UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Development of ear asymmetries in dichotic listening Neufeld, Gordon Arthur 1971

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1971_A8 N48.pdf [ 3.45MB ]
Metadata
JSON: 831-1.0102036.json
JSON-LD: 831-1.0102036-ld.json
RDF/XML (Pretty): 831-1.0102036-rdf.xml
RDF/JSON: 831-1.0102036-rdf.json
Turtle: 831-1.0102036-turtle.txt
N-Triples: 831-1.0102036-rdf-ntriples.txt
Original Record: 831-1.0102036-source.json
Full Text
831-1.0102036-fulltext.txt
Citation
831-1.0102036.ris

Full Text

DEVELOPMENT OF EAR ASYMMETRIES IN DICHOTIC LISTENING by Gordon Arthur Neufeld B.A., University of Winnipeg, 1970 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in the Department of Psychology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August, 1971 In present ing th i s thes i s in pa r t i a l f u l f i lmen t of the requirements for an advanced degree at the Un iver s i t y of B r i t i s h Columbia, I agree that the L ib ra ry sha l l make i t f r ee l y ava i l ab le for reference and study. I fu r ther agree that permission for extens ive copying of th i s thes i s for s cho la r l y purposes may be granted by the Head of my Department or by his representat ives . It i s understood that copying or pub l i ca t i on o f th i s thes i s f o r f i nanc i a l gain sha l l not be allowed without my wr i t ten permiss ion. Department of P s y c h n 1 n S y The Un ivers i ty of B r i t i s h Columbia Vancouver 8, Canada Date A u g u s t . 1971 Abstract Two hundred and eight Ss from grades 2, 4, and 6 were tested for recognition of dichotically presented musical stimuli, sound effects, and CVC nonsense syllables differing in medial vowel or i n i t i a l stop consonant. Ear asymmetry was found to increase from grade 2 to grade 6. The l e f t -ear advantage found for music, sound effects, and vowel-varied stimuli was due to decreasing right-ear performance with age. A right-ear advantage for consonants was the result of increasing right-ear performance and a simultaneous decrease in left-ear performance with age. These results were discussed in terms of a unilateral dominance specific for speech as opposed to a bilateral dominance for both speech and nonspeech material. Sex differences were found in the development of ear asymmetry, g i r l s showing ear asymmetry earlier than boys in the recognition of verbal material and boys showing ear asymmetry earlier than g i r l s in the recogni-tion of sound effects. The results of the study were compared with those of a similar study using adults as Ss. The comparison showed that substantially larger ear asymmetries were obtained with grade 6 Ss than with adults. This difference was found to be due to the children's inferior recognition of stimuli presented to the nonpreferred ear, preferred ear performance being the same for both groups. The possibility of a covert order of report factor influencing the magnitude of the ear asymmetry found was suggested. TABLE OF CONTENTS Chapter Page Abstract i i L i s t of Tables v L i s t of Figures v i Acknowledgements v i i I L a t e r a l Asymmetries i n Dichotic L i s t e n i n g : A Review 1 The Establishment of the Phenomenon 1 Ear Asymmetries with Speech St i m u l i 2 Ear Asymmetries with Non-Speech St i m u l i 4 Proposed Mechanisms. 6 Relation of ear asymmetry to Cerebral Dominance.... 6 The Response-Bias Hypothesis. 8 The D i f f e r e n t i a l - S t o r a g e Hypothesis 10 The Attentional-Bias Hypothesis.... 12 The Perceptual B i l a t e r a l Dominance Hypothesis 13 II L a t e r a l i z a t i o n of the Phonological Processes 17 I I I Method 35 Subjects 35 Procedure 35 Stimulus Materials 36 Design. 37 IV Results 40 V Discussion 54 D i r e c t i o n and Nature of the Ear asymmetries Obtained.. 54 i i i Developmental Trends in Dichotic Listening.. 57 References 64 Appendix • 69 LIST OF TABLES Table Page I Summary of Analysis of Variance of Recognition Scores 41 II Total Recognition Scores with Grade Effects Included. 51 III Total Recognition Scores with Grade Effects Excluded. 52 IV Summary of Separate Analyses of Ear Differences within Levels of Material, Grade, and Sex 53 ,.v LIST OF FIGURES Figure Page 1 Recognition performance as a function of the stimulus material presented 42 2 Increase in recognition of stimulus material as a function of grade 44 3 Ear preference as a function of stimulus material.... 45 4 Recognition scores of stimuli presented to the right and l e f t ears as a function of grade level and type of stimulus material 46 5 Sex differences i n the development of ear preferences 48 6 Nature of the development of ear asymmetries with grade effects included and with grade effects excluded for both right-ear advantage stimuli and left-ear advantage stimuli 50 v i ACKNOWLEDGEMENTS Thanks to: Jim Johnson and Ray Corteen who provided valuable assistance in the formulation and treatment of the problem, as well as i n the writing of the thesis. Dorothy Neufeld, who did much of the real work in collecting the data and who prepared the manuscript. Bob Wilson, who is responsible for my interest in these problems. Frank Spellacy, who gave his permission to use a tape he had prepared and used in a previous experiment. v i i CHAPTER I Lateral Asymmetries in Dichotic Listening: A Review In general, i t has been found that under conditions of dichotic^" stimulation, an ear preference is typically shown in the recognition and rec a l l of some kinds of auditory stimuli. There have been several alternative hypotheses proposed as to the mechanisms involved as well as the conditions necessary to produce these effects. It is hoped that the study described in this paper w i l l contribute to the understanding of ear asymmetries, especially in regard to i t s development and the l i t t l e under-stood left-ear advantage phenomenon. To provide the rationale for the experiment, a brief review of the relevant literature w i l l be presented as well as a review of the mechanisms that have been proposed to account for the findings. A separate chapter w i l l be devoted to a survey of studies that have pro-vided a detailed analysis of the speech signal in an attempt to discover the minimal conditions required to produce the phenomenon of ear asym-metry. The Establishment of the Phenomenon The procedure most frequently used in dichotic listening experi-ments was introduced by Broadbent (1954) to examine the limits of immed-iate memory on two separate channels. In his experiment, pairs of digits i L i c k l i d e r (1953, p. 1026) defines monotic as the stimulus applied to only one ear at a time, diotic as one stimulus applied simultaneously to both ears, and dichotic as different stimuli applied simultaneously, one to each ear. This usage w i l l be observed here. 2 were recorded with one digit on each track of a two-channel tape-recorder, and presented to the subjects through stereophonic earphones. The two digits in each pair were recorded in such a way that they began simultan-eously.; In his most typical condition, a group of three pairs of stimuli were presented on each t r i a l , after which the subject reported a l l the. numbers he heard in any order he chose. Broadbent discovered that i f the material was presented f a i r l y rapidly, the subject tended to give a l l the numbers from one ear before reporting any from the other. Kimura (1961b) was apparently the f i r s t investigator to relate dichotic listening performance to ear la t e r a l i t y effects. Using Broadbent's dichotic digits task, she found that normal right-handed subjects reported digits presented to the right ear more accurately than they did digits presented to the l e f t ear. She attributed the effect to the functional prepotency of the contralateral pathway from the right ear to the language dominant l e f t hemisphere. Thus, the stimulus from a dichotically presented pair which enters the ear contralateral to the hemisphere dominant for language is more li k e l y to be reported correctly than is the stimulus which enters the i p s i l a t e r a l ear. Ear Asymmetries with Speech Stimuli Since Kimura's original report, various investigators have accumu-lated an impressive body of evidence that the ear advantage obtained i n dichotic listening tasks is a real one. Several experimenters have replicated Kimura's finding using Broadbent's dichotic digits task or some modification of i t (Bartz, Satz, Fennel & Lally, 1967; Bryden, 1963, 1970; Dirks, 1964; Kimura, 1963; Knox & Kimura, 1970). Broadbent and Gregory (1964) have also obtained a right-ear superiority using a 3 multiple-choice recognition paradigm rather than a free recall paradigm. Following the dichotic presentation of three pairs of digits, four groups of three digits were presented di o t i c a l l y at the same rate, two of which corresponded to the digits presented dichotically. The task was to choose these two triads. The l a t e r a l i t y effect has been obtained with speech stimuli other than digits. Most of these experiments have used Broadbent's paradigm, only substituting alternative stimuli for digits. With this type of procedure, a right-ear advantage has been obtained with monosyllabic words (Curry, 1967; Curry & Rutherford, 1967), two syllable words (Bartz et a l . , 1967), and nonsense syllables (Curry, 1967; Curry & Rutherford, 1967; Kimura, 1967). Borkowski, Spreen & Stu.tz (1965) found the result using monosyllable nouns matched for i n i t i a l phoneme. This was replicated by Jones & Spreen (1967) using as subjects educable retarded children of six to twelves years of age. Using only one pair of words per t r i a l , Dirks (1964) obtained a right-ear advantage in a free r e c a l l paradigm, as did Knox & Kimura (1970) using monosyllable concrete nouns in an object identification paradigm. Using a multiple-choice recognition procedure with one dichotic pair per t r i a l , Kimura (1967) obtained a right-ear advantage with t r i s y l l a b i c nonsense syllables, and Kimura & Folb (1968) found i t with these t r i s y l l a b i c nonsense syllables inverted, i.e., back-wards speech. Several investigators have reported similar effects in tasks which are very different from the digits test and i t s modifications. In an experiment reported by Oxbury, Oxbury & Gardiner (1967), subjects repeated a continuous string of digits presented to one ear while trying 4 to ignore digits presented to the other ear. Irrelevant digits were paired with 25 or 50 percent of the attended digits. Errors were more frequent when the l e f t ear was the attended ear.. Treisman & Geffen (1967) presented two messages dichotically and asked subjects to tap to target words in either channel while shadowing one of the channels. When the right ear was shadowed, more words were correctly repeated. Corballis (1967) presented digits three at a time. Two digits were presented one to each ear, and a third was presented to both ears. Right-ear responding was more accurate than was left-ear responding. A significance test for this effect was not reported however. Studies splitting up the speech signal have also resulted in right-ear superiorities. These studies w i l l be discussed in the following chapter. Ear Asymmetries with Non-Speech Stimuli A similar phenomenon has been discovered with various nonverbal stimuli. With these stimuli, however, the ear advantage is reversed, identification of stimuli presented to the l e f t ear being superior to identification of stimuli presented to the right ear. Kimura (1964) presented unfamiliar woodwind solo passages dichotically to right-handed subjects i n a multiple-choice recognition paradigm and obtained s i g n i f i -cantly higher recognition scores for the l e f t ear. The same subjects showed right-ear superiority in the dichotic digits test. This discovery was related to Milner's (19,62) finding of more severe decrements in right-temporal lobectomized patients than in left-temporal lobectomized patients on a l l subtests of the Seashore Measures of Musical Talents. The most severe decrements were on the Tonal Memory (recognizing patterned 5 sequences of tones) and Timbre (tone quality) subtests. Shankweiler (1966) found a similar d e f i c i t in the a b i l i t y of right-temporal lobectomized patients to recognize familiar melodies. Subsequently, Kimura (1967) obtained a left-ear advantage where subjects were asked to sing dichotically presented familiar and unfamiliar melodies. In an attempt to isolate the components of music that are c r i t i c a l in the demonstration of an ear preference, Spellacy (1970) presented music and three types of musical components - temporal patterns, frequency pat-terns, and timbre in a simple recognition paradigm, but found a significant left-ear preference only for the complex music stimuli. A similar study by Spreen, Spellacy and Reid (1970) replicated the left-ear advantage for music and also found a significant left-ear preference for tonal patterns. Darwin (1969b), using pure tone sequences to form a melodic line, also obtained a left-ear advantage. Left-ear superiorities have also been found in other nonspeech tasks. In a study by Chaney & Webster (1966), subjects showed a left-ear super-i o r i t y i n identifying dichotically presented sonar signals. The same sub-jects who were found by Curry (1967) to be more accurate with the right ear in recalling dichotically presented words and nonsense syllables were better with the l e f t ear when identifying familiar environmental sounds presented dichotically. A left-ear superiority for sounds made by familiar objects was also found by Kimura & Knox (1970) for the subjects who had shown a right-ear advantage for digits and nouns. On the basis of the data discussed above, i t would appear that in general, subjects tend to be more accurate in identifying verbal stimuli presented to the right ear, and nonverbal stimuli presented to the l e f t ear in dichotic listening tasks. 6 Proposed Mechanisms Relation of Ear Asymmetry to Cerebral Dominance The evidence that the right-ear advantage in dichotic listening i s related to hemispheric dominance i s f a i r l y strong. Most compelling are data from a study by Kimura (1961a) which involved two groups of epileptic patients as subjects. A group of 107 patients who presumably were l e f t -hemisphere dominant for language produced averages of 86.5 and 79.8 per-cent correct responses for digits presented to the right and l e f t ear res-pectively. But a group of 13 patients who were right-hemisphere dominant (as determined by the Wada sodium amytal test) were 88.5 percent accurate on l e f t ear stimuli and only 78.0 percent accurate on right ear stimuli. Kimura presents analyses which indicate that this asymmetry in the direc-tion of ear advantages obtained from subjects with speech represented in the l e f t hemisphere and from subjects with speech represented in the right hemisphere, i s independent of the location of the patient's epileptogenic focus. Subjects which allow comparisons of right-handed and left-handed sub-jects are also relevant to the hypothesis of a relationship between cere-bral dominance and ear asymmetry, but must be interpreted with some caution. The relation between hand preference and cerebral dominance for language i n right-handed subjects i s straightforward; over 90 percent of a random sample of right-handed subjects can be expected to be l e f t -brained for language (Benton, 1965; Milner, Branch & Rasmussen, 1964). Most of the published studies of dichotic listening have used right-handed subjects exclusively, and these studies almost invariably have resulted i n right-ear superiority. What one would predict for l e f t -handed subjects i s not so clear, however. Evidence indicates that only • 7 about two-thirds of normal left-handed subjects have left-hemisphere language dominance and that left-handed subjects as a group tend to be less strongly lateralized than right-handed subjects, regardless of which side i s dominant (Benton, 1965; Milner et a l . , 1964). Both these d i f -ferences might be expected to lead to less pronounced right-ear super-i o r i t i e s when groups of left-handed subjects are studied in dichotic listening tasks using verbal material. The few studies which have com-pared performance of right-and left-handed subjects on dichotic listening tasks have found strong right-ear superiority in right-handed subjects, and less marked right-ear superiority (Bryden, 1970; Curry, 1967; Satz, Achenbach, Pattishal & Fennel, 1965) or no ear superiority (Bryden, 1965; Zurif & Bryden, 1969) in left-handed subjects. There i s some evidence that familial left-handed subjects are more li k e l y to be right-hemisphere dominant for language than the non-familial left-handed (Weinstein & Sersen, 1961) and that familial left-handed subjects are much more l i k e l y to show left-ear superiority in the dichotic listening situation than either right-handed or non-familial left-handed subjects (Bryden, 1967, 1970; Curry, 1967; Zurif & Bryden, 1969). On the basis of such evidence, most authors have accepted as probable a relationship between cerebral dominance for language and the occurrence of right-ear advantages with verbal material. They have not, however, agreed on the nature of this link. Some attribute i t to perception (Kimura, 1961b), some to short-term memory (Inglis, 1965), others to atten-tion (Treisman & Geffen, 1968). Other authors have chosen to explain ear asymmetries without reference to cerebral dominance. It is to their explanation that we w i l l turn f i r s t . 8' The Response-Bias Hypothesis Attempts have been made to explain the ear difference effect by claiming that there is a general tendency to report material entering the right ear before that entering the l e f t , implicitly denying the stimulus specificity of the direction of the effect (Oxbury, Oxbury & Gardiner, 1967). Several investigators have observed that information presented to the ear reported f i r s t (immediate or perceptual channel) is more accurately recalled than information presented to the ear reported second (delayed or storage channel) (Broadbent, 1958; Bryden, 1964; Inglis, 1965). If sub-jects therefore, i n a free re c a l l task tend to report more often from the right ear f i r s t , an ear asymmetry in favor of the right ear would be expected because of the failure to control for a bias in the order in which messages are reported. Although some investigators have indeed observed this order of report phenomenon in favor of the right ear (Bartz et a l . , 1967; Broadbent, 1954, 1957; Bryden, 1963), the evidence regarding the presence of such a tendency is equivocal. Satz et al... (1965) found a slight tendency for the l e f t ear to be reported f i r s t . And Inglis & Ankus (1965) failed to find any consistent tendency for subjects to give the material from the right ear f i r s t in a sample ranging in age from 10 to 70years. It thus appears that the preference for starting the report sequence with an item presented to the right ear is a less universal finding than the overall right-ear superiority. The response-bias model predicts that a l l lateral asymmetries w i l l disappear when order of report is taken into account. Several methods have been used in attempts to measure the magnitude of the ear superiority when the response bias is controlled. One s t a t i s t i c a l method is to score t r i a l s on which the f i r s t stimulus reported was presented to the right ear separately from t r i a l s on which a left-ear stimulus is reported f i r s t . Using the dichotic digits task, both Bartz et a l . ; 1 (1967) and Bryden (1967) found the right ear to be significantly more efficient regardless of whether stimuli presented to the right ear or stimuli presented to the l e f t ear were reported f i r s t . An alternative method of s t a t i s t i c a l control i s to score separately data from the f i r s t reported ear and data from the delayed ear. Comparisons are then made between right and l e f t ears as immediate ear and as delayed ear. Again, both Bartz et a l i , v(1967) and Bryden (1967) found the right ear to be superior as both the immediate and the delayed channel. Satz et al.., (1965) found right-ear performance to be somewhat superior when only the data /from the ear reported last on a given t r i a l were analyzed, although this difference was not significant (p < .20). The results from these studies using s t a t i s t i c a l controls for possible response-bias effects are not very extensive, but they are con-sistent in indicating a right-ear superiority. Another method of determining whether ear advantages can be obtained independently of a response-bias effect i s to use an ordered r e c a l l pro-cedure in which the subject reports the l e f t ear f i r s t on half the t r i a l s and the right ear f i r s t on half the t r i a l s . Bryden (1963) reports a significant right-ear superiority when such controls are used with a dichotic digits task. Borkowski et al.., (1965) used this procedure with one-syllable nouns as stimuli and found significant right-ear superior-i t i e s in delayed channel comparisons as well as in immediate channel com-parisons. Satz et al.-, (1965) obtained similar results using a four-pair-per-trial digits task. 10 Also relevant to the evaluation of the response-bias hypothesis are numerous studies which have ut i l i z e d paradigms that give no opportunity for serial order bias and yet result in significant ear difference effects. These include the studies in which the subject is asked to attend to and report only the stimuli presented to a given ear (Chaney & Webster, 1966; Kirstein & Shankweiler, 1969), studies using a forced-choice rec-ognition procedure (Broadbent & Gregory, 1964; Kimura, 1967; Kimura & Folb, •1968), and studies using a simple "yes-no" recognition paradigm (Spellacy, 1970; Spellacy & Blumstein, 1970; Spreen, Spellacy, & Reid, 1970). It would seem that there i s l i t t l e evidence to support the hypothesis that l a t e r a l i t y effects are due entirely to a response bias. However, i t has been demonstrated that order effects do occur and can accentuate l e f t -right differences in dichotic listening in the free-recall procedure. The Differential-Storage Hypothesis Inglis (1965) has suggested that the phenomenon of right-ear pre-ference in the r e c a l l of verbal stimuli may be explained by a differential storage d e f i c i t . According to this hypothesis, input arriving in the dominant hemisphere is less subject to interference or spontaneous decay than is input arriving in the nondominant hemisphere. That i s , short-term memory processes may be more efficient ; in the l e f t hemisphere than in the right. According to this model, lateral asymmetries w i l l increase as a function of time between storage and r e c a l l . When applied to an ordered re c a l l situation, the differential storage model would predict l i t t l e or no difference between accuracy on the l e f t ear when i t is given f i r s t and accuracy on the right ear when i t is given f i r s t . However, with material reported second, re c a l l of stimuli presented to the right ear would be more accurate than re c a l l of stimuli presented to the l e f t ear. Bryden (1967) in a reanalysis of data collected in a previous study (Bryden, 1963) found that there was at least as large a difference between ears on the immediate channel as on the storage channel. Borkowski et al.. (1965) found a right-ear superiority of 7 percent on the immediate rec a l l channel and one of 6 percent on the stor-age channel. Darwin (1969b) has also failed to find evidence that the ear difference effect i s any smaller on the f i r s t than on the second reported ear. In contrast, Satz et a l . . (1965) reported a 4 percent difference on the immediate channel and one of .11 percent on the delayed channel, supporting the differential-storage model. Upon closer analysis of this study, however, i t appears that the small differences between ears reported on the immediate channel may have been due to the ease of the task resulting in a "ceiling effect." Unless the task is made sufficiently d i f f i c u l t to produce a number of errors on the immediate r e c a l l channel, i t i s obvious that a right-ear superiority w i l l not appear. The differential-storage hypothesis may also be applied to the simple "yes-no" recognition paradigm introduced by Spellacy (1970). The model would imply that ear asymmetry would increase with longer intervals bet= ween the presentation of the dichotic stimulus and the recognition stimulus. Thus, memory decay in the nondominant hemisphere would be minimal at short intervals but increase with longer presentation-recognition intervals. Assuming that the same mechanisms would apply for nonverbal stimuli that are supposedly right-hemisphere dominant, Spellacy (1970) and Spreen, Spellacy & Reid (1970) conducted two studies using the same stimulus materials (music), varying only the length of the interval between the 12 presentation of the dichotic stimulus and the recognition stimulus. Intervals of 1, 5, and 12 seconds were used. It was found that difference between ears i s at a maximum immediately after the presentation of the stimulus and tends to decrease gradually with no significant differences occurring with an interval of 12 seconds. Thus the available evidence with the possible exception of Satz et a l . s (1965) would indicate that the differential-storage hypothesis i s untenable. The Attentional-Bias Hypothesis Treisman & .Geffen .'(1968) have suggested that the ear difference effect arises because of an unequal distribution of attention, the right ear being easier to attend to than the l e f t ear for verbal material. The implication i s that.if attention was controlled, l i t t l e , i f any, ear asymmetry would be obtained. Several of the investigations under dis-cussion have included conditions in which subjects were presumably attending, or at least attempting to attend, to only one ear while stimuli were presented dichotically. In one such study, Dirks (1964) found a significant right-ear superiority when subjects attended to both ears for digits or electronically distorted words as stimuli, but a smaller, non-significant right-ear superiority when subjects.attended to only one ear and heard distorted words. These results are consistent with the attentional-bias model. Most of the relevant studies, however, report data which are inconsistent with this model. Kimura (1967), using words and nonsense syllables as stimuli, found with both types of material that the group instructed to report the right ear only performed significantly better than the group instructed to report the l e f t ear only. Also, in the studies by Chaney & Webster (1966), Kirstein & Shankweiler (1969), 13: and Oxbury et al.., (1967, Exp. .11), subjects were instructed to attend to and report from only one ear at a time, and right-ear superiorities for verbal stimuli were obtained in each of these studies. If ear asymmetries were a result of an unequal distribution of attention, a reasonable expectation would be that sounds which are more easily separated by selective attention would show a greater ear d i f -ference than those which are more d i f f i c u l t to separate. Kirstein & Shankweiler (1969) however, found that when a subject i s asked to report sounds from a particular ear, he makes fewer errors of recall for vowel-contrasted nonsense syllables than for consonant-contrasted nonsense syllables, but in the same experiment, the latter sounds showed a greater right-ear advantage than did the former sounds. That attention may be a necessary factor for the occurrence of right-ear asymmetries is consistent with either a memory or a perceptual explana-tion of ear preference. Evidence that both perceptual and attentional factors may be involved in ear asymmetry is discussed in the following chapter. The attention hypothesis by i t s e l f , however, is inadequate in accounting for ear asymmetries, especially the reverse asymmetries shown for speech and non-speech stimuli. The Perceptual Bilateral Dominance Hypothesis Kimura (1961b) has argued that ear differences in dichotic listening tasks are primarily a perceptual phenomenon. According to this proposal, material arriving at the dominant hemisphere is more readily perceived than material arriving at the nondominant hemisphere. It would appear that this model can best account for the fact that ear asymmetries are obtained with procedures controlling for attention, order of report, and memory factors. The difference, however, between the perceptual model and, for example, the dif f erential-rstorage model, is very subtle and the differential implications of the two notions are not as distinct as might be desired. According to the perceptual hypothesis, errors of reproduc-tion are due to a failure of part of the input to enter the system. According to the differential-storage hypothesis, these errors of repro-duction are due to a decay or distortion of input after i t has entered the system. Although a perceptual model does not imply a denial of the effects of short-term memory decay, i t does assume that both inputs w i l l decay equally over time. In contrast, the differential-storage model implies that input to the nondominant hemisphere decays at a faster rate than i n -put to the dominant hemisphere. Thus the perceptual model in contrast to the differential-storage model would seem to predict ear asymmetries of the same degree on both the immediate and storage channels. With the pos-sible exception of Satz et. al.., (1965), the data discussed f u l f i l l this prediction (Borkowski et a l . , 1965; Bryden, 1967; Darwin, 1969b). For simple recognition procedures, the perceptual model would suggest that the difference between the two channels is at a maximum at short-term intervals with gradual but equal decay of both channels with time. Again, the results of Spellacy (1970) and Spreen et al.. (1970) clearly support such a model. Thus the perceptual model appears to be consistent with most of the data i n dichotic listening experiments. Whereas the previously discussed hypotheses have more or less ignored the left-ear advantage phenomenon for certain nonverbal stimuli, Kimura (1967) accounts for i t by postulating a right-hemisphere dominance for such material. According to Kimura, the auditory asymmetries reflect the functional asymmetry of the brain for the perception of verbal and non-15 verbal stimuli. Thus the superior right-ear scores for the perception of verbal material reflect left-hemisphere specialization for language functions, whereas greater left-ear scores for certain nonverbal materials reflect right-hemisphere specialization for nonverbal functions. An alternative explanation for the left-ear advantage phenomenon has been suggested by Spellacy & Blumstein (1970) and w i l l be discussed in detail in the following chapter. It is largely i n reference to this question that the present investigation was conducted. Kimura has based her explanation of perceptual asymmetry on a hypo-thesized superiority of the contralateral pathway over the i p s i l a t e r a l pathway in dichotic listening. Thus the right-ear superiority for verbal materials may be brought about by the "superior connections" between the right ear and the contralateral regions i n the l e f t hemisphere which are dominant for language. Conversely, the left-ear superiority for nonverbal materials may be a consequence of "superior connections" between the l e f t -ear and the contralateral regions of the right hemisphere dominant for those materials. Kimura suggests that this contralateral prepotency is based upon the greater number of contralateral neurons as well as upon inhibition of i p s i l a t e r a l neurons during dichotic stimulation. In regard to her f i r s t premise, evidence indicates that the crossed auditory pathways are stronger thanthe uncrossed pathways in cats and dogs (Rosenzweig, 1951; Tunturi, 1946) and in man (Bocca, Calearo, Cassinar & Migliavacca, 1955; Hall & Goldstein, 1968; Penfield & Rasmussen, 1950). Such a finding suggests that monotic stimulation should also result in ear asymmetries. Significant ear differences have been reported with monotic stimulation (Bakker, 1968, 1970), but many more subjects have been required to reach significance with this procedure. Corroboration for Kimura's second premise has come from the work of Milner, Taylor & Sperry (1968). The subjects in this study were right-handed patients for whom the main commissures linking the cerebral hemi-spheres had been sectioned to relieve epilepsy. Under dichotic stimula-tion, these subjects were able to report verbal stimuli presented to the right ear, but not those presented to the l e f t ; under monotic stimulation, they performed equally well with the two ears. Milner et a l . attribute the results to suppression of the i p s i l a t e r a l pathway from l e f t ear to l e f t (language) hemisphere during dichotic stimulation, and, of course, to sectioning of the callosal pathway that otherwise would have carried the l e f t ear input from right hemisphere to l e f t . Although the data discussed would seem to support a perceptual bil a t e r a l dominance model for ear asymmetries in dichotic listening, the evidence is far from complete. Recently a number of studies investigating the nature of cerebral dominance in speech perception have shed further light not only on the conditions necessary to produce a right-ear advantage and thus the mechanisms implicated, but also on the nature of hemispheric relations i n reference to ear asymmetries in dichotic listening. It i s to these studies that .we would now l i k e to turn. 17 CHAPTER II Lateralization of the Phonological Processes To say that a given hemisphere is the dominant hemisphere for language is attributing to that hemisphere a set of characteristics which is only vaguely understood. The present knowledge of the brain mechanisms which underly language in general is miniscule. Presumably as more is known about the conditions necessary to produce right-ear superiorities, more productive questions w i l l be forthcoming as to the mechanisms involved in ear asymmetries and hemispheric relations. It would, of course, be incorrect to suggest that nothing is known about language processes. One of the things known is that language i s heirarchically organized into levels of structure, each of which may be considered more-or-less independently of the others. One such level is that of meaning. The studies already mentioned using nonsense speech and inverted nonsense speech as stimuli demonstrate clearly that a right-ear advantage does not depend upon the stimuli being meaningful (Curry, 1967; Curry & Rutherford, 1967; Kimura, 1967; Kimura & Folb, 1968). The next logical step f i r s t taken by Shankweiler & Studdert-Kennedy (1967) was to pull the speech signal apart and to test i t s components in order to determine which aspects of the perceptual process depended upon lateralized mechanisms. The stimuli used were synthetic steady-state vowels. The syllables used were a l l combinations of the stop consonants lb, d, g, p, t, k/ with the vowel /a/. The fifteen possible pairs were presented dichotically, one pair per t r i a l , and the subjects were instructed to.report both syllables on each t r i a l , guessing i f necessary. A similar procedure was followed with five equal duration (300 msec.) 18 steady-state vowels (i,s£, ae, a, u) . A significant right-ear advantage was found for the stop consonants and a small, but nonsignificant right-ear advantage for the vowels. This right-ear advantage for consonants has been replicated for i n i t i a l stops (Darwin, 1969a; Haggard & Parkinson, 1971; Kirstein & Shankweiler, 1969; Spellacy & Blumstein, 1970; Studdert-Kennedy & Shankweiler, 1970) and extended to f i n a l stops (Darwin, 1969a; Studdert-Kennedy, 1970), fricatives (Darwin, 1971), and laterals and semivowels (Haggard, 1971). The findings with vowels have been far less consistent. They have shown no ear preference (Darwin, 1969a, 1971), a nonsignificant right-ear advantage (Kirstein & Shankweiler, 1969; Studdert-Kennedy & Shankweiler, 1970), a significant right-ear advantage (Chaney & Webster, 1966; Darwin, 1971; Haggard, 1971; Spellacy & Blumstein, 1970) and a significant l e f t -ear advantage (Spellacy & Blumstein, 1970). To be able to integrate these results, i t i s necessary to have some understanding of the relevant stimulus properties of consonants and vowels pertaining to speech perception, as well as a more comprehensive description of the conditions, paradigms, and procedure used in the studies themselves. For the former the interested reader i s referred to Liberman, Cooper, Shankweiler & Studdert-Kennedy (1967) and Lindblom & Studdert-Kennedy (1967) although a brief discussion of the relevant aspects of speech w i l l be included here. Analytically, the sounds of speech form a subset of the sounds of the environment since they are subject to phonetic constraints deriving from the anatomy and physiology of the vocal tract and to phonological and allophonic constraints imposed by particular languages. The phonetic con-straints are of two main types, both of which lead to a complex relationship 19 between the perceived phoneme and the acoustic signal. In one case a complex relation arises because the articulatory specifications for some phonemes are incomplete, the articulators which are not specified can then assume a wide variety of positions with a corresponding wide variety of acoustic sequelae. In the second case the complex relation arises from the variation i n size and shape of the vocal tract producing the sound. Phonemes which show extensive restructuring of their acoustic cues as a function of context are referred to as "encoded." Encoding is a property of the stimulus, but i t is possible to dis-tinguish i t from other types of stimulus properties. Encoding is not defined by any single.acoustical correlate, but because the need for the encoding principle appears with the need to move an articulator from one position to another, the amount of encoding tends to be high where there is articulatory and hence acoustical change. These acoustical changes are rapid variations in the frequency spectrum known as "formant transitions." A vowel is characterized by a relatively stable (approximately 100 msec.) set of values of the resonant frequencies of the vocal tract, the formants. When a consonant is articulated, these frequencies change rapidly (30 to 80 msec), but the values to which they go are relative to the neighboring vowel, and to the position i t i s i n with respect to that vowel. Thus the consonant is said to be encoded. This variance is different for each cue of manner, voicing and place. The perception of steady-state and isolated vowels is quite different from the perception of most consonants, depending primarily on the frequency position of the formants. There i s , for these vowels, no restructuring of the kind found to be so common among the consonant cues and, accordingly, no problem of invariance between acoustic signal and perception. Therefore, 20 steady-state vowels are said to be unencoded. However, vowels are rarely steady state in normal speech. Most commonly these phonemes are articu-lated between consonants and at rather rapid rates. Under these conditions vowels do show substantial restructuring - that i s , the acoustic signal at no point corresponds to the vowel alone, but rather shows at any instant, the merged influences of the preceding or following consonant. Thus the speech signal typically does not contain segments corresponding to the dis-crete and commutable phonemes. The formant transition i s , at every instant, providing information about the two phonemes, the consonant and the vowel - that i s , the phonemes are being transmitted in parallel. The dichotic listening paradigm seems especially suited to teasing;; out the conditions necessary for a right-ear advantage. Liberman et a l . (1967) hypothesized that only those aspects of speech which show appreciable con-textual variation (i.e. the encoded aspects) would give a right-ear advan-tage. This category would indicate those sounds containing formant transi-tions. This hypothesis is certainly consistent with Shankweiler & Studdert-Kennedy's (1967) discovery of a right-ear preference for synthetic isolated steady-state vowels. This right-ear advantage for synthetic i n i t i a l stops has been replicated by Haggard & Parkinson (1971) who used the same pro-cedure as Shankweiler & Studdert-Kennedy (1967) for the stops /b, p, d, t/, and also by Kirstein & Shankweiler (1969) who used an attention paradigm where the subject is instructed to preattend and report from only one ear in a given block of the experiment. Using a one-pair-per-trial free r e c a l l procedure, Darwin (1969a) replicated the right-ear advantage finding with synthetic i n i t i a l stops and extended the finding to f i n a l stops in vowel-consonant (VC) syllables, using a special method of scoring to 21 control for order of report. Studdert-Kennedy & Shankweiler (1970) repeated their i n i t i a l experiment using real speech and again found right-ear superiorities for both i n i t i a l and f i n a l stops incorporated into CVC syllables. Spellacy & Blumstein (1970) used a novel recognition paradigm in which, four seconds after each dichotic stimulus pair, the subject was presented with a diotic recognition f o i l which he was to identify as being the same as or different from the members of the preceding dichotic pair. Again, a right-ear preference was obtained for i n i t i a l stops i n CVC syllables. The results of these studies are not inconsistent with the notion that contextual variation i s an important factor in the production of a right-ear preference for verbal material. In a more direct test as to the necessity of ccontextual variation in producing a right-ear advantage, Darwin (1971) used fricatives as opposed to stop consonants. There are two main cues which contribute to the per-ception of fricatives. The f i r s t , and perceptually the most significant, is the spectral peak of the f r i c t i o n i t s e l f . This peak shows relatively l i t t l e variation with vowel context. A secondary cue is the formant transition to adjacent vowels. These transitions show much more contextual variation with vowel context than do the spectral peaks, as they depend on the shape of the whole vocal tract. Darwin used the six fricatives / f , s, J , v, z, 3/ in the syllabic frame /- ep/, employing the one-pair-per-tria l free r e c a l l procedure. The study included the four stimulus con-ditions of f r i c t i o n alone, f r i c t i o n plus steady-state vowel, f r i c t i o n plus appropriate formant transitions, and the appropriate formant transitions plus vowel minus the f i r c t i o n . Although fricatives synthesized from f r i c t i o n alone, without the formant transitions, were clearly identifiable, only the two conditions which included the formant transitions showed a 22 right-ear advantage, thus supporting the hypothesis of Liberman et a l . As was mentioned above, this hypothesis also seems to provide an account of the fact that Shankweiler & Studdert-Kennedy (1967) found no significant ear advantage with synthetic isolated steady-state vowels. This finding however, would be consistent with several other plausible hypotheses, including variations of the contextual variation hypothesis of Liberman et a l . , as well as other completely independent hypotheses. One possibility i s that right-ear superiority depends primarily upon prefer-ential processing in one hemisphere of certain purely acoustical properties that are present i n stop consonants but lacking in vowels. A second pos-s i b i l i t y i s that the right-ear advantage depends c r i t i c a l l y upon the pres-ence of an encoded acoustical feature i n the stimulus that i s present in stop consonants and lacking in steady-state vowels. A third possibility is that a right-ear preference i s dependent upon the actual "perceptual load" of decoding, this load being greater in stop consonants than in vowels. A further alternative i s that right-ear superiority depends only upon the phoneme class distinguishing the response items and not upon stimulus properties, vowels being among the classes not showing a right-ear advantage. Halwes (1969) on the other hand entertained the possibility that where vowels are used as stimuli, a right-ear advantage i s perhaps masked by a ceiling effect.. It was pointed out that in Shankweiler & Studdert-Kennedy's (1967) experiment, vowel identification proved to be a much easier task than i s consonant identification. Investigations by Kirstein & Shankweiler (1969) using the attention paradigm described earlier, and by Darwin (1969a) using the one-pair-per-t r i a l free recall procedure, also failed to obtain significant ear d i f -ferences for synthetic steady-state vowels, giving general support to the 23 above hypotheses. A previous experiment by Chaney & Webster (1966) had shown a significant right-ear advantage.for steady-state vowels, but the lack of adequate controls for channel differences, and the complex pro-cedure in which the subjects were instructed to identify the sound as regards to sex of author, inflection, voice, and ear, render the results d i f f i c u l t to interpret in reference to the various hypotheses. One of the main purposes of a second study by Studdert-Kennedy & Shankweiler (1970) was to determine whether natural vowels embedded in a consonantal frame would show a greater right-ear advantage than the synthetic, isolated, steady-state vowels of the previous study, thus supporting a contextual variation hypothesis or a modification of i t . Although a significant right-ear superiority was not obtained, some ten-dency toward a right-ear advantage for the vowels was evident. The nega-tive findings were explained by a consideration of the fact that these vowels were approximately 300 to 500 msec, long whereas i n natural speech they average only 100 msec. According to Liberman et a l . (1967), the acoustic cues for vowels in slow articulation tend to be invariant. Thus the results of this experiment again appear inconclusive. One attempt to c l a r i f y the issue was made by Haggard (1971, Exp.I) using semivowels and laterals (w, r, 1, j ) . Of a l l classes of speech sounds, the semivowels and laterals are those which vowels have most in common. These sounds are made with a relatively open vocal tract, and have a short steady state which is acoustically similar to a vowel. But they also have formant transitions leading into the neighboring vowel. In the latter respect, they are analagous to stop consonants, but the transitions are slower and more intense, being made with a more open 24 vocal tract. The words "what, rot, lot, yacht" were used in a one-pair-per - t r i a l free r e c a l l procedure. Right-ear superiorities very similar to those found by Shankweiler & Studdert-Kennedy (1967) for stop consonants were obtained. These results suggest very strongly that properties shared between vowels and semivowels such as overall duration and openess of vocal tract are not relevant to right-ear superiority and that some properties shared by semivowels and stop consonants such as the common presence of formant transitions and encoding are relevant. The finding that the most plausible acoustical factor is irrelevant sheds doubt on the hypothesis that right-ear superiority i n dichotic listening tasks depends primarily upon acoustical processing. Normally, the source of encoding for vowels i s the consonantal con-text. This encoding is usually reflected in the formant transitions, thus making i t d i f f i c u l t to separate these two factors i n regard to their relative importance i n the production of a right-ear effect. Haggard (1971, Exp. II) used a novel technique to divorce encoding as an explana-tory principle from the stimulus properties of formant transitions. This was done by using speaker context as a special source of encoding not involving formant transitons. Synthetic steady-state vowels were embed-ded in CVC syllables, creating the words "a kid, a ked, a cud, a cod, a could." These stimuli were recorded in two different fundamental fre-quencies similar to a high voice and a low voice. A right-ear preference was found, demonstrating that the addition of an element of encoding to steady-state vowels i s a sufficient condition for this effect. Thus the idea that right-ear superiority depends upon a property of a phoneme class was ruled out. Also, an analysis of sequential effects showed that .25 the amount of decoding actually required on a given t r i a l was not relevant. The results would seem.to indicate that a sufficient condition for the emergence of a right-ear advantage is the presence of some normally encoded acoustical attribute. The above results were essentially replicated by Darwin (1971) with synthetic steady-state vowels i n the words "a nit, a net, a gnat, a knot, a nut." Each stimulus was produced by two different sized vocal tracts, providing a source of contextual variation. A control condition was included wherein only sounds from the smaller vocal tract were presented. The results were the same as those of Haggard (1971) with the additional finding that the right-ear superiority with vowels depended upon the nature of the discrimination within the framework of the whole experiment rather than within an individual t r i a l . Although no ear advantage was demonstrated by the control group, the same stimuli presented to the experimental group demonstrated right-ear superiority. Here, then, i t is not merely the extraction of the encoded acoustical cue which i s important but also i t s phonetic relevance. The implication i s that the presence of some normally encoded acous-t i c a l attribute does or does not produce a right-ear advantage, depending upon the nature of the task, and the relevance of that attribute to the task. Support for this hypothesis comes from experiments dissociating the particular stimulus feature encoded from the task involved. One such experiment by Darwin (1969b) involved inflecting the fundamental frequency on a high quality synthesized syllable. The study demonstrated that these simple pitch sweeps give a left-ear advantage even though they are carried on a word.. These pitch sweeps did not cue a phonemic distinction. That 26 the stimulus parameter of fundamental frequency is not responsible for this finding was demonstrated in a complementary experiment by Haggard & Parkinson (1971, Exp. I ) . Fundamental frequency was manipulated, being used to cue the linguistic distinction of voicing for stop consonants. A right-ear advantage rather than a left-ear advantage was obtained. Further support that the presence of an encoded stimulus attribute does not necessarily result in a right-ear advantage comes from a second experiment by Haggard & Parkinson (1971, Exp. II). A series of six sen-tences, each varying i n emotional tone was used as stimuli. An attention paradigm was used with babble as the competing stimulus. The task of the subject was to indicate by ticking a response sheet both the sentence and the emotion heard. Recognition of the emotions showed a left-ear advantage, while there was no ear difference for recognition of sentences. Upon analysis of the data, i t was found that in cases where sentences were incorrectly reported, an above chance score was.still obtained on emotions, indicating a possible dissociation of the tasks and implicating attention as a possible source for this dissociation. Thus while the experiment supports the idea that the perceptual task is of prime importance in det-. ermining ear advantages, i t also indicates that the subject's attentional strategy may have an effect upon ear asymmetries. This attentional factor was directly attacked by Spellacy & Blumstein (1970) who used their recognition paradigm described previously. The stimuli were dichotic pairs of CVC nonsense syllables that differed in medial vowel or i n i t i a l consonant. The subject's attention was manipulated by creating an expectation of hearing language or nonlanguage sounds. This manipulation was achieved for the language-expectation group by 27 presenting the test stimuli in a random series along with CVC syllables constituting real English words. A nonlanguage set was established i n the second group by presenting the test stimuli in series along with melodies and sound effects. The vowels showed a right-ear advantage for the language expectation group and a left-ear advantage for the non-language expectation group, consonants showing a right-ear advantage for both groups. As discussed above, the encoded cues are the only stimulus properties by which stop consonants may be identified. It would seem consistent with the results of Spellacy & Blumstein (1970) as well as with the previous studies discussed, that any task or attentional strategy that makes this consonant identification relevant w i l l result in a right-ear advantage. On the other hand, i f a stimulus possesses attributes by which i t may be identified that are both encoded and unencoded, the direction of the ear advantage w i l l be determined by whether or not the subjects have been given a "language set",/ Thus vowels, possessing both formant transitions and steady-state frequencies, show a right-ear advan-tage only i f the encoded attributes are made relevant to the identification task. The assumption, that a right-ear advantage depends upon both the presence of some encoded acoustical cue and a task or attentional strategy that makes that encoded property relevant, appears to reconcile the con-f l i c t i n g results of previous experiments. For example, several investi-gators failed to find a right-ear effect for vowels because they employed stimuli lacking i n encoded attributes, i.e., steady-state vowels (Darwin, 1969a; Kirstein & Shankweiler, 1969; Shankweiler & Studdert-. 28 Kennedy, 1967). And although some investigators used stimuli which might have otherwise shown a right-ear advantage, they either u t i l i z e d a task clearly making the encoded attributes irrelevant (Darwin, 1971; Spellacy & Blumstein, 1970) or employed a procedure that may very well have resulted in uncertainty on the part of the subjects as to the linguistic nature of the task involved (Studdert-Kennedy & Shankweilerj 1970). Where both the appropriate stimuli and a linguistic task are provided, vowels appear to show a distinct right-ear advantage (Darwin, 1971; Haggard, 1971; Spellacy & Blumstein, 1970). Thus investigations into the right-ear advantage phenomenon have pro-duced evidence to suggest the presence of a linguistic mechanism in the l e f t hemisphere that i s involved in some way in the perception of encoded stimulus attributes characteristic of speech. The findings have also sug-gested, however, that the mere presence of these attributes in the stimulus input i s not sufficient to produce a right-ear effect. An additional requirement is a "language set" that makes the encoded stimulus attribute relevant to the identification of the stimulus input. The studies discussed above have almost exclusively dealt with the nature and the conditions necessary to produce a right-ear advantage. Although much understanding concerning the nature of cerebral dominance with regard to language has been gained, almost nothing i s known concerning the nature of cerebral dominance with nonverbal stimuli. As was mentioned above, Kimura (1967) proposed that greater left-ear scores for certain nonverbal functions reflected right-hemispheric specialization for nonverbal functions' Perhaps the most plausible hypothesis would be that there i s some kind of perceptual mechanism i n the right hemisphere that i s specialized for part i -cular stimulus properties characteristic of certain nonverbal sounds, 29 similar in principle to the linguistic mechanism for encoded attributes typical of speech, found i n the l e f t hemisphere. Too few experiments exploring the left-ear advantage phenomenon have been conducted to enable an evaluation of such a hypothesis. An alternative hypothesis has been proposed by Spellacy & Blumstein (1970). It was noted that although a right-ear advantage for vowels was obtained in the language-expectation group, and a left-ear advantage in the nonlanguage group, the actual recognition scores for the l e f t ear remained the same in the two conditions. Both ear differences resulted almost solely from fluctuation in right-ear performance which i s presum-ably due to l e f t cerebral hemispheric function. The suggestion was offered that when the formant transitions of the vowels were attended to, the result was enhanced perception. However, when the formant transitions were not attended to, perception was decreased. That i s , the result of the l i n g u i s t i c mechanism i s enhanced perception when these encoded a t t r i -butes are attended to, and a decrement in perception when these encoded attributes are.not attended to. They speculate that "the l e f t ear super-i o r i t y which was been shown for other nonverbal sounds i s due not to increased perception associated with specialized right hemisphere auditory processing, but rather to a decrease of right ear perception due to the inhibiting effects of the right ear signal passing through a speech sound analysis to which _S does not attend." (p. 438) The primary purpose of the present investigation was to explore this possibility. One approach would be to devise a task in which, by mani-pulating the attentional strategy of the subjects, both right-and l e f t -ear superiorities could be obtained for nonverbal sounds. A simple inves-tigation would demonstrate whether only right-ear recognition was being 30 affected or both. As nonverbal sounds do not possess the necessary encoded attributes required to produce a right-ear effect, such a procedure can not be ut i l i z e d . An alternative approach is to use a developmental paradigm. A dev-elopment of ear asymmetries, reflecting developing cerebral lateralization for the perception of sounds, might be expected with increasing age. Unfortunately, l i t t l e i s known concerning the ages at which ear asymmetries f i r s t appear. Two studies by Kimura (1963, 1967) and one by Kimura & Knox (1970) have found that both boys and g i r l s already show a right-ear effect for verbal material at the age of five, although i n one study (Kimura, 1967) the right-ear advantage for five-year old boys did not reach signi-ficance. The study by Kimura & Knox (1970) also showed a left-ear advan-tage for environmental sounds present at this age. There are several d i f -f i c u l t i e s with these studies however. Although the procedure used was a free r e c a l l paradigm, or a simple modification of i t involving the label-ling of environmental sounds, order of report factors were not controlled for. A study by Inglis and Sykes (1967) indicates that in free r e c a l l procedures order of report factors may have a greater influence upon the dichotic listening performance of children than upon that of adults. That order of report factors may have contaminated the results of Kimura is further suggested by the fact that the.degree of ear asymmetry in the experiments decreased rather than increased with age as would be expected i f ear asymmetry were reflecting the development of cerebral lateralization for language. A further d i f f i c u l t y i s that these results seem to be at variance with neurological studies which used brain-damaged children as subjects. Such studies have demonstrated that through the elementary 31 school years, the likelihood of right-hemispheric damage producing long-lasting aphasic disorders gradually decreases, while the likelihood of left-hemispheric damage producing long-lasting aphasic disorders grad-ually increases, the adult pattern of lateral dominance being clearly established by age 12 or 13 (Lenneberg, 1966; Zangwill, 1960). A study by Bryden (1970) investigating dichotic listening performance with elementary school children i n grades 2, 4, and 6 produced results more i n keeping with the studies of brain-damaged children. A dichotic digits free recall task was used but the results were reported i n terms of the relative incidence of, rather than magnitude of right-ear super-i o r i t y . Although the difference in scoring procedure makes comparisons d i f f i c u l t , there i s some indication that the frequency of right-ear superiority increases with age in right-handed subjects, and that the adult pattern emerges earlier i n g i r l s than in boys. These conclusions are highly tentative, however, and i t was hoped that the present investigation would provide further information as to the development of ear asymmetries in children. The primary purpose of the present study then, was to investigate the nature of ear-asymmetry development, especially in regard to stimulus material showing left-ear superiority. A secondary purpose was to provide confirmation for the premise that left-ear superiority for vowel-varied stimuli i s possible under conditions making the encoded attributes a Com-ponent of a task which e l i c i t s a nonlanguage set. And, as mentioned above, an additional aim was to provide c l a r i f i c a t i o n as to the influence of age and sex factors on ear asymmetry. With the above purposes in mind, the present experiment was planned as to procedure, subjects and stimulus material used. To avoid the 32 exaggerating effect of order of rec a l l factors on ear asymmetry, a simple recognition paradigm was decided upon. And on the basis of the studies of brain-damaged children as well as Bryden's (1970) dichotic listening study, elementary school children were chosen as subjects. In an attempt to pro-duce a nonlanguage attentional set, the consonant-varied and vowel-varied stimuli were presented along with music and sound effects. Regarding the direction of ear asymmetry, a right-ear advantage was predicted for the consonant-varied stimuli, and a left-ear advantage was predicted for the vowel-varied stimuli, music, and sound effects. Although presumably a nonlanguage set would be created, a right-ear advantage was predicted for the consonant-varied stimuli as only the encoded attributes of a stop consonant are relevant to i t s identification. In contrast, a left-ear advantage for vowels was predicted as the nonlanguage attentional set would presumably change the relevant identification cues from the encoded attributes to the steady-state stimulus properties which vowels have i n common with other nonverbal sounds that have produced left-ear effects. The basis for the predicted left-ear superiority of sound effects, music and vowels in a nonlanguage context, was an assumed decrease in right-ear perception due to the right-ear signal passing through a speech sound analysis to which the subjects were not attending. This hypothesis i s in contrast to Kimura's argument that left-ear superiority is a result of increased perception associated with specialized right hemisphere auditory processing. Thus, in the present experiment, i t was hypothesized that any increase in ear asymmetry with left-ear advantage stimulus material would be due to a decrease in right-ear recognition, left-ear recognition 33 Remaining unchanged. As the r i g h t ear e f f e c t i s presumably due to s p e c i a l i z e d processing of encoded a t t r i b u t e s i n the l e f t hemisphere, and as Zangwill (1960) and Lenneberg (1966) have given evidence to suggest that right-hemisphere involvement i n language functions decreases as l e f t -hemisphere involvement increases, i t was hypothesized that any increase i n ear-asymmetry with ri g h t - e a r advantage stimulus material would be the r e s u l t of a simultaneous increase i n right-ear recognition and decrease i n l e f t - e a r recognition. On the basis of Bryden's (1970) developmental study using elementary school children, and the studies using brain-damaged c h i l d r e n , as subjects, i t was predicted that ear asymmetry would increase with age. Regarding sex f a c t o r s i n ear asymmetry, there i s some evidence to suggest that females may show ear-asymmetry with verbal material sooner than males (Bryden, 1970; Kimura, 1967). The di f f e r e n c e s reported were not s i g n i f i c a n t however, and therefore no predictions were made regarding the e f f e c t s of sex i n the present study. In a d d i t i o n to the hypotheses d i r e c t l y r e l a t e d to the stated aims of the present study, two a d d i t i o n a l p redictions were made on the basis of findings i n previous d i c h o t i c l i s t e n i n g experiments. One p r e d i c t i o n was that o v e r a l l recognition performance would increase with age. This r e l a t i o n s h i p between age and d i c h o t i c l i s t e n i n g performance has been reported by Bryden (1970), I n g l i s & Sykes (1967), Kimura (1963, 1967), and Knox & Kimura (1970). I t was also expected that subjects would recognize more vowel-varied s t i m u l i than consonant-varied s t i m u l i . This p r e d i c t i o n was based on observations by Shankweiler & Studdert-Kennedy (1967, 1970) K i r s t e i n & Shankweiler (1969), and Spellacy & Blumstein (1970). 34 In summary then, the present investigation was concerned with the following hypotheses: 1. A right-ear advantage for consonant-varied stimuli. 2. A left-ear advantage.for vowel-varied stimuli, music and sound effects. 3. An increase i n ear-asymmetry with age, due to: a. an increase in right-ear recognition and a simultaneous decrease in left-ear recognition with right-ear advantage stimulus material. b. a decrease i n right-ear recognition with left-ear advantage stimulus material. 4. An increase i n dichotic listening performance with age. 5. A higher level of recognition performance with vowel-varied stimuli than with consonant-varied stimuli. 35 CHAPTER III Method Subjects The Ss used were 208 elementary school children in Grades 2, 4 and 6. Four schools were used. A l l Ss, as judged by their teachers, (a) were right-handed,.(b) were i n their grade by academic pass, (c) had not repeated a grade, (d) spoke English as their native tongue, and (e) had no known auditory impairment. A l l Ss were also required, in keeping with the policy of the school d i s t r i c t , to have a consent form signed by a parent or guardian. Of the 208 Ss, 19 were excluded because of apparatus d i f f i c u l t y , and 9 because of failure to follow instructions. Procedure The apparatus used to present the dichotic stimuli consisted of a Sony TC 630 stereophonic tape-recorder, a six-jack stereophonic listening station, and six sets of stereophonic earphones (Sharpe Pro HA 660). Ss were tested in groups, usually of five, each session taking approximately 30 minutes. Individual desks were placed facing away from the tape-recorder and in a semicircle, to minimize distraction. The tape was pre-sented to the Ss at a low comfortable listening level which was the same for Ss i n a l l conditions. The stimuli were monitored by E_ through a sixth set of earphones connected to the listening station. Tape-recorded instructions outlined the nature of the task and pro-vided two examples as well as an opportunity for questions. If there was any doubt as to whether an _ understood, instructions were repeated by E in a varied form with additional examples, unt i l there was relative cer-tainty as to S_s understanding. This procedure was necessitated most fre-quently when children in Grade 2 were tested. 36 The auditory stimulation procedure consisted of 80 dichotic stimulus pairs, one pair at a time. Each pair was presented twice to make a total of 160 presentations. Each dichotic stimulation was preceded by an indication of the t r i a l number and followed after four seconds with a diotic recognition f o i l . There was an interval of five seconds between the recognition stimulus and the onset of the next dichotic presentation. Ss were instructed to respond by marking an "X" under "yes" on a prepared answer sheet i f the recognition stimulus was the same as one of the mem-bers of the preceding dichotic pair, or under "no" i f the recognition stimulus was different from either of the members of the preceding dichotic pair. Thus a correct response for the right ear would be recorded i f an j3 marked "yes" when the item member of the dichotic pair presented to the right ear was the same as the following recognition stimulus presented diot i c a l l y . And a correct response for the l e f t ear would be recorded i f an S_ marked "yes" when the item member of the dichotic pair presented to the l e f t ear was the same as the diotically presented recognition stimulus. The number of correctly recorded "yes" responses constituted an S/s score for a particular ear on a specific type of stimulus material. To control for any channel differences arising from inequalities i n the apparatus, half of the Ss in each c e l l had the earphone order reversed relative to the other half. Stimulus Materials The stimulus materials were adapted from a study by Spellacy & Blumstein (1970). Four types of test stimuli were used. Twenty pairs of dichotic stimuli were CVC nonsense syllables that differed in i n i t i a l con-sonant. The i n i t i a l consonants varied were the six stop-consonants 37 /p, t, k, b, c, g/. Another twenty pairs of dichotic.stimuli were CVC nonsense syllables differing in the medial vowel. The vowels varied were / i , e, a, 9, o, u/. In addition to these CVC nonsense syllables, twenty pairs of dichotic stimuli were exerpts of sung melodies, made up of melodic repetitions of the CVC syllable /da/. The remaining twenty dichotic pairs were human imitations of animal and machine sounds. A l l sounds were created by the same voice and no deviation from simultaneity of onset was perceptible to E. In recording the stimulus material, the randomized series of 80 dichotic pairs was repeated with channels reversed so that a member of a dichotic pair that would be presented to the l e f t ear in the f i r s t half of the experiment, would be presented to the right ear i n the second half. Also, recognition f o i l s were assigned randomly to dichotic pairs but with the restriction that i f a pair was followed by a neutral f o i l in the f i r s t half (i.e. requiring a "no" response) i t would be followed by a positive f o i l in the second half and vise versa. A further restriction was that 25 percent of the recognition stimuli were identical with right-ear stimuli, and 25 percent were identical with the left-ear stimuli. The remaining 50 percent were recognition stimuli not used in the dichotic presentation. Design A 3x2x4x2 between-within analysis of variance design was ut i l i z e d , the between factors being grade (G) and sex (S), and v.t'hei within factors being type of stimulus material presented (M), and ear of presentation (E). 38 There were three levels of G and four levels of M. Each c e l l contained 30 Ss. An increase in dichotic listening performance with age was expected to be revealed in a significant G effect. An analysis of the linear com-ponent of this effect was planned to determine the shape of the function. A significant main effect for M was also predicted, and a separate signi-ficance test was planned to compare consonant and vowel recognition. A significant MxE interaction was expected, representing a right-ear advan-tage for consonant-varied stimuli and a left-eear advantage for vowel-varied stimuli, music and sound effects. Separate orthogonal contrasts between ears were planned for the predicted right-ear advantage material and for the predicted left-ear advantage material. The predicted increase of ear-asymmetry with age would be evidenced by a significant MxGxE interaction, and sex differences in this develop-ment would be represented in the MxGxExS interaction. To test the hypo-thesis that, where increases in ear asymmetry were evident with right-ear advantage stimulus material, this increase would be the result of a simultaneous increase with age in right ear recognition and decrease with age in left-ear recognition, two orthogonal contrasts were planned, one for each ear. The same procedure was planned for left-ear advantage material to test the hypothesis that left-ear superiority i s due to a decrease in right-ear recognition with age. As developmental effects are necessarily confounded with grade effects in a developmental paradigm, the grade condition totals for each c e l l were adjusted to exclude, or at least minimize the exaggerating effects of grade. This analysis was carried out by calculating the average increase across grades for each sex 39 w i t h i n each s t i m u l u s m a t e r i a l , and s u b t r a c t i n g t h e a v e r a g e i n c r e a s e f r o m t h e a p p r o p r i a t e c o n d i t i o n t o t a l s o f each e a r . (A d e s c r i p t i o n and example o f t h e p r o c e d u r e used i s i n c l u d e d i n t h e A p p e n d i x . ) W i t h t h i s a d j u s t m e n t t o e x c l u d e Grade e f f e c t s , i t was p r e d i c t e d t h a t where i n c r e a s e s i n e a r asymmetry were e v i d e n t , t h e a d j u s t e d d a t a w o u l d show an i n c r e a s e i n r i g h t -e a r p e r f o r m a n c e and a s i m u l t a n e o u s d e c r e a s e i n l e f t - e a r p e r f o r m a n c e f o r : .. r i g h t - e a r a d v a n t a g e s t i m u l u s m a t e r i a l , and a d e c r e a s e i n r i g h t ^ e a r p e r -formance w i t h no change i n l e f t - e a r p e r f o r m a n c e f o r l e f t - e a r advantage s t i m u l u s m a t e r i a l . 40 CHAPTER IV Results A summary of the analysis of variance of recognition scores for the four factors of grade (G), sex (S), ear of presentation (E), and stimulus material presented (M), is presented in Table I. The main effect for G was significant (F=10.73, df=2,174; _<?.001). A highly significant por-tion of the variance attributable to G was due to the linear component (Zlinear =21 • 18, df=l, 174; p_<;.001) of the effect, reflecting increased performance as a function of grade level. The mean recognition score for grade 2 was 7.65, for grade 4, was 8.02, and for grade 6 was 8.52. The maximum recognition score possible was 10. A significant E main effect (F=38.61, d_=l,174; p_<.001) reflects the fact that in general, more stimuli were recognized when presented to the l e f t ear than when presented to the right ear. The mean recognition score for the l e f t ear was 8.27, and for the right ear was 7.86. The main effect for M was also significant (F=20.46, df=3,522; p_<C.001). This effect i s illustrated in Fig 1. Consonant identification proved to be the most d i f f i c u l t task with a mean recognition score of 7.51. Music and sound effect stimuli were apparently much easier to identify, with mean recognition scores of 8.34 and 8.39 respectively. Vowels represented a task of intermediate d i f f i c u l t y with a mean recognition score of 8.00. The difference between recognition performance with consonant-varied stimuli and vowel-varied stimuli was found to be significant (F=l5.05, df=1,522) at the .001 l e v e l l of probability. The main effect for S was not significant, nor were any interactions with S. A significant MxG interaction (F=2.11, df=6,522; p_<.05) reflects the fact that the rate of increase in stimulus recognition as a function \ 41 Table I Summary of Analysis of Variance of Recognition Scores Source SS df MS F P G (grade) 188.03 2 94.02 10.73 <.001 S (sex) 13.03 1 13.03 1.49 GxS - 42.37 2 21.18 2.42 Error (b) 1523.44 174 8.76 M (material) 174.93 3 58.31 20.46 <.001 MxG 35.97 6 6.00 2.11 <.05 MxS 0.77 3 0.26 0.09 MxGxS 9.32 6 1.55 0.54 Error (w^) 1486.38 522 2.85 E (ear) 54.05 1 54.05 38.61 <.001 ExG 4.69 2 2.34 1.67 ExS 1.06 1 1.06 0.76 ExGxS 6.10 2 3.05 2.18 Error (v^) 244.22 174 1.40 MxE 189.79 3 63.26 50.61 <.001 MxExG 17.20 6 2.87 2.30 <.05 MxExS 6.32 3 2.11 1.69 MxExGxS 3.54 6 0.59 0.47 Error (w^) 654.53 522 1.25 Total 4655.74 1439 42 9 -<u M O O CO c o a o o Pi c a 7 -Consonants Vowels Music Sound Effects Fig. 1 . Recognition performance as a function of the stimulus material presented 43 of age varies with the stimulus material presented. This interaction i s illustrated in Fig. 2. Between grades 2 and 4, the rates of increase in stimulus recognition do not vary greatly with the type of material presented. Between grades 4 and 6, however, a large increase in stimulus recognition is evident for sound effects, a smaller increase for consonants and vowels, and the least increase i s evident for music. The mean difference between grades 4 and 6 for sound effects i s 1.05 while the mean difference between the same two grades for music i s only.0.17. The rate of increase i n stimulus recognition remained the same for consonants and music across grades, while vowel and sound effect recognition increased more between grades 4 and 6 than between grades 2 and 4. The MxE interaction was highly significant (F=50.61, df=3,522; p_.<r.001). This effect is illustrated in Fig. 3. Consonant-varied stimuli were more frequently identified correctly when presented to the right ear than to the l e f t ear (F=46.72, df=1,522; p_<.001). The percentage ear preference for consonants was 5.36. This measure was calculated by dividing the d i f -ference between recognition scores for right and l e f t ears by the sum of the recognition scores for both ears, and multiplying by one hundred. Recognition of vowel-varied stimuli, music, and sound effects showed l e f t -ear preferences of 3.32 percent, 6.90 percent and 4.04 percent respectively. The difference between l e f t versus right ear recognition for these stimuli combined was highly significant (F=399.50, df=l,522, £^.001). The one significant t r i p l e interaction, MxExG CE=2-30, df=6.522; p_<.05) reflects the development of ear differences with age, the direction of the ear differences being determined by. the stimulus material being presented. This interaction is illustrated in Fig. 4. Consonants, vowels and sound 44 Fig. 2. Increase in recognition of stimulus material as a function of grade 10 Fig. 3. Ear Preference as a function of stimulus material 10 u 89 w C o •H C 60 o 4J 8 c ro (U S Right ear Left ear Grade 2 4 6 Consonants 4 Vowels 4 Music 2 4 6 Sound Effects 4> Fig. 4. Recognition scores of stimuli presented to the right and l e f t ears as a function of grade level of stimulus material • 47 effects rshow increases i n ear differences between grades 2 and 4, while music shows no evidence of a developmental trend either between grades 2 and 4 or between grades 4 and 6. Consonants show a decrease in ear asymmetry between grades 4 and 6, while recognition of vowels and sound effects increases at about the same rate for either ear. No ear. prefer-ence i s present for vowels in grade 2, while small ear preferences are present for consonants and sound effects, and a relatively large ear preference i s already present for music. Ignoring the direction of the effect, the average ear preference across material i s 3.47 percent in grade 2, 5.97 percent in grade 4, and 6.19 percent in grade 6. Fig. 5 illustrates the development of ear asymmetries from grade 2 to grade 4 separately for.each sex. The increase in ear-asymmetry for consonants between.grades 2 and 4 illustrated in Fig. 4 is apparently due solely to males, females already showing a 6.12 percent right-ear pre-ference i n grade 2, increasing only slightly to 6.50 percent in grade 4. With sound effects, the trend i s reversed for males and females, males demonstrating a 6.14 percent left-ear preference in grade 2, while females show absolutely no ear preference and account for almost a l l of the increase in ear asymmetry from grade 2 to grade 4 shown in Fig. 4. Both sexes show an increase in ear asymmetry in vowel recognition, females showing a faster rate of development than males. Neither sex shows ear-asymmetry for vowels in grade 2. In the recognition of music stimuli^-,, both sexes show a substantial left-ear preference in grade 2 with no sign of further development. Thus, four developmental trends in ear asymmetry are evident, males showing a developmental effect for consonants and vowels, and females for Fig. 5. Sex differences in the development of ear preferences 49 vowels and sound effects. These trends are illustrated separately for right-ear advantage stimuli and left-ear advantage stimuli in Fig. 6. With grade effects included, a large increase i n right-ear recognition (F-22.43, df=l,522; p_<;.001) with negligible change i n left-ear recogni-tion i s observed for consonant-varied stimuli. With left-ear advantage stimuli, the reverse trend i s observed, recognition of these stimuli increasing significantly (F=19.95, df=l,522; p_<:.001) for the l e f t ear with only a slight change in the right ear. With grade effects excluded by the method outlined earlier, recognition of consonants presented to the right-ear increases while recognition of the same stimuli to the l e f t ear decreases. For the left-ear advantage stimuli, recognition of the stimuli presented to the right ear decreases while negligible change is evident in left-ear recognition. A summary of this analysis for grade effects included and for grade effects excluded i s presented in Tables II and III respectively. A summary of the analyses of ear asymmetries made by computing separate analyses of variance for each c e l l i s presented in Table IV. With con-sonants, significant ear differences were found for females in a l l grades and for males in grades 4 and 6 only. With vowels, significant ear pre-ferences were found for females in grades 4 and 6 and males in grade 6. Significant ear preferences were shown for both sexes in a l l grades with musical stimuli. With sound effects, although boys showed a significant ear preference across a l l grades, this preference was evident only in grade 6 with g i r l s . _1_ _l_ 2 Grade 4 Grade Effects Included 2 Grade 4 Grade Effects Excluded — i — — 1 2 Grade 4 Grade Effects Included 1 I 2 Grade 4 Grade Effects Excluded Right-ear advantage stimuli Left-ear advantage stimuli Fig. 6. Nature of the development of ear asymmetries with grade effects included and with grade effects excluded for both right-ear advantage stimuli and left-ear advantage stimuli 51 Table II Total Recognition Scores with Grade Effects Included Grade Sex Material Ear 2 4 F P Male consonants l e f t right 204 212 203 248 0.13 22.43 >.90 <.001 vowels l e f t right 219 217 237* 221** *19.95 ** 0.44 <:.ooi >.85 Females vowels l e f t right 228 232 265* 238** sound effects l e f t right 243 243 255* 240** 52 Table III Total Recognition Scores with Grade Effects Excluded Grade Sex Material Ear 2 Male consonants vowels l e f t 204.0 188.5 right 212.0 225.5 l e f t 219.0 213.7 right 217.0 197.7 Females vowels l e f t 228.0 247.0 right 232.0 220.0 sound effects l e f t 243.0 241.2 right 243.0 226.2 53 Table IV Summary of Separate Analyses of Ear Differences within Levels of Material, Grade, and Sex Material Grade Sex Left Ear Right Ear F p_ M 204 212 0.50 >.50 2 F 207 234 8.55 <.01 Consonants Vowels M 203 241 25.06 <.001 4 F 216 246 16.12 <.001 M 229 248 5.72 <.05 6 F 221 244 13.14 <.001 M 219 217 0.03 >.90 2 F 228 232 0.16 >.35 M 237 221 1.70 >.15 4 F 265 238 13.97 <.001 M 275 254 9.30 <.01 6 F 264 231 13.35 <.001 M 251 216 10.10 <.01 2 F 269 236 19.98 <.001 M 265 229 13.25 <.001 4 F 271 238 15.78 <.001 M 273 243 8.38 <.01 6 F 275 235 23.19 <001 M 243 215 9.97 <01 2 F 243 243 0 1.0 Sound — — — Effects M 258 223 29.58 <001 4 F 255 240 2.80 >.25 M 290 269 13.13 <01 6 F 82 5 4 44 01 Music • 54 CHAPTER FIVE Discussion Direction and Nature of the Ear Asymmetries Obtained The finding of a right-ear superiority with consonant-varied non-sense syllables i s consistent with the results of previous dichotic listening experiments using speech stimuli. The explanation that seems most consistent with this finding as well as with previous findings i s that there i s present in the l e f t hemisphere a linguistic mechanism for processing encoded acoustical cues. A right-ear rather than a left-ear advantage i s presumably obtained because of the greater efficiency of the contralateral auditory pathway over the ip s i l a t e r a l pathway. The present data show that this right-ear preference for consonant-varied stimuli originally develops from an increment in right-ear per-formance and simultaneous decrement in left-ear performance with age. These effects presumably reflect the developing lateralization of speech perception in the l e f t hemisphere. As such, the data are in agreement with the evidence obtained from brain-damaged children, these observations showing that through the elementary school years the likelihood of right-hemisphere damage producing longlasting aphasic disorders gradually dec-reases whereas the likelihood of left-hemisphere damage producing these disorders increases. The parallel development of a right-ear preference for verbal material and left-hemisphere involvement in the processing of language adds strong support to Kimura's hypothesis of a close relationship between the two phenomena. The finding of a left-ear superiority for vowel-varied stimuli i s relatively unique, having been found previously only by Spellacy & 55 Blumstein (1970). Although vowels, as do consonants, contain encoded acoustical cues, no right-ear advantage was hypothesized as the nonlanguage set given Ss presumably would result in these cues being made irrelevant to the identification of the stimuli. Thus, the lack of a right-ear preference with these vowel-varied stimuli supports the hypothesis that the mere presence of encoded cues i s not a sufficient condition to produce a right-ear effect. The finding thus further emphasizes the importance of task variables in influencing the direction of the ear difference obtained when both encoded and steady-state cues are present. The left-ear preferences found for music and sound effects are con-sistent with the data from the relevant investigations discussed above. As to the nature of the left-ear advantage with music, sound effects and vowel-varied stimuli, the data show that this advantage is due primarily to a decrement in right-ear performance, left-ear performance remaining unchanged. This finding presents a challenge to the common practice in neuropsychology of interpreting ear superiority in dichotic listening tasks as a reflection of a perceptual or processing dominance of the cerebral hemisphere contralateral to the ear showing superior performance. According to this interpretation, greater left-ear scores for music and sound effects reflect right hemisphere specialization in the processing of these stimuli. The left-ear advantage with vowel-varied stimuli would probably be explained by Kimura's perceptual bi l a t e r a l dominance hypo-thesis by assuming superior perception in the right cerebral hemisphere of the steady-state "music-like" properties of vowels. The present data, however, do not support such interpretations, the left-ear advantage being the result of a decrement i n right-ear performance rather than an incre-ment in left-ear performance. 56 An alternative to the bilateral dominance explanation would be to interpret the data solely i n terms of a lef ^ hemisphere specialization for the processing of encoded acoustical cues. It is proposed that such a li n g u i s t i c mechanism would result in enhanced perception of stimuli when the encoded acoustical cues are attended to, and a decrement in the perception of stimuli when these cues are not attended to. Thus the right-ear decrement would be the result of a right-ear signal passing through a linguistic mechanism to which S_ is not attending. As there is no decre-ment in perception of the l e f t ear signal, an apparent left-ear advantage would be obtained. Such an interpretation i s more consistent with the present findings. The unilateral dominance model would also seem to be more compatible with our knowledge of the principles governing evolution. Study of the evolution of the vocal tract in relation to the physiological requirements for producing the sounds of speech suggests that man has evolved special structures for speech-production and has not simply appropriated existing structures designed for eating and breathing (Lieberman, 1968). One would reasonably suppose that man has also evolved matching mechanisms for speech perception. Evidence has already been discussed above which shows that speech perception entails peculiar processes, distinct from those of nonspeech auditory perception. That a specialized perceptual mechanism for speech would therefore be necessary seems f a i r l y obvious. No pressing evolutionary reason, however, is readily apparent for a complementary specialized perceptual mechanism for nonspeech sounds. As the evidence is f a i r l y well documented in regard to a specialized speech mechanism in the left-hemisphere, i t would seem more appropriate to consider the effects or •57 resulting cost of such specialization on the auditory processing of nonverbal stimuli rather than to postulate a separate dominance for these stimuli. Certainly such an approach would be more parsimonious given our present state of knowledge. Study of evolutionary history demonstrates that when specialization takes place, i t does so at the expense of more expendable functions. It would seem reasonable to expect then, that auditory specialization for speech perception in man would probably interfere with the more expendable perception of nonspeech sounds. As this specialization for speech perception has taken place in the l e f t hemisphere, i t would be reasonable to expect that the interference with nonspeech perception would be evident i n the same hemisphere. As no specialization for speech per-ception and therefore no interference with nonspeech perception i s evident in the right hemisphere, an apparent left-ear advantage would result. The postulation of a specialized mechanism for speech perception would thus appear to be a l l that i s necessary to account for both right-and left-ear preferences. Such an interpretation i s not only more parsimonious, but i s more consistent with evolutionary history, and definitely more com-patible with the present data. More enlightened speculation as to the exact nature of the effects of this linguistic mechanism must await further investigation into the nature of variables affecting ear asymmetry. Developmental Trends in Dichotic Listening Two major developmental trends are.evident in the present results, one being increased performance with age, and the other, increased ear asymmetry with age. Each trend w i l l be discussed separately. 58 The increase in overall recognition performance with age is reflected in the significant linear component of the grade effect. The phenomenon has been observed by several investigators (Bryden, 1970; Inglis & Sykes, 1967; Kimura, 1963, 1967; Knox & Kimura, 1970) but only Inglis & Sykes (1967) have offered an explanation. They have suggested that increase in performance could be due to an increase in short-term memory storage capac-ity with age. Although the hypothesis would admittedly be relevant to the multiple-pair-per-trial free r e c a l l procedure, i t would hardly seem probable that i t could account for the linear effect observed in the present experiment where only one pair of stimuli per t r i a l was presented, and where a recognition rather than a recall task was employed. Knowledge of the exact mechanism accounting for this developmental trend w i l l have to await further investigation. One of the unexpected findings i n the present study was the differen-t i a l effect of age on the recognition of various types of stimulus material, recognition of some materials increasing at a faster rate than recognition of others. The various types of stimulus material also differed in level of d i f f i c u l t y , vowel recognition proving to be easier than consonant rec-ognition, with the recognition of music and sound effects being easier than both. Although a relationship could conceivably be expected between the effect of age on a task and i t s level of d i f f i c u l t y , comparison of the effects of age on sound effect recognition and music recognition-two tasks of comparable difficulty-negates such a premise. Music recognition is the least affected by age of the stimuli presented, while sound effect recogni-tion i s affected the most. Thus the differential effects of age on separate identification tasks remain unexplained. 59 Regarding the development of ear asymmetry, the present data indicate that in general, the largest portion of the developmental trend takes place before grade 4, only a small increase in ear asymmetry being evident between grades 4 and 6. The actual age at which ear asymmetry is f i r s t apparent is somewhat uncertain. Ear preferences are already present in grade 2 for g i r l s in the recognition of consonant-varied stimuli, for boys in the recognition of sound effects, and for both sexes in the recognition of musical stimuli. For the remaining groups, however, no ear-preference is evident in grade 2. For these groups, ear asymmetry is f i r s t evident, although not necessarily significant in a l l cases, in grade 4. Both sexes show significant ear differences in the.recognition of a l l four stimulus materials by grade 6. Although the evidence is not complete, i t would seem probable that ear asymmetry f i r s t originates between the approximate ages of 5 and 8. The data also show that in general, this asymmetry increases with age through the elementary school years. The data are con-sistent with the results of studies using brain-damaged Ss, these results showing l i t t l e difference in the effects of left-versus right-hemisphere damage before the age of 5, and as discussed above, a gradual increase i n l e f t hemisphere involvement in language function during the elementary school years. The data are also in agreement with those of Bryden (1970) who has reported an increase in the frequency of right-ear superiority with verbal stimuli from grade 2 to grade 6. However, the present findings con-tradict those of Kimura (1963, 1967, 1970) who has shown ear asymmetry to be present already at the age of 4, and to decrease rather than to increase with age. An attempt to explain this discrepancy is presented below. A rather unexpected finding of the present study was the presence of 60 sex differences in ear-asymmetry development. For both consonant-varied and vowel-varied stimulus recognition, boys lagged behind g i r l s , the adult pattern being reached by grade 4 with g i r l s and by grade 6 with boys. Terman & Tyler (1954) present evidence to show that g i r l s excel boys in almost a l l speaking s k i l l s at least in the early school years. If there is indeed a relationship between the earlier development of ear asymmetry with speech stimuli in g i r l s , and their superior a b i l i t y with language, the pre-sent investigation would suggest that this difference may be primarily a developmental one, g i r l s going through the developmental cycle more rapidly than boys, but both arriving eventually at the same level. The basis for this statement would be that while sex differences are found in grades 2 and 4 with verbal material, no such differences are evident in grade 6. The reverse finding in regard to sex differences i s found with sound effect recognition, boys showing ear asymmetry in grade 2 while g i r l s do not do so u n t i l grade 6. There is evidence to suggest that boys are superior to g i r l s in visiospatial a b i l i t y (Terman & Tyler, 1954), but what relation, i f any, this a b i l i t y would have to the recognition of sound effects i s not at a l l clear. The present study has provided information as to the developmental aspects of ear asymmetry from grades 2 to 6, but does not in i t s e l f provide information as to the relation between ear-asymmetry at grade 6 and ear asymmetry at maturity. According to the findings with brain-damaged Ss, the adult pattern of cerebral dominance for speech perception is achieved at about grade 6, the effects of brain damage on speech perception remaining relatively constant from this age on. Such observations suggest that l i t t l e change in ear asymmetry would be evident between grade 6 and adulthood. 61 To evaluate this premise, the consonant and vowel recognition data from the present experiment were compared with the relevant data from the study conducted by Spellacy & Blumstein (1970) with college students as Ss. When such a comparison is made, the mean recognition scores for the preferred ear are found to be almost the same for Ss in grade 6 and Ss in college. The mean recognition scores for the nonpreferred ear, however, are much lower for the grade 6 Ss than for the college student Ss. For consonants, right ear performance i s comparable (mean recognition score of 8.20 vs. 8.21) while the performance on the l e f t ear i s less for Ss in grade 6 than in college (8.50 vs. 7.97). For vowels, the reverse i s found, l e f t -ear performance being similar (8.98 vs. 9.02) while right-ear performance is much less for Ss in grade 6 than in college (8.08 vs. 8.83). Thus, although preferred ear performance has reached asymptote by grade 6, per-formance with the nonpreferred ear is s t i l l increasing with age. The result i s a larger ear difference in grade 6 than at college age, ear asymmetry increasing during elementary school years, and then decreasing to maturity. Unfortunately, there i s a scarcity of relevant data available that can be used to help interpret this phenomenon. As the preferred ear is different for the two materials, the mechanism responsible would appear to be unrelated to cerebral dominance. A differential rate of maturation for the two hemispheres would also f a i l to account for the phenomenon. A third alternative i s the possible influence of order of report factors. Using multiple-pair free recall procedures, both Bryden (1970) and Inglis & Sykes (1967) found an increase in performance with age in favor of the second ear reported. As discussed previously, several investigators have 62 found that the preferred ear is usually reported f i r s t , the nonpreferred ear being reported second. At f i r s t glance, i t would seem highly unlikely that order of report factors would be influencing the results in a one-pair-per-trial recognition procedure. A very tentative possibility could be that an order of report factor in the form of covert rehearsal in favor of the preferred ear i s taking place, and thus interfering^ with the retention of the stimulus presented to the nonpreferred ear. Such a decreased level of performance in the nonpreferred ear would lead to an exaggerated ear preference. This interference with the nonpreferred ear performance would be expected to decrease with age, as order of report factors appear to be negligible in the studies discussed above using adults as Ss. The present findings are compatible with such a notion. Assuming the above premise to be true, one would expect that dichotic listening studies sensitive to order of report factors would find rather large ear differences at comparatively earlier ages with a progressive dec-rease in ear asymmetry with age. On the other hand, a procedure relatively insensitive to order of report factors would show ear asymmetry developing at a later age, this ear difference increasing rather than decreasing with age during the elementary school years. To the extent that order of report factors were influencing the results, a decrease in ear asymmetry would be evident between grade 6 and maturity. In view of these predictions, i t is rather interesting that Kimura (1963, 1967, 1970), using a procedure rather sensitive to order of report factors, found relatively large ear asymmetry present at the age of 4, with ear differences decreasing progressively with age. In contrast, Bakker (1970) in a recent study using a monotic stimulation procedure in which order of report factors would be completely 63 excluded, found a significant increase in ear asymmetry occuring during the elementary school years. The present investigation, which would be relatively insensitive to order of report factors, also shows this increase in ear asymmetry with age. It i s very unclear at the present state of knowledge whether the inter-ference with retention of stimuli presented to the nonpreferred ear i s , in fact, caused by covert rehearsal in favor of the preferred ear, thus resulting in greater ear differences being obtained in childhood than adulthood. It would appear, however, that two separate mechanisms are res-ponsible for the ear asymmetries obtained, one related to cerebral dominance for speech, and the other specific to the identification of stimuli presented to the nonpreferred ear, regardless of whether this nonpreferred ear is l e f t or right. In general, the results of the present investigation support the hypo-thesis of a relationship between left-hemisphere specialization for speech and ear asymmetry. Both the age of onset of ear asymmetry, and the posi-tive relationship between magnitude of ear asymmetry and age during the elementary school years, would support such a premise. The data, however, do not necessitate the postulation of a right-hemisphere specialization for nonspeech stimuli. Both right-and left-ear advantages were found to be the apparent result of differences in left-hemisphere function only. 64 References Bakker, D.J. Ear-asymmetry with monaural stimulation. Psychonomic  Science, 1968, 12, 62. Bakker, .D.J. Ear-asymmetry with monaural stimulation: relations to lateral dominance and lateral awareness. Neuropsychologia, 1970, 8, 103-117. Bartz, W.H., Satz, P., Fennel, E., & Lally, J.R. Meaningfulness and late r a l i t y in dichotic listening. Journal of Experimental Psychology, 1967, 73, 204. Benton, A.I., The problem of cerebral dominance. The Canadian Psychologist, 1965, 6a, 332-348. Bocca, E., Calearo, C , Cassinar, V., & Migliavacca, F. Testing "co r t i c a l " hearing in.temporal lobe tumours. Acta Oto-Laryngologica, 1955, 45, 289-304. Borkowski, J.G., Spreen, 0., & Stutz, J.Z. Ear preference and abstract-ness in dichotic listening. Psychonomic Science, 1965, 3_, 547-548. Broadbent, D.E. The role of auditory localization in attention and memory span. Journal of Experimental Psychology, "$953, 47, 191-196. Broadbent, D.E. Immediate memory and simultaneous stimuli. Quarterly  Journal of Experimental Psychology, 1957, _, 1-11. Broadbent, D.E. Perception and Communication. New York: Pergamon Press, 1958. Broadbent, D.E. & Gregory, M. Accuracy of recognition for speech presented to the right and l e f t ears. Quarterly Journal of Experimental  Psychology, 1964, 14, 359-360. Bryden, M.P. Ear preference i n auditory perception. Journal of Experi- mental Psychology, 1963, _5, 103-105. Bryden, M.P. Tachistoscopic recognition, handedness, and cerebral dominance. Neuropsychologica, 1965, 3_, 1-8. Bryden, M.P. An evaluation of some models of la t e r a l i t y effects in dichotic listening. Acta Oto-Laryngologica, 1967, 63, 595-604. Bryden, M.P. Laterality effects i n dichotic listening: relations with handedness and reading a b i l i t y in children. Neuropsychologica, 1970, 8, 443-450. 65 Chaney, R.B. & Webster, J.C. Information in certain multidimensional sounds. Journal of the Acoustical Society of America, 1966, 40, 447-455. Corballis, M.C. Immediate recall of spoken digits presented three at a time. Journal of Speech and Hearing Research, 1969, 21, 416-424. Curry, F.K.W. A comparison of left-handed and right-handed subjects on verbal and nonverbal dichotic listening tasks. Cortex, 1967, _3, 343-352. Curry, F.K.W. & Rutherford, D.R. Recognition and recall of dichotically presented verbal stimuli by right-and left-handed persons Neuropsychological 1967, 5, .119-126. Darwin, C.J. Laterality effects in the recall of steady-state and transient speech sounds. Journal of the Acoustical Society of  America, 1969, 96, .'114 (abstract), (a) Darwin, C.J. Auditory perception and central dominance. Unpublished Ph.D. Thesis, University of Cambridge, 1969. (b) Darwin, C.J. Ear differences i n the r e c a l l of fricatives and vowels. Quarterly Journal of Experimental Psychology, 1971, 23, 46-62. Dirks, D.D. Perception of dichotic and monaural verbal material and cerebral dominance for speech. Acta Oto-Laryngologica, 1964, 58, 73-80. Haggard, M.P* Perception of semi-vowels and laterals. Journal of the  Acoustical Society of America, 1969,' 46, 115 (abstract). Haggard, M.P. Encoding and the REA for speech signals. Quarterly Journal  of Experimental Psychology, 1971, 23, 34-45. Haggard, M.P. & Parkinson, A.M. Stimulus and task factors as determinants of ear advantages. Quarterly Journal of Experimental Psychology, 1971, 23, 168-177. Hall, J.L. & Goldstein, M.H. Representation of binaural stimuli by single units i n primary auditory cortex of unaneasthetized cats. Journal  of the Acoustical Society of America, 1968, 43, 456-461. Halwes, T. Effect of dichotic fusion i n the perception of speech. Supplement to Status Report on Speech Research, September, 1969, Haskins Labs. Inglis, J. Dichotic listening and cerebral dominance. Acta Oto-Laryn- gologica, 1965, 60, 231-238. Inglis, J. & Ankus, M.N. Effects of age on short-term storage and serial rote learning. British Journal of Psychology, 1965, 56, 183-195. 66 I n g l i s , J . & Sykes , D.H. Some s o u r c e s o f v a r i a t i o n i n d i c h o t i c l i s t e n i n g p e r f o r m a n c e i n c h i l d r e n . J o u r n a l o f E x p e r i m e n t a l C h i l d P s y c h o l o g y , 1967, 5, 480-488. J o n e s , D. & Spreen, 0. D i c h o t i c l i s t e n i n g by r e t a r d e d c h i l d r e n : t h e e f f e c t s o f e a r o r d e r and a b s t r a c t n e s s . C h i l d Development, 1967, 38, 101-105. K i m u r a , D. Some e f f e c t s o f t e m p o r a l l o b e damage on a u d i t o r y p e r c e p t i o n . C a n a d i a n J o u r n a l o f P s y c h o l o g y , 1961, 15, 156-165. (a) Kim u r a , D. C e r e b r a l dominance and t h e p e r c e p t i o n o f v e r b a l s t i m u l i . C a n a d i a n J o u r n a l o f P s y c h o l o g y , 1961, 15, 166-171. (b) Kim u r a , D. Speech l a t e r a l i z a t i o n i n young c h i l d r e n as d e t e r m i n e d by an a u d i t o r y t e s t . J o u r n a l o f C o m p a r a t i v e P h y s i o l o g i c a l P s y c h o l o g y , 1963, 56, 899-902. K i m u r a , D. L e f t - r i g h t d i f f e r e n c e s i n t h e p e r c e p t i o n o f m e l o d i e s . Q u a r t e r l y J o u r n a l o f E x p e r i m e n t a l P s y c h o l o g y , 1964, 14, 355-358. K i m u r a , D. F u n c t i o n a l asymmetries o f t h e b r a i n i n d i c h o t i c l i s t e n i n g , C o r t e x , 1967, 3, 163-178. K i m u r a , D. & F o l b , S. N e u r a l p r o c e s s i n g o f backward speech sounds. S c i e n c e , 1968, 161, 395-396. K i r s t e i n , E. & S h a n k w e i l e r , D.P. S e l e c t i v e l i s t e n i n g f o r d i c h o t i c a l l y p r e s e n t e d c o n s o n a n t s and v o w e l s . Paper t o E.P.A., P h i l a d e l p h i a , A p r i l , 1969. Knox, C. & Ki m u r a , D. C e r e b r a l p r o c e s s i n g o f n o n v e r b a l sounds i n boys and g i r l s . N e u r o p s y c h o l o g i c a , 1970, 8_, 227-237. L e n n e b e r g , E.N. Speech development: i t s a n a t o m i c a l and p h y s i o l o g i c a l c o n c o m i t a n t s . I n E.C. C a r t e r e t t e ( E d . ) , B r a i n F u n c t i o n , Volume 3. B e r k e l y : U n i v e r s i t y o f C a l i f o r n i a P r e s s , 1966, Pp. 37-66. L i b e r m a n , A.M., Cooper, F.S., S h a n k w e i l e r , D.P., & Stud d e r t - K e n n e d y , M. P e r c e p t i o n o f t h e speech code. P s y c h o l o g i c a l Review, 1967, 74, 431-461. L i c k l i d e r , J.C.R. B a s i c c o r r e l a t e s o f t h e a u d i t o r y s t i m u l u s . I n S.S. Ste v e n s ( E d . ) , Handbook o f E x p e r i m e n t a l P s y c h o l o g y . New Y o r k : J o h n W i l e y & Sons, 1953, Pp. 985-1039. L i e b e r m a n , P. P r i m a t e v o c a l i z a t i o n s and human l i n g u i s t i c a b i l i t y . J o u r n a l o f t h e A c o u s t i c a l S o c i e t y o f A m e r i c a , 1968, 44, 1574-1584. L i n d b l o m , B.E.F. & St u d d e r t - K e n n e d y , M. On t h e r o l e o f formant t r a n s i -t i o n s i n v o w e l r e c o g n i t i o n . J o u r n a l o f t h e A c o u s t i c a l S o c i e t y o f A m e r i c a , 1967, 42, 830-843. 67 Milner, B. Laterality effects in audition. In V.B. Mountcastle (Ed.), Interhemispheric Relations and Cerebral Dominance. Baltimore: John Hopkins University Press, 1962, Pp. 177-195. Milner, B., Branch, C. & Rasmussen, J. Observations on cerebral dominance. In C.V.S. de Rueck & M. O'Connor (Eds.), Ciba Foundation Symposium on  Disorders of Language. London: Churchill, 1964, Pp. 200-214. Milner, B., Taylor, L., & Sperry, R.W. Lateralized suppression of dichoti-cally presented digits after commissural section in man. Science, 1968, 161, 184-186. Oxbury, S., Oxbury, J., & Gardiner, J. Laterality effects in dichotic listening. Nature, 1966, 214, 742-743. Penfield, W. & Rasmussen, T. The Cerebral Cortex of Man. New York: Macmillan, 1950. Rosenzweig, M.R. Representations of the two ears at the auditory cortex. American Journal of Physiology, 1951, 167, 147-158. Satz, P. Laterality effects in dichotic listening. Nature, 1968, 218, 277-288. Satz, P., Achenbach, K., Pattishal, E., & Fennel, E. Order of report, ear asymmetry and handedness in dichotic listening. Cortex, 1965, 1_, 377-396. Shankweiler, D. Effects of temporal-lobe damage on perception of dichoti-cally presented melodies. Journal of Comparative Physiological  Psychology, 1966, 62, 115-119. Shankweiler, D. & Studdert-Kennedy,,M. Identification of consonants and vowels presented to l e f t and right ears. Quarterly Journal of Experimental Psychology, 1967, 19, 59-63. Spellacy, F. Lateral preferences in the identification of patterned stimuli. Journal of the Acoustical Society of America. 1970, 47, 574-578. Spellacy, F. & Blumstein, S. The influence of language set on ear preference in phoneme recognition. Cortex, 1970, 6, 430-439. Spreen, 0., Spellacy, F.J. & Reid, J.R. The effect of interstimulus interval and intensity on ear asymmetry for nonverbal stimuli in dichotic listening. Neuropsychologica, 1970, 8_, 245-250. Studdert-Kennedy, M. & Shankweiler, D. Hemispheric specialization for speech perception. Journal of the Acoustical Society of America, 1970, 48, 579-594. Terman, L.M., & Tyler, L.E. Psychological sex differences. In L. Carmichael (Ed.), Manual of Child Psychology. New York: Wiley Press, 1954. 68 Treisman, A.M., & Geffen, G. Selective attention: perception or response. Quarterly Journal of Experimental Psychology,1967, 19, 1-18. Treisman, A.M. & Geffen, G. Selective.attention and cerebral dominance in perceiving and responding to speech messages. Quarterly Journal  of Experimental Psychology, 1968, 20, 139-150. Tunturi, A.R. A study of the pathway from the medial geniculate body to the acoustic cortex in the dog. American Journal of Physiology, 1946, 147, 311-319. Weinstein, S. & Sersen, E.A. Tactual sensitivity as a function of handedness and lat e r a l i t y . Journal of Comparative and Physiological  Psychology, 1961, 54, 665-669. Zangwill, O.C. Cerebral Dominance and i t s Relation to Psychological  Function. Edinburgh: Oliver & Boyd, 1960. Zurif, E.B. & Bryden, M.P. Familial handedness and left-right differences in auditory and visual perception. Neuropsychologica, 1969, ]_, 179-187. 69 Appendix Procedure used to adjust ear condition total in grade 4 for the effects of grade An estimate of the increase from grade 2 to grade 4 due to the effect of grade was calculated separately for each sex within each type of stimulus material. This was done by dividing the difference between the grade 6 condition total and the grade 2 condition total by four. The resulting sum was subtracted from both the left-ear and the right-ear condition totals in grade 4 to provide an estimate of recognition per-formance when the increase due to the effect of grade was excluded. The following is an example of the procedure with consonant-varied stimuli presented to males. Data Condition totals for: grade 6 = 477 grade 2 = 416 grade 4 left-ear = 203 grade 4 right-ear = 241 Conditions Adjusted grade 4 left-ear condition total = grade 4 left-ear condition total - /grade 6 condition total - grade 2 condition t o t a l ^ = 203 -^477-416 ^  = 188.5 Adjusted grade 4 right-ear condition total = grade 4 right-ear condition total -/grade 6 condition total - grade 2 condition tot a l \ \ • 4 ; = 241 - ^ 477-416^ = 225.5 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0102036/manifest

Comment

Related Items