lHlell ? I .THS A STUDY OF THE PERCEPTION 0F THE DURATION 0%: CONYiNUCUS: WARILED AND Pb’LSED SEGNALS Thesis for the Degree 0; M. A. MECBEGM SEEK? UNPJERSITY Patricia Sheafor 196-3 LIBRARY Michigan State University A STUDY OF THE PERCEPTION OF THE DURATION OF CONTINUOUS, WARBLED AND PULSED SIGNALS By Patricia Sheafor AN ABSTRACT Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS College of Communication Arts, Department of Speech ABSTRACT A STUDY OF THE PERCEPTION OF THE DURATION OF CONTINUOUS, WARBLED AND PULSED SIGNALS by Patricia Sheafor The purpose of this study is to ascertain the accuracy of human perception of the duration of various auditory signals of the same frequency which vary in duration and method of present- ation. The twenty-five subjects utilized in this study had nor- mal hearing for the frequency involved in the test stimuli and were all college students with a mean age of 25.6 years. Pairs of 1000 cps continuous, warbled and pulsed tones, 6.2-8.6 sec- onds in duration were randomly arranged and transcribed onto magnetic tape. They were presented at 80 db with the first stim- ulus always being the reference signal and the second, the vari— able. The subjects were required to make a three choice judge- ment of.longer, equal or shorter. The subjects were tested in groups and the total time required for the test procedure was ap- proximately one and a half hourso The findings of this study indicate that subjects had greater difficulty discriminating duration of the shorter stimuli and that the method of presentation did not affect the per cent of correct judgements of duration. Subjects were able to judge duration within .28, .#8 and .51 of the standard for each method Patricia Sheafor of presentation respectively. The conclusions drawn from this study suggest that.one should counterbalance stimuli for methods of constant stimuli used.in con- nection with judgements of duration. Longer time durations are per- ceived correctly more often. The method of presentation does not af- fect the per cent of correct responses but the durations necessary to make these judgements do vary with the method of presentation. The subjects show a tendency to use some counting method to esti- mate the time of duration of sounds. The study of the perception of duration seems to leave itself open for much further research. A STUDY OF THE PERCEPTION OF THE DURATION OF CONTINUOUS, WARBLED, AND PULSED SIGNALS By Patricia Sheafor A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS College of Communication Arts, Department of Speech 1963 ’m e TABLE OF CONTENTS LIST'OF TABLES . . . . . . . . . . . LIST'OF FIGURES. .. . . . . . . . . . n v . . . Chapter I. STATEMENT OF THE PROBLEM . . Introduction Hypotheses Importance of Study Definition of Terms II. REVIEW OF LITERATURE , , Time Perception . Pitch and Intensity Discrimination Time Error.. Duration . III. SUBJECTS, EQUIPMENT, AND TESTING PROCEDURES. Subjects Equipment. Procedure. IV. RESULTS AND DISCUSSION . Results. Discussion V. SUMMARY AND CONCLUSIONS. Summary. ii Page iv H CDO\O\4-‘N .11' . 12 .15 .15 .17 ~22 .22 .29 .32 .32 TABLE‘OF'CONTENTS (Continued) Page Conclusions . . -. . . . . . . . . . . . . . . . 3} Implications for Future Research . . . . . . . . 3h BIBEIUGRIPE!“ ... . . . . . . . . . . . . . . . . . . . . 36 APPENDICES .. ... . . . . . . . . . . . . . . . . . . . . 38 iii Table 1. 2. LIST OF TABLES Analysis-of-Variance Table for One Variable. Mean Percentages of Correct Responses. . Summary of Analysis-of-Variance Random Order of Presentation of Test Stimuli on the Original Test Tape . Subject Response Sheet Raw Scores -- Continuous, Iarbled and Pulsed Data . iv 0 Page 22 23 23 39 #3 Figure 1. LIST OF FIGURES Record and Timing System Employed in Production of Test Tapes . . Playback System for Presentation of Test Tapes. . . . . . . . Graphic Representation of Responses for Continuous Data Graphic Representation of Responses for Warbled Data. . . . . . . Graphic Representation of Responses for Pulsed Data . . . . . . . . . Page l6 17 25 26 27 CHAPTER I STATEMENT OF THE PROBLEM Introduction The research reported in this thesis has as its purpose to ascer- tain the accuracy of human perception of the duration of various audi- tory signals, which though of the same frequency and intensity vary in duration and method of presentation. The body of this material contains a statement of the problem which will incIude a hypothesis, a discussion of the distinctiveness of the study, and a definition of the pertinent terms used in this re- search.' This is followed by the review of literature pertaining to the area of sound or signal duration. A discussion of procedures of the study includes (l) a description of the subjects, (2) enumeration of the equipment utilized, and (3) reports the procedures of the exper— iment. The statistical and graphic analysis of the data and a discus- sicn of these procedures are included as the.fourth section of the thesis. The results and significant data are discussed in the.summary and conclusions of the study. _Implications for further research are also discussed. Finally, the raw data, and samples of procedural items may be found in the appendices. Hypotheses The initial impetus for this study came from a discussion of a portion of a research study being performed for the Office of Civil De- fense within the Speech and Hearing Science Laboratory at Michigan State University, East Lansing, Michigan. Iarning signals.and their intensity, duration and alerting properties were under investigation. The question which was raised pertained to the effect of length and type of signal presentation on judgement of signal duration. This discussion transpired into the topic under investigation in the pres- ent study. The specific questions raised for study at this time are: I. How accurately are human beings able to judge the dur- ation of signals of the same frequency and intensity which vary random- ly as to duration and method of presentation, i.e., continuous, warbled, or pulsed? 2. with reference to the units of duration, how great a difference is necessary to make a judgement of difference in duration? 3. Are signals of one type, i.e., continuous vs. warbled nor pulsed, more easily differentiated than those of another type? A The third of these three questions lends itself to statistical analysis. The following hypothesis has been formulated on the basis of the third question to facilitate the analysis. There is no.difference in the per cent of correct judgements of duration as a function of the method of signal presentation. Igportance of Study It becomes evident as one reviews the literature that there has been very little research done in the specific area of judgement of signal duration. This may be more fully seen in the review of litera- ture'whichgis“presented in Chapter II. - There have“been numerous studies in the areas of acoustic.per- caption, detection of pitch and discrimination of pitch and loudness. "Rnwever,vthe*majerityof.these studies have made very little,.if any, ' “mention of the methods used to-determine the duration of their.sigu naIs.or the effects which these durations had on the auditory per- ception of their subjects. There are a few studies which hare dealt with signal duration but these have either dealt with signals of such short duration that they are classified as clicks, or they have stud- ied thererception of duration of signals introduced in a background of white: noise . . In the review of literature Hirshl.makes.a statement that more of the perceptual system than the ear itself is required to make judge- ments involving auditory perception. Current audiometric tests util- ice.the perception of tones and small changes in their.intensity as a diagnostic tool in the determination of the location of the lesion.2 It would seem that further research in the area of perception of sound dur- ation might well lead to new diagnostic tools for central auditory func- tions. Thus.it would:seem that this study is justified not only because it deals with signal duration of at least 6.6 seconds nith.no background noise but.because this area of investigation certainly is a little stud- ied one. It is hoped that this investigation will yield information 1Ira J. Hirsh, "Auditory Perception of Temporal Order," The Jour— ,_n§l;of the Acoustical.Society of America, XXXI (June . 1959), pp. 7593767. 2James Jerger, ”Hearing Tests in Otologic Diagnosis," ASHA, IV (May, 1962). pp. 139-1u5. which will be valuable in terms of adding to the fund of information concerning humanraudition. Definition of Term; There are a number of terms which will be used throughout this discussion that will need to be defined. The Psyghophysical Method of Constant Stimuli is the method which has been utilized in the presentation and analysis of stimuli and data in this study. Hirsh defines psychoghysics as: ;the study of the relations between physical.stimuli and the ‘responses'to which they give rise . . . .The method of con- stant stimuli is one of three classic psychophysical precede urea for measuring the differential or absolute threshold. Stimuli are presented in discrete categories of the independ- ent stimulus dimension. ‘The observer responds with a "yes" or "no;” "same" or "different" after each stimulus presenta- tion..3' Applying this definition to this study we find that the stim- uli were divided into two categories. The reference signal which was always presented first was 1000 cps, constant.stimuli 7.6 seconds in duration. ‘The variable signals were randomly arranged in duration and method of presentation and ranged in duration from 6;6 - 8.6 seconds for the continuous and warbled signals and 6.2 - 8.6.seconds for the.pulsed data. The subjects responded with judgements of "longer,” ”equal" or "shorter” after the presentation of each stimu- lus.pair. The Difference Limen (DL) is one of the methods of analysis of psychophysical data. It is ". . .the increment in a stimulus which is just noticed in a specified fraction of the trials."l+ lhen three , 3Ira J. Hirsh, The Measurement of Hearing (New York: McGraw- 3111 Book Company, Inc.,"l‘952), pp.‘3i¢1,"'339. itIbid., p. 336. category judgements, i.e., greater, equal, and less, are used the DL may be defined as ". . .the interval of uncertainty IU (or half of it) between the 50-per-cent point on the 'less' and the SO-per-cant point on the 'greater' curve."5 Guilford states simply that the DL is a stimulus.difference that is noticed fifty per cent ofthetime.6 The Point of Subjective Equality (P3E7"indicates.the central tendency of the judgements. It may be computed as a by-product when limens are the chief point of interest.7 The three methods of presentation of stimuli should be defined in order to clarify exactly what methods were used in this study. The continuous tone was a pure tone of 1000 cps which did not fluctuate as to amplitude or frequency. It was presented as the ref- erence.signal throughout the study but was also used as a test stim- ulus. The warbled tone had a center frequency of 1000 cps and a 5% warble at a rate of 3 and 1/3 cps. The pulsed tone was also a 1000 cps tone having an off time of three tenths of a second and an on time of three tenths of a second. All of these tones were delivered by the same pure tone oscillator from an Allison Audiometer Model Zl‘B. w—V 51bid., p. 13. 6J. P. Guilford, Psychometric Methods (New York: McGraw-Hill Book Company, Inc., 195%), p. 22.‘ 71bid., p. 25. CHAPTER II 'REVIEW“0F“THEmfiITERITURE Time and its duration are of intense interest to the human race. There never seems to be enough time or perhaps it is simply that man perceives time with some constant error in his method of perception; The duration of time as perceived in the routine of daily life seems to be affected by factors which are both psycholog- ical and physical in nature. The desire for a better control of time, an understanding of its duration, and a comprehension of its effects upon man physically and psychologically has been a constant source of research in many areas of knowledge. In an attempt to gain a better insight into the area of time study undertaken in this research, the writer has undertaken the fol» lowing.review of literature. The studies of time have been numerous and-ascertaining.which have dealt with the perception of signal dura» tion has led the uriter to.divide the pertinent literature into seVb eral related areas, (l).time perception, (2) pitch and intensity dis- crimination, (3) time error and (k) duration. Time Perception A study by loodrow1 on time perception has.been valuable in gaining an over all perspective of the areas involved. v—v WV 1Herbert Woodrow, "Time Perception." Handbook of.§§gggimental Ps cholo . ed. 8. 3. Stevens (New York: John liley and Sons, Inc., 1953). 7 Chapter 32, l22h-1235. Woodrow discusses the temporal stimuli ordinarily used in time studies and describes them as one of two types: (1) empty in- tervals which are bounded by, for example, two short flashes of light orwsound and, (2) sounds or lights lasting continuously over a period of time. The accuracy of discrimination of both the intervals and durations of these types of stimuli have been made for (1) empty in- tervals bounded by clicks, (2) continuous tones and (3) continuous lights. In-a study reported by Woodrow which utilized continuous tones the just noticeable differences between tones was approximate- ly 10 per cent to 12.5 per cent of the-standards varying from 0.k - 2.0 seconds and a considerably larger one for durations of 4 - 30 seconds.2 Time perception is the result of a special type of reac- tion to a stimulus. This reaction which facilitates the judgements when comparisons must be made, are affected by objective or environ- mental factors as well as by the attitude of the subject. The ob- jective factors include the arrangement of the experimental condi- tions, whether the subject sees a light prior to the onset of the tone, thether he has had practice in making this type of judgement and how.long he.has been involved in the experimental process. The attitude of the subject refers to the manner in which.the subject ‘goessabout his task. Some subjects perceive and describe the stim- ulus as a whole, others.seem to pay attention to the.first stimulus and then, in some manner, start reproducing it as the second stimu- .lus.begins. loodrow continues his discussion of various time study prob- ""‘ ‘ '.' 21bid., p. 1225. lems and concludes ception. The term to the duration of cal duration. One protensity depends the extensity of a with a section dealing with theories of time per- protensity is used in this section and pertains which one is aware, as distinguished from physi- view of time perception holds that the amount of on the physical duration of the experience, as horizontal line depends on the length of the line. However, the concept of protensity has not led to any specif- ic theory to explain the comparison of durations. Another theory holds that perception of duration can only arise with change. It suggests that time is always indirectly judged by some process which serves as a cue. These cues may be a fading memory image, or a brain trace or they may be connected with attention in such a way that physiologic changes, chiefly some sort of strain, are utilized as the basis of judgements of durations. Some authorities include breathing sensations and counting which is considered to be the most common and accurate method for estimating time in minutes, as temporal cues. In conclusions Woodrow states: Time is not a thing that, like an apple, may be perceived. Stimuli and patterns of stimuli occupy physical time; and we react to such stimuli by perceptions, judgements, compar- isons, estimates, etc. Whether some mental variable as dura- tion or protensity is an immediate prOperty of our perception of temporal stimuli, or of mental processes in general, is a matter of some given prOperty, like the value disagreement. If there is no such immediately it follows that time is a concept, somewhat of pieces 0f money, that attaches to percep- tions only through a judgmental process. 3 Pitch and Intensity Discrimination This area of study was included in this research not basically 3Ibid., p. 1255. for its content but in an attempt to determine what intervals and durations were used in previous studies dealing with time. Green, g£_al.u undertook a study to determine how signal amplitude and duration effect the detectability of a pure tone which was partially masked by random noise. Three experiments were performed. In the first, signal duration was held constant while amplitude was varied; in the second study, signal energy was constant while duration and amplitude were varied; and in the third, amplitude was held constant and duration varied. Not all the find- ings of this study were pertinent to the present investigation. A few interesting points were made with regard to duration. The au- thors state that until recently (1956) very little research has been done with signal duration, and the question of how duration ef- fects the hearing mechanism is an important aspect of the study of duration. They included a rather extensive coverage of previous studies concerned with signal detection and the writer was inter- ested in these chiefly for the time intervals stated. Yet even these were not specifically applicable to the present study. 5 In a study conducted by Hirsh, he examined the judgement of temporal order and the kinds of physical changes that are neces- sary in order that there be perceived changes. The stimuli util- ized in this research were very short and were classified as clicks. “David M. Green, Theodore G. Birdsall, and Wilson P. Tanner, Jr., "Signal Detection as a Function of Signal Intensity and Dura- tion," he Journal of the Acoustical Society of America, XXIX (April, 1957) . pp 5 523-531". 5Ira J..Hirsh, "Auditory Perception of Temporal Order," The Journal of the Acoustical Society of America, XXII (June, 1959), pp. 759-757- 10 He found that a temporal separation of 2 msec. is sufficient to give rise to the perception of two sounds and that a separation of 20 msec. enables subjects to determine 75% of the time which of the sounds came first. He concludes that more of the perceptual system than the ear itself is required to make such a judgement and thus temporal order judgements do not seem to be closely related to factors which have been shown to be important for the peripheral auditory‘system. Konig's6 study deals with the effect of interstimulus in- terval on pitch discrimination. Here again the chief points of interest to the writer had to do with the duration of signals. Konig points out that the pitch of a tone depends on its duration and that as the duration is lessened the tonal characteristics dis- appear and the tone sounds like a click. He found that the best performance of his subjects was attained with an interstimulus in- terval of 2.5-5 seconds. However, the studies cited previously have used intervals of from 20 mesc to 4 seconds (Hirsh, Green, Woodrow). The actual duration of the test stimuli in these studies have varied as much as the intervals. Thus one is left with little specificknowledge of which durations are most applicable to the present research, and also with very little understanding of the factors used to determine the above mentioned intervals and durations. 63. Konig, "Effects of Time on Pitch Discrimination Thres- holds under Several PsychOphysical Procedures; Comparison with In- tensity Discrimination Thresholds," The Journal of the Acoustical Society of America, XXIX (May, 1957), pp. 606-612. 11 Time Error The function of the time error in psychophysical methods is of importance and two rather extensive studies were reviewed on the subject. 7 Koester's study was done in relation to pitch and loudness discrimination but deals to a great extent with time error. lhen a subject involved in a psychophysical eXperiment compares two successive stimuli which are equal or slightly different in qual- ity or intensity, the second stimuli is often over or underesti- mated, with reference to the first. Over-estimation is the more common type of error and is referred to as a negative time error. When the second stimuli is under-estimated this is referred to as a positive time error. The study contains eight experiments which deal chiefly with the pitch and loudness discrimination. The results which dealt with time-error seem to indicate that time-error cannot be accounted for by a single cause. The major sources of the con- stant error seem to be (1) stimulus level, (2) the type of impres- sion, (3) amount of practice and (h) experimental procedure. It is also pointed out that the amount of time between stimuli does not seem to have as much effect as the above mentioned factors.on the time error. Another study dealing with time error is that reported by 7Theodore Koester, "The Time Error and Sensitivity in Pitch and Loudness Discrimination as a Function of Time Interval and Stim- ulus Level," Archives of Psychology, ed. R. S. Woodworth, No. 297, (May. 19#5). p. 69. 12 Stott.8 The purpose of his investigation was to ascertain the oc- currence of time-error which is found in the comparison of short continuous tones. He was also interested in the location of the indifference duration which he defines as the duration at which time-errors disappear. Most of the work was concentrated on dur- ations of 0.5-2.0 seconds. The two major findings relevant to the present study in- dicate that the previous experience of a subject in comparing durations is a very important factor in determining time-errors. Also Stott feels that in order to minimize the effects of exper- ience large numbers of subjects should be used and only the judge- ments made in the early portion of the experimental sessions should be considered. Duration Studies dealing specifically with the perception of dura- tion of signals seem to be difficult to find. In her research the writer has found only two studies which have had duration as their chief point of concern. The first study concerned with signal duration was done by Small'and Campbell and the results were published in A232 Psychologies.9 Thus the information available is rather limited. The subjects in the study were asked to judge whether the second number of a stimulus pair was longer or shorter than the first. 8 L. H. Stott, "Time-Order Errors in the Discrimination of Short Tonal Durations," Journal of Experimental Psychology, XVIII (1935), pp. 7&1-766. 9A. M. Small and R. A. Campbell, "Differential Sensitivity for Temporal Intervals with Auditory Stimulation," Acta Psycholog- ica’ XIX (1961), pp. 557-5580 13 The standard stimuli was maintained constant and the duration of the variable stimulus was randomly varied. The findings include the following: (1) the inter-stimulus interval became less im- portant as a variable as duration lengthened, (2) listeners gen- erally over-estimated the duration of the standard stimulus and (3) the data of the study did not make it possible to determine whether listeners were basing their judgements on duration alone. loudness alone or a combination of the two. If loudness is the determining factor in such judgements then it should be much more carefully controlled. The second study found by this researcher was a Ph.D dis- sertation entitled "Human Discrimination of Auditory Duration."lo The study utilized a two-alternative-forced—choice pro- cedure. Two sine-wave signals of the same amplitude and fre— quency which differed only in duration were presented in sequence at each trial. The order of the presentation of stimuli was ran- dom and the subjects' task was to state, for each trial, which of the signals had occurred first, the longer or shorter one. The signals were presented in a background of white noise. The results of this study indicate that duration measure- ment was accomplished by a "counting mechanism" and that the source of these impulses was random. Limitations on perform- ance were assumed to come iron uncertainty regarding the end- 10 C. D..Cree1man, "Human Discrimination of Auditory Dura- tion," (Ph.D. dissertation, University of Michigan), Dissertation Abstracts, XXII, (July, 1961), p. 331. gift; a Origins-nun 13'an”... .1... HIM»! cluuurwflfi.‘r 1# points of the time interval, and from memory limitations. No definite findings are stated as to the perception of duration by the subjects in this study. CHAPTER III SUBJECTS, EQUIPMENT, AND TESTING PROCEDURES Subjects The twenty-five subjects utilized in this study were stu- dents at Michigan State University, East Lansing, Michigan. Thirteen of the subjects were not affiliated with the area of Speech’and Hearing Science, the remaining twelve subjects were graduate students in Speech and Hearing Science. AII of the sub- jects were screened at 15 db at 1000 cps, the frequency involved in the test procedure, and were judged to have normal.hearing for that frequency. There were thirteen female subjects and twelve male-subjects.ranging in age from twenty to forty-six years. The mean age of the subjects was 25.6 years. Equipment .Hearing screening for all subjects was accomplished with an-audiometer. (Audiovox Model 7 BT, serial number 2232) with a matched pair of earphones (Telephonics"TDHé39). The equipment utilized was: 1. Audiometer, (Allison Model 21 B); 2. Tape recorder, (lollensak Model T 1500); 3. Recording tape, (Scotch Tenzar Backing Magnetic Tape 311); #. Tape recorder for the production of the test tapes, (Ampex Model 601); 15 16 5. Tape recorder for the presentation of the taped stimuli, (Ampex Model 602); 6. Earphones, a panel of twelve binaural earphones, (Telephonics TDH-39); 7. Timer, (Hunter Timer, Model 100 B) which was capable of presenting pulses at lengths var- iable from one millisecond to ten seconds in steps of one millisecond and, 8. Graphig Level Recorder (Bruel and Kjaer Model 2305 . Two block diagrams of the.equipment used for the recording and presentation of the test tapes follow. Figure 1 -- Record and timing system employed in production of test tapes. 1- 3w.— ”“2 31:1 “P‘ _ FA. r Recorder :tglnfmt | x e ___°“1"‘F: “ (4) (3) T H E (2) R __—__T 'fiape DeCK (1) 17 Figure one illustrates the instrumentation for the tape re- cording of the test stimuli. It includes, as numbered in the fig- ure, (1) the tape recorder (Ampex 601) from which the taped stimu- lus tones were sent through (2) the timer (Hunter Timer, Model 100 B) where the duration of the signal was determined, into (3) an Ampex Stereo Mixer from whence the signals progressed to (4) a tape recorder (Ampex 602) on which the test tapes were recorded. Figure 2 -- Playback system for presentation of test tapes. n. Amp Lme 9m (1) (2) (3) (h) Figure two illustrates the equipment used for the playback system. Here the (l) tape recorder (Ampex 602) was the source for the signal presentation, (2) the Ampex Stereo mixer is again util- ized, (3) two amplifiers are added and (h) the test stimuli are fed into the panel of twelve earphones (Telephonics TDH-39). Procedure As the subjects entered the testing room, their hearing was screened by the researcher. The test procedure was carried out in the Speech and Hearing Science Laboratory at Michigan 18 State University. This area was relatively quiet throughout the experimental proceedings. The pure tone stimuli used in the test tapes were present— ed in pairs and distributed as follows: A-C were practice pairs at 500 cps, 1-165 were test pairs at 1000 cps. The first tone in each test pair was always a continuous signal and was considered to be the reference tone. This tone always was 7.6 seconds in length. The second tone in each pair ranged randomly in length from 6.6 seconds to 8.6 seconds and varied also as to method of presentation. All pairs were presented to the subjects at 80 db relative to 0.0002 dyne/cma. The subjects were seated in a panel of twelve chairs with the-earphones plugged into jacks on each chair. The subjects were given a response sheet (sample in appendix) on which to place their judgements and then were given the following instructions: le are performing this test in order to ascertain how ac- curately length of sound is perceived with varying types of “signals. You will hear three types of signals--continuous, pulsed and warbled. They will be presented in pairs with the first sound always being the reference tone. You are to make a judgement as to the length of the second sound as compared to the first or reference signal. You may use any method of measurement you desire provided it does not involve a watch or does not disturb ydur neighbors. Thus tapping your finger or foot is not permissible unless you can do it unobtrusively. Your response sheet is made out in such a way that all you must do is mark the apprOpriate square following the num- bered pair. You will be told the number of each pair prior to hearing the signals. The first three pairs of sounds are 'practice pairs. He will do these first to acquaint you with the procedure. Place a mark in either the longer, equal or shorter column when you have made your judgement as to the length of the second sound as compared to the first. 19 Immediately following these instructions the subjects ad- justed their earphones and the three practice pairs were initiated. The practice pairs differed from the test pairs in that they were made up of 500 ops tones and the differences in duration were of greater magnitude, being of at least one second in length. Upon completion of the three practice pairs, the subjects were told what the responses should have been. If there were any subjects who had not made any correct responses to these gross differences, the practice items were given again. Following the practice items the.subjects were asked to remove their watches and the following instructions were given. Now we will start the experiment. You will have to pay very close attention because many of the differences are much smaller than those you heard in the practice pairs. As there were twenty-five subjects, the testing was per- formed in three separate sessions. Four subjects were tested in- itially, nine subjects in the second session, and twelve in the final testing session. The testing time was approximately one and one half hours, with a short break at the half-way point while the tapes were being changed. It was felt-that the subjects who heard the test tape in sequence from 1-165 might do poorly on the first part of the tape due to unfamiliarity with the task, and also on the second half of the test due to fatigue. However, they might also profit from the learning experience of the first half of the tape and so show im- provement on the second section of the tape. Therefore, the first thirteen subjects heard the test tape in sequence from 1-165. The second group, consisting of twelve subjects, heard the test tape 20 from 1-0, 82-165, and 1-81. This procedure was followed in order to.minimize the learning and fatigue factors that might occur. After the test procedure was completed, the subjects were asked to note on the back of the response sheet which, if any, method they had used to time the signals. Upon completion of the experimental testing, the exact duration of the pulsed stimuli was brought under question due to its characteristic three tenths' of a second "on-off” pattern. Thus,rthe duration of the pulsed signals could only be exactly as planned if the timer happened to turn on exactly in the mid- dle of an "on" pulse and also turn off in the middle of an "on” pulse. The pulsed signals could vary as much as six tenths of a second from the expected duration depending on the point at which the timer initiated the signal. In order to determine the recorded characteristics of the pulsed signals, it was decided that all pulsed signals should be put through the Bruel and Kjaer Graphic Level Record- er to test their durations. During this investigation it was found that (l) the Hunter timer had produced a constant error of four tenths of a second and had thus shortened all signals by this amount, (2) the duration of the pulsed signals varied greatly from signal to signal and (3) that the signal relation- ships.for randomly selected continuous and warbled signals,‘,w though.shortened, were the same. Thus, the original time.inter- val of eleven steps in duration from seven to nine.seconds was modified to be eleven steps in duration with two tenths of a second difference from 6.6 seconds to 8.6 seconds. There are 21 nineteen steps in duration of the pulsed data ranging from 6.2 ‘seconds to 8.6 seconds. The exact order of the original tapes may be found in the appendices with the apprOpriate revisions made for the pulsed data. CHAPTER IV RESULTS AND DISCUSSION Results The results of the experiment were tabulated and subjected \ to two statistical procedures. The analysis-of-variance technique as described by Dixon and Massey1 was utilized. The formuli are presented in the table below. Table l. Analysis-of-variance table for one variable m Sum of Squares df Mean Square Estimate of 2 2 Means 11‘ + — T++ i N K-l sma 5'2 + nd’m2 n1 2 2 2 Within 22 xij2- ‘- Tf N-k 6p 0’ hi ’4‘ Total 22 x132- T++2 N4 fN The data, over time, for the judgements for each method of presentation were converted to percentages and the one way analysis- of-variance technique described above was utilized to ascertain if there was any relationship between per cent of correct judgements of duration and methods of presentation. The tables below show the data involved in this analysis. 1Wilfred J. Dixon and Frank J. Massey, Jr., Introduction to” Statistical Analysis (New York: McGraw-Hill Book Company, Inc., 1957), PP- 1h5-152. 22 23 Table 2. Mean percentages of correct responses Continuous Iarbled Eulsedn w Mean ‘ 53089 SSORS 51e87 ' Table 3. Summary of Analysis-of-Veriance i - Source df BSD—J_ 8 MS 1 Columns 2 88.75 h#.37 .15# within 36 1031+l.88 fl 287.27 Total 38 10u30.63 The findings of this analysis indicate that there was no significant relationship between the per cent of correct judge- ments and the method of presentation. Following this procedure the data for each method of pre- sentation were plotted on graph paper and lines of best fit were drawn by inspection. Prior to determining the difference linens, a least squares procedure, as described by Blalock2 was utilized to verify the lines of best fit. It was found that those drawn by in- spection had in most cases been exact or very close to those cal- culated by using the formuli given below. b = 1;!!! - (21) (2!) NEXZ - (5.1)2 a = g! - ng N Yp = a + bx zfiubert M. Blalock, Jr., Social Statistics (New York: Mc- Graw-Hill Book Company, Inc., 1960), pp. 283-283. 2# The three graphs that follow on pages 25, 26, and 27, con- tain the data plotted with the calculated lines of best fit for the "shorter" and "longer" judgements. The lines of best fit for the ”equal" data were drawn by inspection. It was found after plotting the data that the results were skewed toward the short end of the time scale. This is due to the presence of a positive time error. In other words, the sub- ject's perceived the first or reference signal as being longer than it was. This caused them to make judgements of the duration of the variable signal in error as compared to the standard. They consistently judged the variable signal to be shorter than it was, thus skewing the data. Such a phenomenon is due to the bias which may arise when the stimuli are presented successively. In this study the first stimulus was always the reference signal and it would appear that this has caused the positive time error, meaning that the apparent magnitude of the first stimulus is enhanced as time passes.3 Koester discusses the "time-error" and states that: . . . whatever is retained (memory image, muscular set, trace. etc.) as a basis for a comparison judgement must undergo some sort of progressive change during the time-interval which e- lapses before the second stimulus is applied . . . Gestalt psychologists such as Koffka (19) have found here the experi- mental basis for an elaborate tract theory according to which the occurrence of the "time" error is attributed to the fact that sensory impressions are represented in the brain as elec- trochemical excitations or traces which undergo fading, dis- placement, and transformation with time. 4 sBurton G. Andreas, Experimental Psychology (New York: John Iiley and Sons, Inc., 1960), pp. 121-122. “Koester, loc. cit., pp. 5-6. Figure 3. Graphic representation of responses for continuous data Values 81 *1e5 _ +1.0 -2.0 l 25 x..._s Continuous shorter o___.= Continuous longer --.n.= Continuous equal a .. j x ‘fii RSE' L. \\\\\ Z 2970 ‘ll‘llfjll 141 6.6 6.8 7.0 7.2 7.8 7.6 f 7.8 8.0 8.2 8.u 8.6 Signal Duration in seconds Figure 4. Graphic representation of responses for warbled data a Values Si +1.5 +1.0 V 26 x__._: Warbled shorter o = Warbled longer --_.-= Warbled equal 1.4 .RSEE l. 2H0 Z5. 14: \\ ~2-0 J I J! l I I 11 1 - h 1 n gg4p_\e 6.6 6.8 7.0 7.2 7.4 7.6 7.8 8.0 8.2 8.4 8.6 Signal Duration in seconds h Figure 5. Graphic representation of responses for pulsed data +1.5 +1.0 Sigga Values ‘200 2? x—————= Pulsed shorter o = Pulsed longer ---— = Pulsed equal £50 PSEE 1-9 5.9:. M 7 7, m g 1 3 I I J, J' l [1,1 (A l l 11 n I 1 11 1 1 6.2 6°56.66‘7 7 7.27; 7.# 7.6 7.87.9 8.28.38.88.58.6 .o 7~1 7.5 7.7 Signal Duration in seconds 28 The positive time error is apparent in all three of the graphs but is especially noticed for the pulsed and continuous data. The present researcher did not design her study in such a way that the correction for time error could have been made by countsrbalancing. This would have involved presenting the refer- ence tone both as the first and as the second stimulus. This could have been accomplished, without informing the subjects of the possibility of change, and could have been allowed for in the scoring procedures. Difference limens were obtained for the three methods of presentation. The interval of uncertainty (IU) for the continu- ous data.was from a lower DL of 6.76 seconds to an upper limen of 7.70 seconds, an interval of .96 seconds. The DL being half of the IU, was .88 seconds. The warbled data had an IU of .56, from 7.40 to 7.96 seconds, and a DL of .28 seconds. The pulsed data had the largest IU of 1.02 seconds, from 6.82 to 7.88 seconds, with a corresponding DL of .51 seconds. The point of subjective equality (PSE) was computed using the arithmetic means of the "equal" data.5 The continuous PSE was 7.54 seconds, 106 seconds from the point of equality set in the study--7.6 seconds. The warbled PSE was 7.5 seconds, .1 of a second from the stated point of equality and the pulsed PSE was at 7.#7 seconds, .13 of a second from the point of equality for the study stated above. 5Guilford, loc. cit., p. 138. 29 Discussion In the statement of the problem in Chapter I, three ques- tions were raised with regard to the research to be accomplished. lith the completion of the compilation and analysis it becomes readily apparent that answers, or partial answers have been found to these questions. It also becomes obvious that other questions have been raised which open areas for future research. The analysis of the data for per cent of correct responses and their relationship to the method of presentation indicates that one method of presentation did not facilitate correct judgements of duration any more than another. This can also be seen by an in- spection of the graphs. Thus, one fails to reject the hypothesis proposed in Chapter I, i.e., there is no difference in the per cent of correct judgements of duration as a function of signal presentation. Subjects had greater difficulty differentiating the shorter stimuli and this would seem to be due to the presence of a posi- tive time error. This indicates that the duration of the first stimulus was.lengthened with the passage of time. Thus the lower DL's were larger for all methods of presentation than the upper DL's. In future research it would be interesting to.design the study and counterbalance the stimuli to eliminate time error as much as possible. The duration necessary for judgements of difference to be made, varies with the methods of presentation. These durations are expressed as the DL, and it must be kept in mind that the DL 30 is indicated by the boundaries of the interval of uncertainty, and at this time interval the subjects are only getting fifty-per-cent or more correct responses. It is difficult to determine the neces- sary duration for 100 per cent correct responses from the data col- lected in this study. The highest per cent correct was 88.8 per cent at 8.6 seconds on the continuous signal data. It would seem, however, that the difference in duration must be of at least one second in order to have 100 per cent correct responses. The reason for the difference limen being placed as it is from a fifty-per-cent point is because with a three choice re— sponse--longer, equal, or shorter--the subjects had a .33 chance of guessing correctly. Thus when the difference limen is drapped from a fifty-per-cent point this is greater than the probability of chance alone. It was found that the subjects required a duration of at least .28 of a second to make difference judgements for warbled signals, 388 of a second for continuous signals, and .51 of a sec- ond for pulsed signals. The point of subjective equality indicates what signal dur- ation the subjects felt was equal to the specified equal test stimulus. The PSE for pulsed data was 7.47 seconds, for warbled signals, 7.5 seconds, and for continuous signals 7.54 seconds. -An interesting by-product of this research has been an in- formal notation and tabulation regarding the methods used by the subjects to time the duration of the signals.- It will be remem- bered that the subjects were asked to note on the back of their re- 31 sponse sheet, what, if any, method they had used for timing. Creel- man states, "Duration measurment was assumed to be accomplished by a 'counting mechanism,‘ Operating on impulses generated over the relevant duration."6 It was found that the majority of subjects used some count- ing device such as marking, tapping, chewing, feeling their pulse or simply counting. A few reported holding their breath and a‘ few stated that they "felt" the duration of the signals. A great many subjects used more than one method of timing during the hour session. 6Creelman, loc. cit., p. 331. CHAPTER V SUMMARY AND CONCLUSIONS Summary Much of the emphasis in research on perception of time has been placed on discrimination and detection of various signals. The aspect of time study covered in this investigation, i.e., the perception of duration, has not been studied exhaustively. The study of perception of duration presents many possibilities for further research, as well as the possibility of being of benefit to man as a possible diagnostic tool for central auditory func— tions. The purpose of this study has been to ascertain the accur- acy of human perception of the duration of various auditory signals in an effort to gain a greater understanding and knowledge of this area of endeavor. .A review of the literature pertaining to this study indicates that little research has been performed with the duration of the sig- nals being the major concern. Duration has been a part of most stud- ies of time perception but has not been investigated in any great de— tail with reference to perception. Thus, this study seems to be one of a limited number. The subjects for this study were twenty-five college students with a mean age of 25.6 years. It was determined that all subjects had normal hearing for the frequency utilized in the study. The test 33 stimuli consisting of pairs of randomly arranged continuous, war- bled and pulsed tones of from 6.2 - 8.6 seconds in duration and of 1000 cps in frequency were transcribed onto magnetic tape and pre- sented to the subjects at 80 db (relative to 0.0002 dyne/cm). The first stimulus was always the reference signal. The subjects were instructed to make judgements of longer, equal, or shorter of the second signal with reference to the first; The subjects were al- lowed a short practice session and were tested in groups. The test procedure including instructions, practice and a short break was approximately one and a half hours in length. The findings of this study indicate that subjects had great- er difficulty discriminating duration of the shorter signals and that the method of presentation, i.e., continuous, warbled or pulsed, did not affect the per cent of correct judgements of dura- tion. A positive time error was found and would seem to have been caused by the lack of counterbalancing of stimuli in the study. Subjects were able to judge durations-of stimuli within .28 of a second for warbled stimuli, .#8 of a second for continuous stimu- li and .51.of a second for pulsed stimuli. Thus judgements of dur- ation were most easily made with warbled stimuli. __Conclusions Within the experimental framework of this investigation the folloming conclusions seem to be in order. 1. None of the three methods of presentation facilitated the per cent of correct judgements of duration. That is, the subjects seem to make as many correct judgements of duration for the continu- 34 ous stimuli as for the warbled and pulsed stimuli. 2. Subjects tend to overestimate the duration of a constant reference tone, thus presenting results skewed by a positive time error. 3. Subjects had greater difficulty perceiving duration of the shorter auditory stimuli. 4. The duration of the auditory stimulus which was neces- sary for judgements of difference to be made, varies with the method of presentation, i.e., continuous, warbled or pulsed. 5. There is a tendency to use some method of counting to time the duration of signals. 6. It would appear that stimuli used in psychOphysical methods of constant stimuli should be counterbalanced. Implications for Future Research This study has reported several tendencies regarding.the per- ception of duration of auditory stimuli. There is as yet a great deal of information to be gathered on this subject. This study has in reality been a.pilot study with many questions remaining unan- swered. A few which might be considered in future research are: 1. What effect would a larger time duration of test stimuli have on the perception of duration of an auditory stimulus? 2. What is the duration necessary to obtain 100 per cent cor- rect responses in judgements of duration? 3. What effect does intensity have on perception of duration of an auditory stimulus? 4. How much does practice effect the judgements of duration? 35 Continued research in this area would add valuable informa- tion to an area of knowledge in the field of Speech and Hearing Science and Acoustics that has not as yet been fully explored. “h 5.73 BIBLIOGRAPHY Books Andreas, Burton 0. Experimental Psychology. New York: John Wiley and Sons, Inc., 1960. Blalock, Hubert M., Jr. ,Social Statisticg. New York: McGraw-Hill Book Company, Inc., 1960. Dixon, Wilfrid J. and Massey, Frank J., Jr. Introduction to Stat- istical Analysis. New York: McGraw-Hill Book Company, Inc., 19573 Hirsh, Ira J., The Measurement of Hearing. New York: McGraw-Hill Book Company, Inc., 1952.. Guilford, J. P. ,Psyghometric Methods. New York: McGraw-Hill Book Company, Inc., 1954. Stevens, 8. S. (ed.). ,Handbook of Experimental Psychology. loodrow Herbert. "Time Perception." New York: John Wiley and Sons, Inc., 1958. Periodicals Creelman, C. D. "Human Discrimination of Auditory Duration," Disser- tation Abstracts, 22 (July, 1961), p. 331. Green, David M., Birdsall, Theodore G., and Tanner, Wilson P. "Sig-- nal Detection as a Function of Signal Intensity and Duration," _ The Journal of the Acoustical Society of America, 29 (April, 1957). pp. 5233531. Hirsh, Ira A. ‘"Auditory Perception of Temporal Order," The Journal of the Acoustica1.Society of America, 3l.(l959), pp. 759~767. Jerger, James. "Hearing Tests in Otologic Diagnosis," ASHA, 4 (May, 1962), pp. 139-1t5. Koester, Theodore. "The Time Error and Sensitivity in Pitch and Loud- ness Discrimination as a Function of Time Interval and Stimu- lus Level," Archives of Psychology, 297 (May, 1945), 69 P. Konig, B. "Effect of Time on Pitch Discrimination Thresholds under 37 Several PsychOpnysical Procedures: Comparison with Intensity Discrimination Thresholds," The Journal of the Acoustical So- ciety of America, 29 (May, 1957), pp. 606-612. Small, A. M., and Campbell, R. A. “Differential Sensitivity for Temporal Intervals with Auditory Stimulation," Acta Psycho- logica, 19 (1961), pp. 557-558. APPENDICES 38 11 . 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23c 24. 25. 26. 27". 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38.‘ 39- 4o. 41. L720 43. Table 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000" 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 cps cps cps cps cps cps cps cps ops ops~ cps' cps ops cps cps cps cps cps CPS” cps cps cps cps cps ops cps ops ops cps ops cps cps cps ops cps cps cps ops ops cps ops cps cps 4. Random on Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous ContinuOus Continuous Continuous Continuous ~Continuous Continuous -Continuous Continuous Continuous Continuous Continuous rContinuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous O O O O O O O O O 0 0 0‘0 0 O O O O O O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\0\O\O\O\O\O\O\O‘\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O‘\O\O\ wnwwwnnuwwunnflvwwwnvuwwuwwnnwnnwwnflwwwnwwww APPENDIX A seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds 39 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 .1000 1000 ops ops ops cps cps cps cps cps cps cps cps cps cps cps ops cps cps cps cps cps cps cps cps ops cps ops cps ops ops cps ops ops cps ops ops cps cps ops cps cps ops cps cps order of presentation of test stimuli the original test tape. Warbled larbled Warbled .Pulsed O‘O‘O 0‘ seconds seconds seconds seconds Continuous 6.8 seconds Iarbled Pulsed Continuous 8.4 Misrecorded Pulsed Pulsed Warbled Pulsed Iarhled Warbled Pulsed Iarbled , Continuous Pulsed Continuous Pulsed Pulsed Warbled Iarbled Continuous Continuous Continuous Pulsed Warbled Continuous Warbled Pulsed Warbled Pulsed Pulsed Continuous. Continuous Pulsed Continuous Pulsed ' Warbled NU‘IOC CNNk‘OWONONm-t-F'CbmmmOflmcfiwm rooms-meow 0 O O 0 O t O o . g 0 e O o . O . nmmmouvmunnmxmwmwfimmwanqwunamvuu ‘7.8 seconds Continuous 7.2 Continuous 7.8 seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds 44. 45. 46. 47." 48. 49. 50s 51. 52. 53. 54. 55. 56e ‘EWR 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 7o. 71. 72.' 74. 75. "76. 78. 79. 80. 81. *82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93- 94. 95c 1000 1000 1000 1000 1000 1000 "1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 cps cps cps cps cps ops ops cps ops- cps ops cps cps cps ops ops cps cps cps cps cps cps cps *cps ops ops cps cps cps cps ops cps cps ops cps cps Ops ops cps ops Cps cps cps cps cps- cps CPB cps cps cps cps ops 40 APPENDIX A Table 4 —- Continued Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous ‘Continuous Continuous Continuous Continuous 'Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous I I I I I I I I I I I I I I I I I I I O I I I I I I I I I I I I I I ~ I I I I I I I I I I I I I D o O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\ \1\1\1\J\)\)\1\1fl\1\)\1\3\)\1\)\1\)\'l\Iflflflflflflflflflflflflflflflflflflflflflfl\1\J\1\)\)\)\J\1Q9 seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 Pulsed Warbled Pulsed larbled Continuous Pulsed Pulsed Pulsed Continuous Continuous Pulsed Iarbled larbled larbled Iarbled Pulsed Continuous Continuous Pulsed Iarbled Pulsed Pulsed Continuous Pulsed Pulsed Pulsed Pulsed 'Iarbled Pulsed Continuous Pulsed larbled Continuous Iarbled Continuous Pulsed larbled Continuous Pulsdd Continuous Continuous Iarbled larbled Iarbled Iarbled Continuous Continuous larbled Iarbled Continuous Pulsed Pulsed momomomtmmmmmmmmmom:Nvtwmmummmmummmomommrmowrrormumm eeeeO 'Q\1\)\1CD\'J moooommx': mqwummflmmo‘auo‘unuflv \3'\‘l OVQVODO‘CDQO‘N) moooooox'axzmoooooom seconds seconds .seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds .seconds seconds seconds seconds seconds seconds seconds seconds seconds ,seconds seconds seconds .seoonds seconds seconds seconds seconds seconds seconds 96 . 97. 98. 99. 100. 101. 102. 103. 10“ O 105. 106. 107. 108. 109. 110. 111.' 112. 113. 114. 115. ll6e 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137e 138. 139. 140. 141. 142. 143. 144. 145. 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 -1000 1000 1000' 1000 1000 1000 1000 1000 1000 1000 1000 cps ops ops ops ops ops cps cps ops cps cps ops cps cps ops cps cps cps cps ops ops cps ops ops ops ops cps ops ops ops ops' ops ops ops cps cps ops cps ops cps cps cps cps cps cps ops ops ops ops cps APPENDIX A Table 4 — Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous Continuous I I I I I I I I I I I I I I I I I I I 0 '\)\]\1\1\]\1\]\'I\'l\)\)\1\)\1\1\)\)\1\)\1\‘JflflQVQVQQVV\Iflflflflflflflflflfiflflflflflflflfl O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O‘\O\O\O\O\O\O\O\O\O\O\O\ seconds - 1000 seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 11000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 Continued cps ops cps cps cps cps ops cps sps cps ops cps cps cps cps ops ops cps cps 0P5 cps cps cps cps cps cps cps cps cps ops ops ops cps ops cps ops cps cps ops ops ops cps cps cps ops cps ops ops cps cps Continuous Pulsed Pulsed Pulsed Pulsed Continuous larbled Pulsed Continuous Continuous Iarbled Continuous. larbled Iarbled Continuous larbled Pulsed Iarbled Warbled Pulsed Pulsed Continuous Continuous Continuous larbled Pulsed Continuous Pulsed Continuous Pulsed Pulsed Continuous larbled Continuous Pulsed Continuous Continuous Warbled warbled Continuous Continuous Warbled Continuous Continuous larbled Warbled larbled larbled Continuous Warbled , eeseeeseeeeeeeeeeesssseesee OV‘Q'GJQV O\\)\]O\‘\J\)O\m\1\)m\1 O\\] mfiflflmflmmflmmm \Im eeeeeeeeeeeeeeeesee CD\1 O\\)\) OOO\OOOO\100\3\1 oooox‘nox seconds seconds seconds seconds seconds seconds seconds 'seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds seconds 42 APPENDIX A - Table 4 Continued 146. 1000 cps Continuous 7.6 seconds - 1000 cps Warbled 6.6 seconds 147. 1000 ops Continuous 7.6 seconds - 1000 cps Pulsed 7.2 seconds 148. 1000 cps Continuous 7.6 seconds - 1000 cps Pulsed 7.3 seconds 149. 1000 cps Continuous 7.6 seconds - 1000 cpsularbldd 6.8 seconds 150. 1000 ops Continuous 7.6 seconds - 1000 cps Continuous 6.6 seconds 151. 1000 ops Continuous 7.6 seconds - 1000 ops Continuous 7.4 seconds 152. 1000 cps Continuous 7.6 seconds - 1000 ops Continuous 7.4 seconds 153. 1000 ops Continuous 7.6 seconds - 1000 ops Pulsed 8.4 seconds 154. 1000 cps Continuous 7.6 seconds - 1000 cps Pulsed 8.5 seconds 156. 1000 ops Continuous 7.6 seconds - 1000 ops Continuous 8.0 seconds ”157} 1000 ops Continuous 7.6 seconds - 1000 cps Continuous 7.8 seconds 1584 1000 cps Continuous 7.6 seconds - 1000 ops Continuous 7.0 seconds 159. 1000 cps Continuous 7.6 seconds - 1000 ops Pulsed 6.7-seconds 160. 1000 ops Continuous 7.6 seconds - 1000 cps Pulsed 7.3 seconds 161. 1000 ops Continuous 7.6 seconds - 1000 ops Warbled 8.4 seconds 162. 1000 cps Continuous 7.6 seconds - 1000 cps Iarbled 7.6 seconds 163. 1000 cps Continuous 716 seconds - 1000 cps Continuous 6.6 seconds 164. 1000 ops Continuous 7.6 seconds - 1000 cps larbled 8.0 seconds 165. 1000 ops Continuous 7.6 seconds - 1000 ops larbled 8.0 seconds Practice pairs: a. 500 cps Continuous 5.0 seconds - 500 ops Continuous 4.0 seconds b. 500 ops Continuous 5.0 seconds - 500 cps Warbled 6.0 seconds c. 500 cps Continuous 5.0 seconds - 500 cps Pulsed 5.0 seconds APPENDIX B Table 5 Subject Response Sheet 1+3 .F was a nun mm mm mn.oL oa.oz ass 1 sea in as as sm.sz a..z r and em” mm. we . mm.oz w.oz 1 es. . one, as W as ss.oz s.ss mma ems so .n. so sm.oz m.oz :mH MNH mm mm mm.oz m.oz mma mma mm mm mm.oz s.oz ems ans - am as an.oz m.oz ond oma om on om.oz m.oz ass sea We as- am.sz s.sz wed wad mm mm mm.oz .0 mod mad um an .n m:H.oJ oHH.o mw.o£ mm.o$ .s m m a m a m a n m . oesnosm m . asses a . nomsoa “Ken 44 M mus“ mes mas _mw mm.oz s~.oz m sea sea :HH em en.oz wm.oz . mus med mas mm nm.oz mm.oz was mes was mm mm.oz 3m.oz and and Has Hm Hm.oz mm.oz one one oaa ow om.oz mm.oz mos mma mos ms ms.oz am.oz mos wma mos we w:.oz o~.oz sea Rs s2 2. 3 .oz miss one one mos ms 0:.oz wa.oz use an a: as 3 .02 T52 ems ems sea on e:.oz wa.oz nos mma mos mu m:.oz ma.oz med mma mos ms ms.oz :H.oz Hos ans son as H:.oz wa.oz owe one ooa on oe.oz wa.oz one mma mm mm W mm.oz ma.oz m m a m a m A m m a ,1 m . noenosm m - asses a - tempos Axum “om4 "oamz 45 APPENDIX C Table 6 Raw Scores - Continuous, Warbled and Pulsed Data 46 47 PPENDIX C Raw Scores — Continuous Data i:— a; ‘ Judgements Time' Longer Equal Shorter 6.6 19 33 68‘ 6.8 22 41 62‘ 7.0 27 47 51‘ 7.2 39 39 46‘ 7.4 37 56 ‘30‘ 7.6 49 54‘ ’ 22 7.8 58' 49 19 8.0 88‘ 31 6 8.2 67‘ 35‘ _ 23 8.4 106‘ 13* 5 8.6 111' 11 3 f. 1...; es- W Raw Scores - larbled Data. _‘1 4‘ 6.6 10 26 88’ 6.8 18 46 61* 7.0 28 42 79‘ 7.2 '17 51 57’ 7.4 42 45 38* 7.6 45 48' _ “32' 7.8 59* 36 28 8.0 ”58* 27 16 8.2 ”79* 34 13' 8.4 88: 29 8 8.6 105’ 17 3 ‘ Correct responses 43 APPENDIX C _ Continued Raw Scores - Pulsed Data“ “W Judgements Time Longer Equal Shorter 6.2 3 5 17‘ 6.5 5 4 16‘ 6.6 12 29 84" 6.7 18 39 69* 7.0 19 20 36‘ 7.1 9 a a. 7.2 25 31 44? 7-3 ‘73 60 66' 7.4 5 9 . 11: 7.5 8 10 7' 7.6 12 10‘ 3 7.7 12‘ 7 6 7.8 49° 32 19 7.9 104‘ 48 23 8.2 17‘ 5 3 8.3 11‘ 8 6 8.4 68‘ 24 7 8.5 54‘ 15 7 8.6 41" 6 3 _r ' Correct responses RS ”'TIT/ifiLHITILlllfi’ W RIES “1111111111?“