results of our experiments. And it would amount to the same thing whether two quickly successive flashes of a light fell both on one and the same member of a pair of " corresponding points," or whether the first fell upon one member, the second upon the other. But the experiments show that the effect in the two cases is widely different. Talbot's law is not applicable to the double retina, that is, to the two retinae functioning together in binocular vision. The experimental results go to disapprove the existence of any such fusion or interference between the apsychical or even the subperceptual events arising from corre- sponding retinal points. At most they indicate hardly discernible traces of such interference (Experiment i). They indicate, on the contrary, that such simple forms of binocular perception as have been dealt with here are themselves fusions of elaborated uniocular sensations. Since left and right end-results emerge pure, " hybridization " has not mixed the early stages in their evolution.

But the difference between the modes of stimulation left and right is a difference that, although it should be potent if the left and nght physiological machinery were conjoined to unity, should constitute no difference when left stimulation is compared with right stimulation by the perceptual product which each yields. The left eye and right eye flickering visual images^ each viewed singly, do NOT (apart from the faint cross-line for recognition) differ to introspection. If the sensations derived from the left eye and right eye respectively appear under introspection indis- tinguishably alike, what ground is there for mutual interference between them? It is much as though, of the left and right lan- tern images each were seen by one of two observers, with similar vision, and as though the minds of the two observers were com- bined to a single mind. It may be recalled that binocular unifi- cation of images, as we possess it, seems a comparatively late achievement of phylogenetic evolution.

When the visual product of the two retinae is thus regarded it is not surprising that Talbot's law fails for the binocular cyclo- pean retina. It fails because the binocular sensation is a fusion of uniocular sensation and from no two similar sensations can a resultant sensation be compounded different from its com- ponents. Were Talbot's law to hold in the above sense for the binocular retina there would, under the " alternate left-right arrangement" (symmetrical flicker), at rates of intermission too high for flicker, result from an image L of brightness x, and an image R of similar brightness x, a combined image LR of brightness x + x, the value of the summed brightness being in accord with the Weber-Fechner rule of summation of sensual intensities. But, as shown, not only does this summation not occur, but nothing like it occurs. The binocular result most often does not perceptibly differ from either of its two co-equal components.

But the experiments with uniocular components dissimilarly flickering, and with flickering components concurrent with steady components, do evidence (unlike the other experiments) interference between the two eyes. This result might be inter- preted as the outcome of community of the physiological mechanisms attaching to the paired " corresponding retinal points." But the other experiments negative the existence of this community. And the explanation just offered for the

absence of interference in the other experiments will account for the presence of interference in these. From two components perceptibly differing between themselves in regard to some quality {.e.g. flicker) a single combined sensual quality is ob- tained, intermediate between that of the two components taken singly. If the perceptible difference, e. g. in flicker, between the components is wide, the fusion is liable to phasic oscillations of predominance of one or other component. Where the difference in flicker is wide, such " rivalry " between the right and left com- ponents is in fact not unfrequently seen. One component may at the height of its phase be alone perceptible at the focus of attention, the other component being inhibited out of focal atten- tion or even out of conscious vision altogether. The inference is that only after the sensations initiated from right and left " corresponding points " have been elaborated ^ and have reached a dignity and definiteness well ameftable to introspection, does inter- ference between the reactions of the two {left and right) eye-systems occur. The binocular sensation attained seems combined from right and left uniocular sensations elaborated independently.

And in harmony with this view stands the evidence adduced for the rule formulated regarding the relation of binocular to uniocular brightness. Further, the difference between the sen- sual result of superposition of two similar images upon one and the same area of a single retina, and upon twin areas of the two retinae, could hardly be so great as it is, did apsychical or sub- sensual reactions underlying " brightness " combine or interfere in the two retinal systems. The binocular combination must be a synthesis of a left-eye with a right-eye sensation. Similarly, the " prevalence of contours " in binocular vision, and the phe- nomena of *' retinal rivalry," are explicable if each member of a pair of corresponding points yields a sensual entity which, when not widely dissimilar from that yielded by its twin point, fuses with that to a binocular sensation. In " retinal rivalry " we have an involuntarily performed analysis of this sensual bicompound. The binocular perception in that case breaks down, leaving phasic periods of one or other of the simpler component sensa- tions bare to inspection.

W. McDougall, 2^^ in applying to " retinal rivalry " and " prevalence of contours" his principle of competition of inter- related nerve-elements for energy, also argues a " separateness of the visual cortical areas for the two eyes." He brings for- ward striking experiments in evidence of this. In one of these he ^^ shows that an after-image, left from excitation of one retina, is more strongly revived by subsequent weak diffuse excitation of that same retina than of its fellow. More recently, in experiments proving reinforcement of visual sensations by the activity of the ocular muscles, as evidenced by after-image observations, he ^^^a shows that activity of the intrinsic muscles of an eye sends up to the brain an influence, reinforcing the activity of the cerebro-retinal tract of that eye, while it exerts no such effect upon the corresponding tract of the other eye, or exerts it in a minor degree only. With this separateness of the mechanisms, wherein are produced the sensations generated in the two retinae, our results by a different line of experimentation accord well.

The compounding together of right and left images really nonidentical but not widely dissimilar, is (Panum, Hering) the basis of visual " relative depth-perception." The compounding of visual images partly dissimilar — flickering with unflickering — seems a simpler case in the same category of synthetic actions. In our flicker experiments the visual components do not differ as to space-attributes, and their combination has therefore no resultant differential space-attribute. But the synthesis gives in each case a compromise between the components in regard to the attribute wherein they do differ; in the flicker experiments, that is in regard to the sensual steadiness of the brightness. This amounts to the same as the rule formulated above for binocular combination of brightness of different intensities, but steady.

Our experiments show, therefore, that during binocular regard of an objective image each uniocular mechanism develops inde- pendently a sensual image of considerable completeness. The singleness of the binocular perception results from tmion of these elaborated uniocular sensations. The singlcfiess is therefore the product of a synthesis that works with already elaborated sen^ sations contemporaneously proceeding. The cerebral seats of right-eye and left-eye visual images are thus shown to be separate. Conductive paths no doubt interconnect them, but are shown to be unnecessary for visual unification of the two images. The unification of a sensation of composite source is evidently associated with a neurone arrange- ment different from that which obtains in the synthesis of a reflex movement by the convergence of the reflexes of allied arcs upon its final common paths.

Here we seem to have therefore contemporaneity of itself sufficing for sensual synthesis, without necessarily any spatial fusion of the neural processes or mechanisms involved, i, e. without spatial confluence to a unit apparatus. As mentioned above, W. McDougall's experiments on after-images lead to a like conclusion. The foundation of new correspondences be- tween retinal points in cases of squint (Tschermak) strengthen the same view. McDougall has recently well summarized the position. But it is one not generally admitted by physiologists or psychologists. Ziehen 2*2a pyrites : " Schon physio) ogisch ist die Verschmeltzung der beiden Netzhautbilder dadurch vor- bereitet dass die Erregungen welche auf den linken Halften beider Netzhaute auftreten, vermoge der eigenthiimlichen par- tiellen Sehnerven-kreuzung zusammen in die rechte Grosshirn hemisphare gelangen, und umgekehrt." And this was the view of Joh. Miiller and of Aubert, and is advanced on histological grounds by Ramon-y-Cajal.