§ 98. Theories of Qualitative Perception. — When we are looking for the physiological conditions of extent and duration, our problem is to discover some property of the sense-organs that can bring them into correspondence with the spatial and temporal aspects of stimuli. Physical objects lie in physical space; physical events occur in physical time. Hence the explanation of extent and dura- tion is of the same general kind as the explanation of quality and intensity; the various aspects of the stimulus must be correlated with certain modes of sensory excita- tion. Our present problem is of a different kind. We § g8. Theories of Qualitative Perception 353 have to discover the physiological conditions both of the unitariness of the qualitative perception and of the possi- bility of its analysis. We begin with tonal fusion.
Stumpf, to whom we are chiefly indebted for our know- ledge of the facts of tonal fusion, thinks that analysis is conditioned by peripheral, fusion by central factors; the ear analyses, if we may so phrase it, and the brain blends. He ascribes the blending to a specific synergy of the nerve- centres, to a determinate mode of cooperation between the nervous structures affected by the tonal excitations. In the present state of brain physiology, this theory, as Stumpf admits, is little better than a form of words; it simply warns us that we shall fail to find in the ear a mechanism for the appearance of fusion. Ebbinghaus, on the other hand, believes that fusion can be explained in terms of the peripheral mechanism; the Helmholtz theory is adequate both to analysis and to blending.
We know that, according to the Helmholtz theory (§ 26), the ear is an analyser: let us see how the theory accounts for fusion. We may begin with the simplest case, that of the octave. Sup- pose that two tones, whose pitch-numbers are 300 and 600 re- spectively, are sounding together. They will set into vibration the basilar fibres whose normal vibration-rates are 300 and 600 in the 1 sec. But they will also, Ebbinghaus says, set in vibration the harmonic undertone fibres; nodes will be formed, and the fractional parts of the undertone fibres will take on the rhythm of the primaries. Thus, the 300-stimulus causes the fol- the 6oo-stimulus causes the following fibres to vibrate at the lowing fibres to vibrate at the rates shown: rates shown: 2A 354 Qualitative Perceptions and so on. It is clear that the 300-fibre is asked to vibrate both as a whole (X 1) and in halves ( X 2), the 150-fibre both in halves and in quarters. The fibre takes up the easier, that is, the slower vibration, and ignores the other, so that the higher tone, that of 600, loses some of its body; the lower tone steals from it. Con- sequently, the upper tone becomes a mere shadowy parasite of the lower tone, and we have the fusion of the octave. It is matter of observation that in duple fusions the lower tone carries or domi- nates the fusion.
Now consider the tones 300 and 480, a minor sixth. The fibres stimulated will be and so on. Here there is no identity of fibres in the two columns; but 150 stands near to 160, and 100 to 120. Since the basilar membrane vibrates, not in single fibres, but in narrow strips (p. in), the 160- and 120-fibres will be hampered in their vibration. The lower tone again steals from the upper, though less cleanly and to a less extent than in the instance of the octave.
It should be added that Ebbinghaus does not deny the possi- bility of a central factor in fusion; he is concerned only to show that the peripheral mechanism affords a plausible explanation. There is, however, one point of observation upon which he and Stumpf are sharply at variance. Stumpf declares that fusion re- mains the same, whether the tones are heard under the ordinary conditions of binaural hearing, or are heard separately by the two ears, or are represented in imagination; and it would be curious, he says, if in the first of these cases the fusion should be a pe- ripheral, and in the others a central matter. Ebbinghaus main- tains that, if very weak and fairly low tones, say, of 400 and 600 vs., are heard binaurally, the higher tone is practically lost in the lower; whereas, if the tones are carried separately to the two ears, they are heard " with perfect clearness and distinctness side by side."
Refere?ices for Further Reading 355 Whether Ebbinghaus has made his point is a question that will be answered differently by different psychologists; he has at all events given a theory that is definite in outline and that appeals to known physical principles. Yet it seems that qualitative perception in general must be re- ferred to central, and not to peripheral conditions. Where the fusion occurs between qualities of separate senses, as in the taste-blends, this conclusion cannot be escaped. But even in the case of the touch-blends there is evidence that the blending depends, not upon peripheral irradiation, but upon processes within the nervous system. What these processes are, and how it is possible that now the total perception and now its sensory constituents may be- come focal in consciousness, we do not know.
References for Further Reading «$§ 96-98. C. Stumpf, Tonpsychologie, ii., 1890, 127 ff., 184 ff.; H. Ebbinghaus, Psychologie, i., 1905, 318 f., 344 ff. Wundt gives a psy- chological theory of fusion in the Physiol. Psychol, ii., 1910, 116 ff., COMPOSITE PERCEPTIONS § 99. Simple and Composite Perceptions. — The percep- tions that we have so far discussed may be termed simple perceptions, since they rest upon a single sensory basis, upon the sense-attribute of extent or duration, or upon the concurrence of sensory qualities. There are also various types of composite perception. Thus, the perception of a movement in the field of vision or of touch is both temporal and spatial; the movement has duration, and it has at the same time extension (spatial magnitude) and direction. The perception of melody is both qualitative and temporal. The perception of a thing, an object, is qualitative and spatial; the perception of a scene, a situation, an event, is qualitative, spatial and temporal.
It is not necessary to take up all these composite percep- tions in detail; for the most part, their analysis follows at once from that of the simple perceptions. We must, how- ever, say something of the perceptions of movement and of melody.
§ 100. The Perception of Movement. — We need not dis- cuss how a moving object is localised, or how we perceive the magnitude, direction and duration of the movement; these questions have already been answered. The difficult thing about movement is its continuity; and the difficulty is resolved in the alternative ways, by nativistic and genetic theories. Some pyschologists regard the specific experience of movement as ultimate and irreducibb; they even speak of sensations of movement, not in the familiar § ioo. The Perception of Movement 357 sense of the sensations aroused by movement of the body or limbs, but in the literal sense; they believe that the moving stimulus arouses a sensation of moving, what we might call a travel-sensation. Other psychologists find the sensory basis of continuity in the positive after-image (p. 68), the persistence of sensation after the cessation of stimulus. By help of this after-image, they say, we are able, within the mental present, to see or feel a stimulus as extended over the whole space between the point which it has just left and the point to which it has just come; the tailing-off of the after-image, its gradual loss of intensity, forbids us to perceive this extension as a spatial extent pure and simple; and the recognition of the moving object as less extended than its path, and as identical at all points of its course, clinches the perception of movement. The author inclines to accept the genetic view, though there are certain observations which it has, so far, failed to explain.
A good deal of work has been done upon the quantitative as- pects of the perception of movement, — the minimal and maximal rates at which movement may be perceived, the differential limen of rate, and so on. We notice only two points. The first, which indicates the insistent character of the moving stimulus (p. 269), has to do with the extensive limen of movement. In direct vision a moving stimulus, to be perceived as moving, must traverse a dis- tance sensibly equal to that which permits of the local distinction of two stationary points: the limen of spatial duality and the ex- tensive limen of movement are practically identical. But in in- direct vision, and in the cutaneous space-field, the stimulus is perceived as moving when it has traversed only about a quarter of the distance required for the perception of duality. It is clear that, as we said above (p. 271), movement makes a very special appeal to the organism. The second point is that the 358 Composite Perceptions discrimination of rates of visual movement is, within certain limits, subject to Weber's Law; it seems, then, that kinaesthetic sensa- tions play the same part here that they play in the discrimination of the moderate time-intervals (§ 93).
On the side of extent, the perception of movement is simply a mode of the perception of spatial magnitude. It may be worth while to remark — though the fact is implied in previous discus- sions— that estimation in terms of eye-movement is very un- certain, unless there is somewhere in the field of vision a fixed point of reference. Movements of the eyes, to and fro, are con- tinually going on, and are rarely remarked.
Illusions of Movement. — We can, again, notice only a few typical illusions. If a stimulus moves over the skin at uniform rate, we take the movement to be quicker where localisation is more accurate, slower where it is less accurate (cf. p. 326). Illusions of visual movement are very frequent, and are due to a great variety of conditions. It is a general rule, e.g., that a fixated object is seen at rest. Hence, if we fixate a tree from the window of a moving train, the tree itself appears to standstill, while the objects on this side of it move backwards, and those beyond it move with the train forwards. Yet the moon, seen between moving clouds, seems to move, and the clouds seem to stand still! Some other principle is evidently at work: possibly the principle that small objects are more likely to move than large.
The Synthesis 0/ Movement. — The perception of movement may be synthetised by means of the stroboscope, in which discrete phases of some objective movement are thrown in rapid succes- sion upon the retina: the instrument is familiar as a toy, and is popularly known as the zootrope. The stroboscopic effect has usually been referred to the persistence of sensation in the pos- itive after-image. But the cylinder may be turned so slowly that the bridging of the gaps by after-images is out of the question, and the perception of movement still continues. It follows that conscious predisposition (cortical set) is of great importance for the perceptipn of movement, — a fact which seriously complicates the problem se'i us hv certain movement-illusions.
In the stroboscope, our virion of the pictures is periodically § ico. The Perception of Movement interrupted by the solid parts of the cylinder-wall, between the slits; if it were not for this interruption, the pictures and their background would run together into a meaning- less blur. For the same reason, the photographic rib- bon of the kinematograph, or moving-picture machine, is passed before the lens of the projection lantern not con- tinuously, but by jerks; every picture is allowed to remain stationary for a moment before Fig. 59. James' Artificial Waterfall. — W. James, Mind, O. S. xii., 1887, 517; Principles of Psychology, ii., Very striking after-images may be obtained by the slow rotation, on the colour-mixer, of a white disc upon which has been painted in black an Archimedean spiral. — E. Mach, Grundlinien d. Lehre von d. Bewe- gungsempfindungen, 1875, 59.
FIG. 58. B. Bourdon, La perception visu- elle de I'espace, 1902, 194.
it is displaced by the next succeed- ing picture.
The Effect of Conscious Predis- position.— If two like objects ap- pear in succession, at different points in space, under conditions which do not preclude the idea of movement, we almost inevitably perceive the movement of a single object. Figure 58 represents a white disc, upon which a series of similar figures A, B have been painted in black; before the disc stands a screen with a small win- dow for observation. Let the disc rotate, in the direction of the ar- row, at such a speed that an in- terval of some quarter-second elapses between the appearances of two neighbouring figures. As the figure is drawn, A has already passed the window, and B is com- ing into view. But B does not, 360 Composite Perceptions rise from below; at the moment indicated, the observer has the distinct perception of an abrupt descent of A. The succession A-B is perceived as the movement of a single A.
After-images of Movement. — If we look for a while at the water over the side of a moving vessel, or gaze at a waterfall, or keep our eyes upon the moving roll in a piano-player, and then turn to the planks of the deck, or the banks of the stream, or the name- plate on the cover of the instrument, we get the perception of a reversed movement, a sort of negative after-image of the original. This after-effect cannot be referred to involuntary movements of the eyes, since it appears only at that part of the field which was occupied by the stimulus. The most plausible explanation, on a genetic theory, is this. The moving object leaves in the eyes a mass of shifting and fading after-images, which, if seen for them- selves, would simply continue the perception of movement in the original direction. Their qualitative character is, however, merged and lost in the qualities of the field upon which they are projected. Since, nevertheless, they still suffice to give a cue to the percep- tion of movement, their passage over the field must arouse the illusion that this is itself in motion, and in the opposite direction. The explanation is not altogether satisfactory, and it must be con- fessed that, in this matter, a nativistic theory of the perception of movement has the advantage.
§ 1 01. The Perception of Melody. — As movement is both temporal and spatial, so melody is both temporal and qualitative. It presupposes rhythm; and it presupposes the formation of a musical scale, and the classification of the intervals of this scale as consonant and dissonant.
Rhythm we have already discussed. The conditions that led to the discovery and the selection of melodic intervals are still obscure; Stumpf refers melody to the fact of fusion, while Wundt distinguishes a number of factors, chief among which is what he terms the direct relationship of the compound tones.
§ ioi. The Perception of Melody 361 According to Stumpf, consonance depends on fusion, and de- gree of consonance upon degree of fusion. And fusion is operative in the case of successive, as it is in that of simultaneous, tones; the second tone fuses with the image (§61) of the first, or, if both tones have ceased, the image of the one fuses with the image of the other. Homophonic and polyphonic music, melody and har- mony, thus have their common root in tonal fusion.
Here, however, an objection arises. Was not primitive music homophonic? And could the hearing of successive tones bring the differences of fusion- degree so clearly to consciousness that a musical scale should result? Stumpf replies that primitive music was not wholly homophonic. Men and women, or men and boys, chanted together, and the difference of register would give the intervals, at any rate, of the octave and the fifth; untrained singers, who think that they are singing in unison, will sing, ac- cording to the compass of their voices, in unison, in octaves, or in fifths. Moreover, musical instruments are a very early invention; and the fusion-degree of the fourth, the thirds, etc., occurring by chance, might strike the fancy of the primitive musician, and so the intervals might be fixed and employed. Finally, the addition of a drone-bass, vocal or instrumental, would help to keep the singer within certain tonal limits. — It may be added that, while we are ac- customed to think of a scale as beginning in the bass and contin- uing upwards to the treble, primitive scales, at least in very many instances, run from treble to bass. The earliest melody was probably somewhat like our recitative; and the voice naturally falls at the end of a sentence. Now the descending fourth is the ordinary drop of the voice in speaking, as the ascending fifth is its ordinary rise in questioning. It seems, then, that the interval of the fourth may have been fixed, in primitive music, as early as the octave and the fifth, but fixed in the descending direction.
According to Wundt, consonance derives from four principal sources, two of which are metric and two phonic. In the first place, the most consonant tones are those which, given together, arouse the fewest difference-tones, — the tones of the octave and the fifth. These intervals, fixed by group-singing, set the limits within which the single voice or instrument is to move. Secondly, the ear is 362 Composite Perceptions able, within certain limits, to compare and to equate tonal distances (cf. p. 209); and, as the absolute sensible discrimination is here constant, the octave is thus bisected into fifth and fourth, the fifth itself into major and minor thirds. Thirdly, to pass to the phonic principles, we find that consonance depends upon the direct and indirect relationship of the compound tones; that is to say, upon the number and intensity of their common overtones, and upon the nearness of their relation to a common fundamental. Fourthly, consonances are characterised by a peculiar mode of fusion: the fusion is less intimate than, e.g., in the prime, or the continuative noise (p. 97); it is a 'distinct' fusion, a fusion of discriminable elements, as contrasted with the ' diffuse ' fusion of the noise; and it is a fusion whose dominating tone owes this dominance, not to any characteristic of its own, but to the concurrence of the three conditions of consonance which we have already mentioned.
These four factors in the perception of consonance combine, under different circumstances, in different ways. The succession of pure tones in a melody depends upon the metric principles and upon indirect relationship; the succession of compound tones de- pends almost entirely upon the phonic principles, and especially upon direct relationship.1 Both in harmony and in melody, con- sonance is, for Wundt, primary, and tonal fusion only secondary.
A choice between the rival theories is, at the best, very difficult. The questions involved are, for the most part, beyond the reach of experiment, while our knowledge of primitive music, as it exists in the world to-day, is imperfect, and the results so far obtained have not been systematised. Where experiments are possible, their outcome is often rendered uncertain by the musical training of the observers; aesthetics may cut across psychology. We shall therefore do well to suspend judgment.
It is natural that, as the scale becomes complex, the rules of melody become precise; but these rules, as well as the affective aspects of rhythm and interval, lie beyond our present horizon. The unit of our modern scale is, of course, the semitone; and it is 1 Stumpf does not deny the influence of direct relationship upon the primi- CS oerception of me'odv. but he confines it to the intervals of octave and fifth.
References for Farther Reading 363 worth noting that this, while it is by no means the least difference of pitch that the ear can distinguish (p. 98), is the least difference that can be accurately sung. The larynx is the earliest musical instrument, and the singing of grace-notes, with a just noticeable difference of laryngeal adjustment, may be responsible for the final form of the musical scale.
References for Further Reading § 100. B. Bourdon, La perception visuelle de Vespace, 1902, 176 ff.; H. Ebbinghaus, Psychologie, i., 1905, 489 ff.; W. Wundt, Physiol.
§101. C. H. H. Parry, The Evolution of the Art of Music, 1896; H. L. F. von Helraholtz, On the Sensations of Tone, 1895, 234 ff.; C. Stumpf, Konsonanz und Dissonanz, in Beit rage zur Akustik und Musik- wissenschaft, i., 1898, 1 ff.; VV. Wundt, Physiol. Psychol., ii., 1910, 440 ff.; iii., 1903, 138. A 'motor' theory of melody is outlined by W. V. D. Bingham, Studies in Melody, 1910 (Psychological Review Monograph Supplement, 50); here may also be found references to the work of the most recent investigators of the subject.
THE PSYCHOLOGY OF PERCEPTION • § 102. Pure and Mixed Perceptions. — Let us assume for the moment that any perception may be analysed, without remainder, into a number of sensations. It would still be true that the mere enumeration of these sensations is not an adequate account of the perception. For the sensations which we find, in the particular case, form a group; they have been selected, singled out, marked off, from the other contents of consciousness. Their fewness is due to the limited range of attention (§ 80), and their present appear- ance as a group at the conscious focus depends upon con- ditions which we have already discussed (p. 270). The grouping itself, the special mode of connection of the sen- sations, we have tried to explain in the preceding Sections.
The simplest kind of perception, then, — what we may call the pure perception, — implies the grouping of sensa- tions under the laws of attention. But it is clear that perceptions are, as a rule, not made up solely of sensa- tions; we see and hear and feel more than is presented to eye and ear and skin; the given sensations are supple- mented by images. Most of our perceptions are mixed per- ceptions, complexes of sensory and imaginal elements; and the life of perception is, far more than one is apt to sup- pose, a life of imagination.
We may have at any rate an approximation to the pure percep- tion (cf. pp. 50 f.) both in the laboratory and in everyday life. When, e.g., we are comparing two linear extensions by eye, or when we are listening to a tonal fusion, with secondary criteria so § 102. Pure and Mixed Perceptions 365 far as possible ruled out, consciousness is practically restricted to the sensation-group. So, if we are unpacking the parts of a new and complicated instrument, and come upon something the use of which is not immediately apparent, the few seconds of intent scrutiny are taken up with a pure perception. The author was once shown a photograph, which consisted of a circular field scrawled all over with random and zigzag marks, and was asked what it was. Here was a pure perception; there was no supple- menting by images. Then the suggestion came: Look at the back! On the back was a date, and the date was that of a great earthquake. The perception at once became mixed; the photo- graph was a seismogram.
Ordinarily, however, our perceptions come to us as mixed. To realise how inevitably they are mixed, devote a couple of days to the following observation: whenever any casual object strikes your attention, do not be content to pass it by with a hasty identi- fication, but go up to it and examine it. You will find that the bit of glass in the grass is really a bit of grey limestone upon which the sun is shining; that the twig which caught your eye because it seemed to move did move — and is a caterpillar; that the enamelled bowl in which a careless servant has left the water is dry and empty. You will, in a word, be amazed to find how little you ' perceive ' and how much you ' imagine.' We have touched on this point above (p. 199), and have suggested that the general reliability of perception may be due, in part, to the different set- ting in consciousness of sensation and image, in part to an in- trinsic difference in the conscious stuff of which the two processes are made. We may now add that, where the image is incorpo- rated in the perception, it obeys the same laws as sensation, a/»d ^-*. that this likeness of behaviour, while in certain cases it, favours *-^N ■ illusion, must also on the whole favour an adequate apprehension. m of stimuli. It is hardly too much to say -that the occurence of.. x illusions is a guarantee of general reliability: glass in the*grass does look like that, things that grow on shrubs are twigs, water left in the bowl would give this appearance. We return to the The images that supplement the pure perception are different in 366 The Psychology of Perception