Newopsychologia, Vol. 31, No. 4, pp. 347-353, 1993. Printed in Great Britain.

0028-3932/93 $6.00+0.00 Pergmon Press Ltd

DYNAMIC GENDER-RELATED DIFFERENCES IN DICHOTIC LISTENING PERFORMANCE

PAM MUNRO and ERNEST GOVIER

Department of Psychology, University of East London, U.K.

(Received 8 June 1992; accepted 2 November 1992)

Abstract-This experiment investigated dynamic gender-related differences in perceptual asymmetry (PA) in a dichotic task. Twenty right-handed males and 20 right-handed females performed a 200 trial directed attention dichotic listening task using consonant-vowel&consonant nonsense syllables. Men showed a greater right-ear advantage (REA) at the beginning of the test than did women. Over the course of the test the REA in men declined significantly (P<O.OOl) whilst in females the REA showed a significant increase (P<O.OOl). Thus by the end of the test the REA was greater in females than males. However, the reduced REA in males was entirely due to improved left-ear (LE) performance and the increased REA in females was due to improved right-ear (RE) performance. The change in PA in women is consistent with a dynamic improvement in left hemisphere (LH) syllable processing and in men a dynamic improvement in the transmission of information through the right hemisphere (RH) to the language processing areas in the LH.

INTRODUCTION

IN STUDIES using the dichotic listening procedure significant sex differences have been reported [12, 131 in which 94% of men but only 67% of women showed a right-ear advantage (REA) for verbal stimuli; thus supporting the notion of greater left hemispheric lateralization for language in males. A significantly larger REA in males has also been reported [6]. Conversely, a trend towards a bigger REA advantage for women has been found [S], whilst no sex-related differences in a sample of 10 right-handed men and 10 right- handed women has also been demonstrated.

None of these studies, however, controlled the order of report and thus results are vulnerable to the effects of preferential processing or reporting strategy on the part of the subject (see Refs [3] and [4]). Unfortunately, in studies which have controlled for such effects findings are equally inconsistent. Thus whilst one study [2] found no evidence of sex difference in ear asymmetry, another [15] reported a highly significant REA in males but only a weak non-significant REA in females. Nevertheless, in a major review of the area [3] it has been argued that there is evidence of a greater REA for verbal stimuli in males than females although it is small and of marginal significance.

Few studies have investigated sex-related differences for non-verbal dichotic material. However, one study [l l] found that the typical left-ear advantage (LEA) for the perception of melodies (e.g. Ref. [8]) did not vary by sex, while another [15] found that, whilst men showed only a small but insignificant LEA for melodies and environmental sounds, females demonstrated a highly significant LE superiority suggesting greater lateralization in females for non-verbal stimuli.

More recently, researchers [21] have reported sex differences in the dynamic aspect of

341

348 P. MUNRO and E. COWER

cerebral functional asymmetry (CFA). At the beginning of a 120 trial fused dichotic word test males showed a greater REA than females. Over the first half of the test the REA in males declined whilst in females the REA increased. For the latter half of the test the two groups had equal mean asymmetry scores. The main effect of sex or time was not significant.

Although differences over time on dichotic listening tasks have been reported for subjects who differ in defensive coping style [22] this dimension of function had not previously been considered in comparisons of the sexes. However, it has been suggested [21] that this difference in the dynamic aspect of CFA could account for at least some of the inconsistencies which appear to be a major feature of studies of sex differences in laterality.

The present experiment was designed to test the robustness of the reported sex differences in the dynamic component of CFA. In contrast to the original study [21] which used a fused dichotic word test recorded by a male speaker and required Subjects (Ss) to make a manual response, the current study used a directed attention dichotic procedure with consonant- vowel-consonant nonsense syllables recorded by a female speaker and required Ss to make a verbal response. Sidedness (hand, eye, foot, ear preference) was used as an indicator of cerebral lateralization for language function as opposed to handedness which has on its own been shown to be a lesser predictor [18, 191.

Subjects

METHOD

Forty subjects participated, 20 males (age range 22-38 years, mean 29.9 years), and 20 females (age range 2240 years, mean 31.8 years), from a range of professional occupations and undergraduate courses. Subjects were co- opted from associates of the experimenter. All subjects were right-sided as assessed by a lateral preference questionnaire [17]. No subjects reported any hearing problems, speech or language impediments or neurological deficits. English was the first language of all subjects.

Procedure

The tape had been constructed specifically for dichotic listening procedures. The six stop consonants/b, d, g, p, t and k/ with an intermediate vowel forming unfamiliar consonant-vowel-consonant (CVC) syllables, spoken in a female voice comprised the stimulus set. All possible combinations had been employed and those with meaningful connotations, e.g. kot, peg, omitted. This resulted in a total of 67 usable syllables which were paired so they differed only in respect to the initial consonant, the medial vowel or the final consonant, e.g. dat dag, ged ped. These had been reversed and mirrored forming 100 pairs of syllables and randomized into 5 sets of 20. Each set of 20 syllable pairs was designated as I block and each syllable pair 1 trial. Six randomly selected syllable pairs comprised a practice block.

RE syllables had been recorded at an interstimulus interval of 6 set on to one channel of a BASF reel to reel tape using a Ferrograph Y722H running at 15 in./sec, at O-3 vol. units, using a Shure Unisphere I Model 565 microphone. This was transcribed onto a Ferrograph 7522H and monitored through Alba H6 stereo headphones whilst LE syllables were simultaneously recorded onto another channel. The recording was then transcribed onto a Maxwell stereo cassette tape. The test stimuli were presented through a Marantz stereo cassette player Model CP230 and received through Alba stereophonic headphones Model H6. Volume was set at the mid-level as indicated by the volume level indicator.

Subjects were tested individually in a quiet room, free from external noise and distraction. Standard instructions were given throughout the experiment. Each session lasted between 45 min and 1 hr.

Subjects were informed that they would be required to listen through headphones to a tape recording of pairs of unfamiliar syllables. Example stimuli were presented and reading time allowed. It was explained that one of the pair ofsyllables would be played in the left ear whilst the other would be presented simultaneously in the right ear. It was explained that the syllables would be presented at 6-set intervals. Subjects were informed that they would be instructed to concentrate on what was being played in one particular ear only and to report verbally what was heard in that ear.

Headphones were then placed on the subject in the manner dictated by the experimental condition (see below) and the practice session played. This consisted of six trials. Before the trials were presented subjects were randomly assigned to attend to their right or left ear. After the first three trials the tape was stopped and the subject instructed

DIFFERENCES IN DICHOTIC LISTENING 349

to attend to the opposite ear for the remaining trials. Opportunity was provided for questions before the experimental session began.

In the experimental session each of the five blocks of 20 trials was played twice making a total of 200 trials. Between each 20-trial block there was a lo-set interval. The order of block presentation was counter-balanced over subjects. For each block subjects were intruded to attend to either their right or left ear only, the order ofwhich was also counter-balanced. For each 40-trial time segment subjects were right-ear (RE) directed on 20 trials and left-ear (LE) directed on 20 trials. In addition headphones were reversed over subjects counter-balancing for channel differences. Overall this produced 20 different experimental conditions to which subjects were allocated.

Subjects were asked not to omit any trials throughout the procedure and to respond with “pass” if any syllables could not be clearly distinguished.

On completion of the dichotic task subjects were asked to complete the lateral preference questionnaire. At the end of the session subjects were debriefed.

Scoring

One point was given for each correct response. Separate scores were determined for each block of 20 trials. Scores were divided into five time periods:

Period 1: Trials 1- 40 Period 2: Trials 41- 80 Period 3: Trials 81-120 Period 4: Trials 121-160 Period 5: Trials 161-200.

Differences in performance between ears for each subject within the five time periods were calculated by subtracting correct LE responses from correct RE responses. Perceptual asymmetry scores for each subject were calculated by further dividing these results by the sum of the correct LE and RE responses.

RESULTS

Mean laterality quotients indicate both males (j= 14.73) and females (j= 14.82) were strongly right-sided (maximum score 16 points) with minimal difference between the two groups.

Means and standard deviations of RE and LE responses for the five time periods and for overall performance are shown in Table 1. As can be seen a REA is present over all time periods for both males and females.

Perceptual asymmetry scores [(RE - LE)/(RE + LE)] are displayed in Fig. 1.

Table 1. Means (I) and standard deviations (S.D.) of correct right-ear (RE) and left-ear (LE) responses for the five time periods

Trials Females Males

RE LE RE LE

l-40

41-80

81-120

121-160

161-200

Overall

S.xD.

S.“D.

S.xD.

S.L.

S.xD.

S.xD.

9.90 8.65 13.05 9.10 (4.76) (3.90) (3.39) (3.45) 10.25 8.15 13.45 10.15 (3.65) (3.17) (3.27) (3.05) 10.65 8.80 13.15 10.60 (3.87) (3.71) (2.32) (3.12) 11.20 7.70 12.40 10.25 (3.55) (3.39) (3.27) (3.05) 11.45 8.15 12.60 10.75 (4.24) (3.22) (2.96) (3.42) 10.69 8.29 12.93 10.17 (4.00) (3.45) (2.96) (3.27)

350 P. MUNRO and E. GOVIER

0.24 r q Malea

01 I I I I J Triatr l-40 Trialr ‘,l* Trial; 81-120 m 121_1~Tri&s 161-200

Fig. 1. Changes over time in perceptual asymmetry for male and female subjects.

TREATMENT OF RESULTS

Lateral preference

The lateral preference scores were subjected to an independent t-test. Males and females did not differ in sideness (t= -0.303, d.f. 38, P>O.O5).

Dichotic task

Data from the dichotic listening task was subjected to a 3-way crossed nested (n> 1) ANOVA with sex, subjects and ear as the main effects. The main effect sex was significant [F (1, 38)=5.078, P<O.O5] with males better overall at the task than females. Performance also differed significantly between subjects [F (38, 320) =24.619, P<O.OOl]. The main effect ear also proved highly significant [F (1,38) = 75.2471 the right ear being more efficient than the left. However, the interaction effect sex x ear was not significant [F (1, 38) =0.4522,

P>O.O5].

Changes over time

When the differences between RE and LE performance were subjected to trend analysis females showed a highly significant increase in REA over the five time periods (Pages L = 981, P < 0.001). In contrast males showed a significant decline in REA over the five time periods (Pages L= 1006, P<O.OOl).

Trend analysis of male RE performance showed no significant increase over the five time periods (Pages L = 853, P> 0.05) whilst there was a significant increase in LE performance (Pages L = 975, P< 0.05). In females, RE performance increased significantly over the five time periods (Pages L = 953, P < 0.05) whilst LE performance showed no significant increase (Pages L = 874, P> 0.05).

DISCUSSION

The results demonstrate a significant REA reflecting left-hemisphere specialization for language functions. Overall performance was better in males than females but the sex-by-ear interaction was non-significant. Differences in perceptual asymmetry did, however, emerge when performance over time was considered. Men showed a greater REA at the beginning of the test than did women. This sex difference decreased during the first half so that by the third

DIFFERENCES IN DICHOTIC LISTENING 351

time period (trials 81-120) mean asymmetry scores were almost the same. This pattern of change over time replicates that reported by WEXLER and LIPMAN [21]. During the remaining trials, however, the REA continued to decline in men and to increase in women, hence by the end of the test the REA was greater in women than men. Thus, over time there was a complete reversal in the sex difference in PA (Fig. 1).

According to KIMURA’S [7,9] structural hypothesis ear advantages on dichotic listening tasks arise because of the greater efficiency of the contralateral, than ipsilateral auditory pathways from each ear to the cerebral cortex. Verbal stimuli presented in the RE follow the more efficient contralateral route direct to the language specialized left hemisphere. Information presented to the LE must first go to the right auditory cortex before transfer across the corpus callosum to the left hemisphere for final processing. This causes both delay and decay of the LE signal.

Rather than a structural advantage KINSBOURNE [lo] has stressed attentional factors in determininglaterality effects. Kinsbourne argues that orientation to one hemispace serves to prime the contralateral hemisphere making it more sensitive to stimuli in the attended location. Anticipation of a verbal task triggers the language specialized left hemisphere activity and attention is biased to the right with consequent superior processing of information in the right sensory field.

Evidence suggests that neither view is completely correct, rather both structural and attentional components appear to be relevant to the production of dichotic laterality effects [3]. This situation has been made even more complex by several studies (e.g. Refs [14] and [ 161) which have demonstrated that the perceived position of a sound source rather than its actual position, or ear of entry is important in determining behavioural asymmetries. Thus hemisphere activity is not governed wholly by stimulus attributes.

WEXLER and LIPMAN [21] attribute their findings to initial sex differences in relative hemisphere activation which are superimposed upon these multiple processes and components which contribute towards the usual REA in dextrals. They argue that in males task anticipation (or commencement) is associated with left hemisphere activation, whilst in females it is associated with right hemisphere activation. This approach-related left hemisphere activation in males leads to a temporary increase in the usual REA, while the approach-related right hemisphere activation in females leads to a temporary decrease in the usual REA. These differences are seen to reflect different modes of attention with which women and men approach tasks; women are assumed to adopt a holistic approach whilst men approach tasks in a more analytic manner.

The present study exactly replicates that of WEXLER and LIPMAN’S [21] in demonstrating an increase in female REA and a decrease in male REA over time emphasizing the robustness of the findings. However, it goes beyond the Wexler and Lipman study in three ways; (1) non- fused stimuli were used, (2) dynamic gender-related differences in REA were demonstrated over a substantially longer period of time and (3) the final REA in females is actually greater than the final REA in males.

Significantly, however, the change in perceptual asymmetry in women is due to an increasing accuracy of RE performance; not a reduction in LE performance as the putative temporarily increased activation of the RH diminishes (Fig. 2).

In men LE performance improves whilst recall from the RE remains fairly constant (Fig. 3). Current findings therefore are not readily explained by the model of hemisphere activation proposed by WEXLER and LIPMAN [21]. Rather the pattern is consistent with increasing left hemisphere activation in women and in men of increasing right hemisphere

352 P. MUNRO and E. GOVIER

01 I I I I I Trials l-40 Ttis 41-80 Trials 81-120 TV 121_160 Tri& 161-200

Fig. 2. Changes over time in mean number of correct responses for right and left ears (n = 20 each ear) for female subjects.

1

4.5 - rlRightear

3.0 - AL&ear z

1.5 -

0 I I I I I Trials l-40 Tti 41-80 Trials 81-lZOTw 121_160 Tzkls 161-200

Fig. 3. Changes over time in mean number of correct responses for right and left ears (n = 20 each ear) for male subjects.

activation. Therefore if differential approach-modes were to be implicated at all it would seem to be in terms of structurally based attentional habits which the subject learns to modify with practice.

Undoubtedly current findings to not present an easily interpretable picture. Nevertheless, it is possible that the increasing lateralization in women is a reflection of their more distributed representation of language functions. With greater distribution of function it may be that there is greater competition/interference between pairs of simultaneously presented dichotic stimuli. This would result in reduced performance overall. However, with prolonged practice female Ss learn to improve RE performance possibly as a consequence of greater LH language involvement.

Conversely, in males the strong lateralization of function limits the possibility of competition/interference effects and confers a clear REA. Practice effects under such circumstances may serve to improve the transmission of information through the RH to the language processing areas in the LH.

DIFFERENCES IN DICHOTIC LISTENING 353

The REA found in language-related dichotic tests results from the net effect of multiple neuronal processes and components. Changes in function of any of these factors therefore has the potential to alter the magnitude of the PA [20]. Current data indicate that at least some, if not all of these processes are sensitive to changes over time. Furthermore, the resultant effects are markedly different in males and females. Whatever the underlying causal mechanism(s) for these differential responses, present findings clearly demonstrate that the phenomenon is far from transient as suggested by WEXLER and LIPMAN [21]. Evidence of changes over time in PA adds another dimension to the complex arena of possible sex-related differences in CFA. Undoubtedly sex differences in this dimension of function could account for at least some of the inconsistencies found in the literature on sex differences in laterality.

REFERENCES 1. BRYDEN, M. P. Tachistoscopic recognition, handedness and cerebral dominance. Neuropsychologia 3, 1-8,

1965. 1. BRYDEN, M. P. Sex-related Differences in Cognitioe Function, M. A. WITTIG and A. C. PETERSEN (Editors),

pp. 121-139. Academic Press, New York, 1979. 3. BRYDEN, M. P. Handbook of Dichotic Listening: Theory, Methods and Research, K. HUGDAHL (Editor),

pp. 143. Wiley & Sons, Chichester, 1988. 4. BRYDEN, M. P., MUNHALL, K. and ALLARD, F. Attentional biases and the right-ear effect in dichotic listening.

Brain Lang. 18, 236248, 1983. ,’ 5. DEMAREST, L. and DEMAREST, J,&e interaction of handedness, familial sinistrality and sex in determining

cerebral dominance for verba&materi&Paper presented at the annual meeting of the Eastern Psychological Association 1979 cited M. P.jBRYDE~~Lateralfty. Academic, New York, 1982.

6. HARSHMAN, R. A.‘, REMMINGTON, R. ana KRASHEN, S. D. Sex, Language and the Brain. Paper presented at UCLA Conference on Human Brain Function. Los Angeles 1974.

7. KIMURA, D. Cerebral dominance and the perception of verbal stimuli. Can. J. Psychoi. 15, 166-171, 1961. 8. KIMURA, D. Left-right differences in the perception of melodies. Q. J. exp. Psychol. 16, 355-358, 1964. 9. KIMURA, D. Dual functional asymmetry of the brain in dichotic listening. Cortex 3, 163-178, 1967.

10. KINSBOURNE, M. The cerebral basis of lateral asymmetries in attention. Acta gsychologia 33, 193-201, 1970. 11. KING, L. Brain asymmetry in t of verbal and non-verbal sounds. _A,_Asis’ London Canada:

University of Western Ontario, McGlone, Sex differences in func $” J lonal brain asymmeir :a- survey. Behav. Brain Sci. 3, 21

12. LAKE, D. A. and BRYDEN, M. P. Handedness and sex differences in hemispheric asymmetry. Brain Lang. 3, 266-282, 1976.

13. MCGLONE, J. and DAVIDSON, W. The relationship between cerebral speech laterality and spatial ability with special reference to sex and hand preference. Neuropsychologia 11, 105-113, 1973.

14. MORAIS, J. Spatial constraints on attention to speech. J. REQUIN (Editor), pp. 245-260. Attention and Performance VII. Lawrence Erlbaum, Hillsdale, NJ, 1978.

15. PIAZZA, D. M. The influence of sex and handedness in the hemispheric specialisation of verbal and non-verbal tasks. Neuropsychologia 18, 163-176, 1980.

16. PIERSON, J. M., BRADSHAW, J. L. and NETTLETON, N. C. Head and body space to the left and right front and rear-l. Unidirectional competitive auditory stimulation. Neuropsychologia 21, 463473, 1983.

17. PORAC, C. and COREN, S. Lateral Preferences and Human Behauiour. Springer, New York, 1981. 18. SEARLEMAN, A. Subject variables and cerebral organization for language. Cortex 16,239-254, 1980. 19. STRAUSS, E. Hand, foot, eye and ear preferences on a dichotic listening task. Cortex 22, 47-82, 1986. 20. WEXLER, B. E. and HALWES, T. Dichotic listening tests in studying brain-behavior relationships.

Neuropsychologia 23, 545-559, 1985. 21. WEXLER, B. E. and LIPMAN, A. J. Sex differences in change over time in perceptual asymmetry. Neuropsychologia

26,943-946, 1988. 22. WEXLER, G., SCHARWTZ, WARRENBURG, S., SERVIS, M. and TARLATZIS, I. Effects of emotion on perceptual

asymmetry: Interactions with personality. Neuropsychologia 24, 699-710, 1986.