§§85-92. General references are: H. L. F. von Helmholtz, Hand- buck der physiologischen Optik, 1896, 613 ff.; B. Bourdon, La percep- tion visuelle de Vespace, 1902; O. Zoth, An%enbewegungen mid Gesichts- ■wahrnehmungen, in NagePs Handbuch, hi., 1905, 283 ff.; H. Ebbing- haus, Grundziige der Psychologies i., 1905, 432 ff.; ii., 1908. 37 ff.; W. Wundt, Grundziige der physiologischen Psychologies ii., 1910, chs. xiii., xiv.
References for Farther Reading 339 § 87. An historical sketch of stereoscopy and pseudoscopy will be found in the author's Experimental Psychology, I., ii., 1901, 257 ff.
§ 90. D. Starch, Perimetry of the Localisation of Sound, in Univer- sity of Iowa Studies in Psychology, iv., 1905; v., 1908 (Psychological Review Monograph Supplements, 28, 38). On the warning sense of the blind, see a series of articles in the Zeits.f. experiment. Paedagogik, iii, 1906 ff.; also L. Truschel, Das Problem des sogenannten sechsten Sinus der Blinden, in Arch.f. d. gesamte Psychol., xiv., 1909, 133.
§91. W. Wundt, Die geometrisch-optischen Tauschungen, 1898; T. Lipps, Raumaesthetik und geometrisch-optische Tauschungen, 1897; V. Benussi, Zur Psychologie des Cestalterfassens, in A. Meinong, Unter- suchungen zur Gegenstandstheorie und Psychologie, 1904, 303 ff. A select bibliography of works upon optical illusion is given by the author, Exper. Psychol., I., ii., 1901, 305 ff.; a shorter list by Ebbinghaus, Psychologie, ii., 1908, 51 f. For eye-movements, see C. H. Judd and others, in Yale Psychological Studies, N. S., i., 1905 (Psychological Review Monograph Supplement, 29); for perspective illusions, J. E. W. Wallin, Optical Illusions of Rei'ersible Perspective, 1905.
§ 92. A brief account of Wundt's theory will be found in W. Wundt, Outlines of Psychology, tr. 1907, 113 ff.; of Hering's, in E. Hering, Beit rage zur Physiologie, Heft 5, 1864, § 124.
TEMPORAL PERCEPTIONS § 93. The Sensory Attribute of Duration. — All sensa- tions have the attribute of duration; all, even the briefest, may be observed as courses in consciousness, as mental processes. And all sensations may be localised in time, as occurring before or after some given sensation. The attribute of duration is, in the author's view, primitive and ultimate, like that of spatial extent; it corresponds, in consciousness, to the rise, poise and fall of the excitatory nervous process. Localisation in time may also, perhaps, be regarded as analogous to localisation in space. The conscious present (p. 19) varies greatly in objective dura- tion; but there can be no doubt that it may last for a considerable period of time: it is ' now ' during the whole hour that we spend in the dentist's chair, or during the whole morning that we devote to some baffling problem. Within these longer mental presents, at any rate, there will be occasion for the running together of like and the setting apart of diverse qualitative experiences; and the conscious grouping and separation may not only give the cue to temporal localisation, but may itself be localisation of a crude sort. Duration, the moving extension of a time-field, thus appears as the foundation upon which all forms of temporal consciousness are built up.
Time is usually regarded as a linear extension, a one-dimen- sional manifold. To the author, it seems that psychological time is rather a surface, a bidimensional manifold, and that its two dimensions are simultaneity and succession. It is true that simul- § 93- The Sensory Attribute of Duration 341 taneous sensations go on ' at the same time '; but this same time, in which they go on, is physical time; psychologically, each one of them goes on in its own time. There are thus a number of linear times proceeding side by side; and lines generate a sur- face. The great difference between bidimensional time and bidimensional space is that the latter is given once for all, and is simply articulated in the course of our experience, while time is made as we live; the time-field is constantly extending.
The existence of a conscious time-field, an extended present, is vouched for by our perception of melody, of rhythm (p. 289), of a polysyllabic word. Under laboratory conditions, this mental present reduces to a period of a couple of seconds: durations themselves are most accurately estimated at about 0.6 sec.; the natural rhythmical unit occupies 1 sec.; accommodation of atten- tion (p. 298) requires 1.5 sec.; the accuracy with which two suc- cessive stimuli can be compared increases up to a limit of 2 sec. after the presentation of the second stimulus. Nevertheless, as the following figures will show, even the least of these fields per- mits of temporal localisation; and it may be added that certain experiments seem to indicate a mental present of 6 sec. duration.
What, then, is the lower limit of temporal localisation? If stimuli follow one another with great rapidity, we have a con- tinuous and uniform sensation; if they come at a lesser rapidity, we have continuity without uniformity, — in vision, shimmer or flicker; in tones, harshness; in noises, rustle or rattle; in touch, roughness. If the rate of succession is still further reduced, we get a true temporal discontinuity, with the distinction of before and after; this appears, in rough average, for sight at an objective interval of ^ or ■£$ sec. (dark and light adaptation), for touch at 4^ sec, and for hearing at, perhaps, T^¥ sec. (noise). It is clear, therefore, that a conscious present of no more than 0.6 sec. still gives full scope for localisation.
While, however, these results are important, as showing that the conscious present is always a time-field, and not a time-point, it is to the longer presents of everyday life that we must look for the cues to temporal localisation. The two or three hours of steady attention (p. 293), the whole hour during which a young child 342 Temporal Perceptions amuses itself with a new toy, the evening devoted to a cram before examination, — periods of this sort, with their diversity of qualitative content and their succession of focal experiences, rep- resent the time-fields within which events are dated.
Temporal Discrimination. — A great deal of work has been done upon the comparison of durations. Unfortunately, it was supposed, at the outset, that the simplest time-experience is the interval of ' empty ' time between two limiting stimuli. In fact, of course, there is no such thing as an empty mental time; the interval enclosed between two clicks or taps is the duration of something, e.g., of some organic sensation; and the simplest time-experience is the ' filled ' time, the duration of a tone or colour or pressure, of some overt and obvious mental content. It has been found, accordingly, that the comparison of intervals is a complicated matter; it is differently grounded for intervals up to about 0.6 sec, for intervals between this and 4 or 5 sec, and for intervals of still longer duration. Short intervals are com- pared, not as spaces of time between limiting stimuli, but by reference to the stimuli themselves; every stimulus has its own duration, its time-halo; and the comparison is based upon the total temporal impression made by the two pairs of stimuli. Long intervals, again, are compared indirectly, by the help of secondary criteria, and principally by reference to the number of mental processes that run their course within the two stimulusperiods. The middle intervals, from 0.5 to 5 sec, are compared as durations, as spaces of time. Since the differential limen is approximately constant, so that the comparison of the intervals falls under Weber's Law, it is natural to suppose that the internal sensations which here carry the time are qualitatively constant and intensively variable; and introspection points to sensations of strain, whether due to the expectant attitude of the whole body or, more particularly, to the adjustment of the sense-organ to which the limiting stimuli are addressed, as the vehicles of the temporal judgment.
Temporal Illusions. — We are subject to gross illusion regarding the rate at which time passes (p. 7). A period that is rich in ex- periences is short in the living, and long in memory; as it passes, § 93- The Sensory Attribute of Duration 343 we have no time to attend to its time-value; when it has passed, we judge it to have been long by the number of experiences it embraced. There are moments, again, when time appears to stand still; and there are also, if the author may trust his own observation, occasions when time appears to be running backward: when, e.g., one is retracing, in profound absorption, the steps that have led to a conclusion which one desires to explicate or to justify. These are everyday illusions, due to a great complication of conditions. But illusions are found, also, under the stricter control of the laboratory.
Thus, within certain limits, an interval that is filled by a discon- tinuous series of sensations appears longer than an empty interval of the same objective length (cf. p. 336, line 2). Or again: when we listen to a rapid series of taps or clicks, we find ourselves forced, as it were, to accent some more strongly than others; the sounds fall into a rhythm. Suppose that we have three clicks, i.e., two inter- vals. If we accent the first click, the first interval appears the longer; if the second, the second; if the third, the first again: the effect of accent is to lengthen the following and to shorten the preceding interval. If the series of clicks really increases in loud- ness, the intervals seem to grow shorter; if it decreases, they grow longer.
The Index of Change. — There are two great groups of temporal perceptions, the continuative and the discrete. Typical of the latter is the perception of rhythm, which we discuss in the follow- ing Section; typical of the former is the perception of change, which appears in three distinguishable modes. We may perceive change, first, by the aid of secondary criteria, and especially by kinaesthesis (inhibition of breathing, the swell of inspiration, eye- movement, etc.). We may perceive it, secondly, as a specific pattern of the connection of elementary processes, a temporal over- lapping of qualities or intensities within the conscious present. And we may perceive it, thirdly, by what has been termed the in- dex of change: a peculiar modification of quality or intensity such that the attribute under observation shows confusedly, not as a point upon the sensory scale, but rather as something that can be referred only to a region or a section of the. scale. The physio- 344 Temporal Perceptions logical basis of this modification is to be sought in the overlapping of the excitatory processes in sense-organ and nervous system.
§ 94. The Perception of Rhythm. - - When we walk or run, we have a fairly regular alternation of stronger and weaker sensory complexes. The legs are pendulums, swinging from their point of attachment to the trunk; but the motor functions of the body are ordinarily asymmetrical, — we are right- or left-handed, as we say; and the superior muscular development of the one (usually the right) side means that, in the course of the double pace, the one foot comes down more energetically and swings more quickly than the other. The accent thus introduced into movements of locomotion is reinforced by the sympathetic swing and jerk of the arms.
In the movements of walking, of dancing, of speech and song, we have a tactual basis for the perception of rhythm. It is probable that this basis is primary, though we are now inclined to think of the rhythmical perception rather as auditory than as kinaesthetic; speech and song imply hear- ing, and even the rhythm of marching and dancing may come to consciousness most emphatically in auditory terms. Sounds are, indeed, the better material for the perception of rhythm: for the limbs, being fixed to the trunk, can give only the most rudimentary, duple rhythms, while sounds, whose stimuli are free, can be divided into groups of any perceivable complexity. Nevertheless, the kinaesthetic component persists. We mark time, beat time, as we listen to music, by free movements of head or foot or hand; and we manage, by spacing or reversing the movements, to imitate in kinaesthesis the complicated rhythms of audition.
It is a moot point whether the perception of rhythm may appear in complete independence of kinaesthesis. In the case of speech and song we have, of course, not only the muscular adjustments of § 94- The Perception of Rhythm 345 the larynx, but also the rhythmical play of the respiratory muscles. The author was formerly disposed to attribute a separate rhythmi- cal perception to hearing, but recent observation has convinced him of the existence of kinaesthetic sensations due to the contrac- tion of the tensor tympani of the middle ear. It is true that change of pitch may determine the character of an auditory rhythm. But pitch may here be regarded as a substitute for, or as itself the equivalent of, intensity: the former, if it arouses asso- ciations which, referred to their ultimate ground, are of an inten- sive character.; the latter, if strain sensations are evoked both by intensity and by quality of the sound stimulus.
Auditory rhythm may, in fact, be obtained by subjective accen- tuation (pp. 289 f., 343), and by variation of the duration, tem- poral separation, intensity, and pitch of the stimuli employed. Subjective accent is a matter of the insistence (p. 55) of certain elements in the stimulus complex; it is favoured by the observer's general attitude, and is oftentimes supported by large organic fluctuations (respiration, swing of the whole trunk). The primi- tive form or material of rhythm — for the level from which the perception of rhythm may be dated is still in dispute — is a dis- crete series of equally intensive impressions, whose members are separated by equal pauses: one finds it, e.g., in the syllabic reading of young children. Out of this grows, first the duple rhythm, probably in the order spondee, trochee, iambus, though the prior- ity of the trochee is disputed; and then the triple rhythm, probably in the order dactyl, anapaest, amphibrach. The limits of rhythmi- cal complexity have been discussed above, pp. 289 f.
The perception of rhythm may be aroused by visual impressions, whether by simple series of discrete stimuli, presented under lab- oratory conditions, or by the sight of rafters on a corridor ceiling, or of the recurring ornaments on a facade. In the author's opinion, this rhythm is always kinaesthetic, based upon eye- movement, upon slight movements which tick off the successive impressions, or upon some other form of intermittent kinaesthesis. Even those writers who believe in a purely visual rhythm acknowl- edge that kinaesthetic associates are almost invariably present, and are exceedingly difficult to suppress.
346 Temporal F Perceptions § 95. Theories of Time-Perception. — In principle, we have in the case of time the same divergence of psycho- logical theory that we noted in that of space. There are psychologists who derive, or generate, time from the in-" timate connection of processes that are themselves consid- ered to be timeless. There are also psychologists who make the two great temporal characters, duration and order, ultimate and irreducible. We have ourselves followed the lines laid down for a theory of space-perception; we have assumed that duration is an attribute of all sensations, and have regarded temporal localisation as a corollary of duration.
As illustrative of the genetic theories, we may take Wundt's ac- count of the origin of temporal ideas. " A sensation thought of by itself," Wundt declares, " can no more have temporal than it could have spatial attributes." The fundamental datum, in time as in space, is order, arrangement; duration, like extent, is for Wundt a secondary formation. Sensations are ordered in time by the help of temporal signs, just as they are ordered in space by the help of local signs. The temporal signs are fusions of af- fective with sensory elements: the affective qualities of tension and relaxation (p. 250) blend with the intensively graded series of kinaesthetic (especially strain) sensations. The time-percep- tion is thus " a fusion of the two kinds of temporal signs [the inten- sive and the qualitative] with each other and with the objective sen- sations arranged in the temporal form." The fixation-point of time, the ' now ' of consciousness, is determined primarily by affective processes; since these change, the fixation-point is constantly changing; and this change of fixation-point is what we mean when we speak of the flow of time.
As illustrative of the nativistic theories, we may take Ebbinghaus' account of temporal perceptions. Sensations, according to Ebbing- haus, have two classes of attributes: the individual or specific, and the general or common. In the latter class he includes space References for Further Reading 347 and time (extent and duration), movement and change, likeness and difference, unity and multiplicity. Rhythm falls under the heading of unity and multiplicity. Temporal order, succession, is merely discreteness of duration, the alternation of 'duration ' and 'interval '; and the distinction of duration and interval is itself merely a matter of direction of attention; we call the temporal attribute ' duration ' when we are attending to some attribute of the durable process, and ' interval ' when we are indifferent to this process, but attentive to its limiting impressions.
Wundt's theory is open to the objection urged against his the- ory of space. The blending of affective process with sensation means, elsewhere in the mental life, not time but feeling; and we cannot understand how, in this particular case, the new product should arise. Ebbinghaus' category of general or common sense- attributes seems to the author to take too much for granted; we are bound to push our analysis as far as it will go. Ebbinghaus appears, as regards both temporal localisation and the percep- tion of rhythm, to halt before he has reached the psychological goal.
All theories of time recognise the importance of secondary criteria, whether of duration or of localisation. Length of time may be estimated by number and variety of experiences, by our boredom, by the strain of expectation, by reference to a familiar time-standard; the date of an experience may be settled by range and particularity of memory, by reference to some salient event, by verbal association, and so on. These things have, however, nothing to do with the specific perception of time.
References for Further Reading §§ 93_95- Wundt, Physiol. Psychologie, iii., 1903, 1 ff.; Ebbinghaus, Psychologie, i., 1905, 432 ff., 480 ff., 504 ff. A brief account of Wundt's theory will be found in his Outlines, tr. 1907, 170 ff.
A bibliography of the work done upon temporal discrimination is given in the author's Experimental Psychology, II., ii., 1905, 393 ff. For the perception of rhythm, see ibid., I., ii., 1901, 337 ff.; R. MacDougall, The Structure of Simple Rhythm Forms, in Harvard Psychological Studies, i., 1903, 309 ff. (Psychological Review Monograph 348 Temporal Perceptions Supplement, 17); C R. Squire, A Genetic Study of Rhythm, in Ameri- can Journal of Psychology, xii., 1901. 492 ff.; K. Koffka, Experimental- Untersuchungeti zur Lehre vow Rhytkmus, in Zeits. f. Psychol, Hi., 1909, 1 ff. For the mental present, see L. W. Stern. Psychische Prasenz- zeit, ibid., xiii., 1897, 325 ff.; for the perception of change, the same author's Psychologie der Verandcrungsaujfassung, 1898.
QUALITATIVE PERCEPTIONS § 96. Qualitative Perceptions. — We have had instances of qualitative perception in the musical or compound tone (§ 25), in the various taste-blends (§ 34), in the touch-blends (§ 50), and in certain organic complexes, such as nausea and hunger (p. 188). It is characteristic of all these ex- periences, first, that the component qualities blend, fuse, run together, so that the perception appears simple, or at least unitary; but, secondly, that the components can still be identified, so that the perception may be analysed, under rigorous scrutiny, into a number of elementary processes. It follows from the sensory blending that the qualitative perception may become focal in consciousness as a whole. A recent writer has declared that the ' texture ' of qualita- tive perception, due to the ' massing ' of its sensory ele- ments,--it is difficult to find words to indicate precisely what is meant, — may, in certain spheres, be as important in creating apparent qualitative differences as is the quality of the single sensation; * and James, in a well-known pas- sage, has argued that the taste of lemonade comes to us, at first, as a simple quality.2 This fact, then, is important: that we may attend to the perception as a whole, and that the blending may be so complete as. to give us the illusion of qualitative simplicity. But the other fact, that system- atic observation always reveals the complexity of the perception, is of no less importance. We may attend separately to the separate components; and, if we take 1 E. Murray, Organic Sensation, in American Journal of Psychology, xx., 1909, 446. 2 W. James, Principles of Psychology, ii., 1890, 2.
350 Qualitative Perceptions this attitude, the perception breaks up into a number of really simple, sensory qualities. Thus, an observer who has never tasted limes will, if he is versed in introspection, distinguish the cold, the peculiar aroma, and the sweet, sour and bitter of limade, while his non-psychological host, who has put the ingredients together, will regard the • taste ' as a simple and single experience. The writer just quoted remarks, in another connection, that " many of the differences regarded at first sight as ultimate resolve them- selves, on analysis, into differences in the consolidation and coordination of the component sensations," and adds that " the dissection here carried out is not purely hypothetical, but was effected in almost every case directly by introspection." 1 James warns us here against two erroneous inferences. The one is " that because we gradually learn to analyse so many qual- ities we ought to conclude that there are no really indecomposable feelings 2 in the mind." We have spoken of this matter above, pp. 50 f. The other is " that because the processes that produce our sensations are multiple, the sensations regarded as subjective facts must also be compound." We referred to this source of error, the stimulus-error, on p. 218. In some cases, the stimulus-error is ruled out by the observer's ignorance of the conditions under which the mental process laid before him for analysis is produced; in general, however, it can be overcome only by long training. A third inference, which is certainly as dangerous as the two men- tioned, is this: that the psychological elements, just because they are elementary, are chronologically the first things in mind, so that perceptions grow, are formed, by the interconnection of orig- inally separate sensations. The elements are, as we have seen (PP- 37 f-)» the results of analysis; the perceptions are the original 1 E. Murray, A Qualitative Analysis of Tickling: its Relation to Cutaneous and Organic Sensation, in American Journal of Psychology, xix., 1908, 315 ff.
2 'Feelings' here mean 'mental processes'; 'sensations,' in the next quoted sentence, mean any cognitive processes that give us bare acquaintance with a fact.
§ 97* Tonal Fusions things, and the sensations are found in them by observation; per- ceptions are given us, and we discover that they are analysable. Misunderstanding here is fatal to the student of psychology, for it means misapprehension of the central psychological problem.
§ 97. Tonal Fusions. — The classical instance of the qualitative perception is the tonal fusion. We saw in § 25 that the musical tone is a complex of fundamental and overtones, and that its wave-train may be analysed into simple waves whose vibration-ratios are 1: 2, 3, 4, etc. In other words, the musical tone is a tonal fusion. It is, how- ever, a fusion of a complicated kind; for the overtones vary in intensity, and are present (p. 102) in some numbers.
We get tonal fusion at its purest by sounding together two simple tones at the same intensity. Under these con- ditions, we find that there are degrees of fusion. The fol- lowing Table shows the observed facts.
Degree or Fusion Illustrative Tones Vibration-ratio of Stimuli 0) C, Cl 2 Fifth c,g 3 Fourth c,f 3 4 Major third c, e 5 Minor third 5 6 Major sixth c, a 3 5 Minor sixth 5 8 Subminor fifth.
c,dp 15 Major seventh.
c,bp 9 These results, obtained both with simple and with musical tones, have been confirmed both by the individual judg- 352 Qualitative Perceptions ments of trained observers and by the collective observations of untrained persons. They mean that the tones of the octave, when all secondary criteria have been ruled out, are heard as a blend, whose unitariness is nearly akin to the simplicity of a tonal sensation, while the tones, e.g., of the major seventh fall apart as they are heard. Recogni- tion of the component tones makes no difference to the degree of fusion; we may know that an octave is sounding, and may be able to identify the tones that compose it; but so long as we are listening to the octave, so long as the fusion itself is focal, we still hear the unitary tone-blend.
The phenomena of tonal fusion have been studied in great de- tail, — with assemblages of more than two tones, with varying in- tensities of the components, with distribution of the stimuli to the two ears, in intervals beyond the octave, in mistuned musical intervals. For our purposes, however, it is unnecessary to go fur- ther. The essential thing is to grasp the twofold character of the fusion: its relative unitariness, and its recognisable complexity. The unitariness is given, sensibly, as we hear it; analysis leaves it as unitary as it was before. Yet the unitariness is never that of a chemical transformation; analysis, for the normal ear and after practice, is always possible.