Gender differences in a dichotic listening and movement task: lateralization or strategy?

Neuropsychologia 39 (2001) 25 – 35 www.elsevier.com/locate/neuropsychologia

Gender differences in a dichotic listening and movement task: lateralization or strategy? Timothy N. Welsh *, Digby Elliott Department of Kinesiology, McMaster Uni6ersity, 1280 Main Street West, Hamilton, Ont., Canada L8S 4K1 Received 1 February 2000; received in revised form 14 July 2000; accepted 17 July 2000

Abstract Although the dichotic listening procedure has been used as a non-invasive neuropsychological technique for assessing laterality of speech perception, it has tended to underestimate the proportion of the right-handed population that is left-hemisphere lateralized for speech perception [Segalowitz SJ, Bryden MP. Individual differences in hemispheric representation of language. In: Segalowitz SJ (Ed.), Language Functions and Brain Organization. Toronto: Academic Press, 1983, pp. 341 – 72]. These underestimations may be due to traditional dichotic procedures being susceptible to attentional biases, order of report effects, and/or memory effects that obscure functional differences between the cerebral hemispheres. In the present study, we used an adaptation of the dichotic listening procedure that was designed to be less sensitive to these confounding effects. Participants were required to move as quickly as possible to one of the two color-coded targets following verbal cues presented via headphones. Conditions of cue-word presentation were monaural, (e.g. ‘blue’ in one ear and a blank track in the other), dichotic-same (e.g. ‘blue’ in both ears), and dichotic-different (e.g. ‘green’ in one ear and ‘blue’ in the other). Ninety-three percent (26 of 28) of the participants demonstrated a right ear advantage (REA) for correct responses. There was also a REA for reaction time, movement time, and the total response time. The pattern of reaction time and movement errors, however, suggest that gender differences found utilizing this dichotic procedure may be due to differences in strategic approach to the task rather than to differences in cerebral laterality. Overall, results suggest that this new adaptation of the dichotic listening procedure is very sensitive to lateralization for speech perception. © 2000 Elsevier Science Ltd. All rights reserved. Keywords: Intrahemispheric versus interhemispheric communication; Dichotic; Neuropsychology

1. Introduction The dichotic listening procedure was first used to study attentional strategies and the ability of a ‘central processor’ to filter out unwanted information [1]. Following this work, Kimura [19] adapted the procedure in an attempt to develop a non-invasive neuropsychological technique for identifying the cerebral hemisphere specialized for language processes. The early procedures involved the simultaneous presentation of word pairs to the participant’s two ears, and the participant was then simply required to recall as many words as possible (free recall). It was found that right-handed people generally (85–89% of the population, see Bryden [4] for * Corresponding author. Tel.: + 1-905-5259140, ext. 23567; fax: +1-905-5236011. E-mail address: [email protected] (T.N. Welsh).

a review) exhibited a right ear-left hemisphere advantage (REA). That is, right-handers recalled more of the words presented to the right ear than to the left ear. The explanation of this finding was that, although there are both ipsilateral and contralateral neural pathways that connect the ears to the cerebral areas responsible for speech perception, during the performance of the dichotic listening task, the minor ipsilateral pathways are inhibited allowing the dominant contralateral pathways preferred access to the areas specialized for speech perception [20]. Thus, because the right ear has ‘direct’ access to the hemisphere best able to perceive and decode speech sounds, people typically perform better with their right ear than with their left ear. Recent brain imaging research employing positron emission tomography (PET) is consistent with the notion that the left cerebral hemisphere plays a greater role in the processing of speech sounds ([9,23]; see also [13]).

0028-3932/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 8 - 3 9 3 2 ( 0 0 ) 0 0 0 9 4 - 4

26

T.N. Welsh, D. Elliott / Neuropsychologia 39 (2001) 25–35

Although the free recall procedure was effective, it was not without its critics. The free recall aspect of the protocol was criticized for confounding functional differences between the hemispheres with attentional biases and short-term memory limitations (e.g. Kinsbourne [21]). Indeed, PET and fMRI studies have shown that directing attention to one ear increases the regional cerebral blood flow to the contrlateral auditory corticies ([25,32]; see also [14]) and, therefore, perhaps primes these areas for the reception and processing of the relevant information. Thus, since the first experiments, the dichotic listening procedure has been adapted to reduce the impact of these other influences. For example, in an attempt to decrease the possible effects of these right-side biases inherent in the free recall of information, Bryden [2] instructed participants to report what they heard in either the right or the left ear first on different blocks of trials. Although Bryden found a distinct REA, the REAs found that while controlling the order of report were smaller than those found while employing the free recall procedure. The dichotic listening procedure has also been criticized for lack of sensitivity. Specifically, although the estimates of the right-handed population that are left hemisphere specialized for speech perception are fairly accurate (85–89%) based on the results of the dichotic listening procedures, it has been pointed out that the tests underestimate the proportion of the population (95.5%) predicted from clinical studies [29]. Perhaps part of this discrepancy is because dichotic tasks are generally well performed, making differences between the two ears small [4]. Just as ear advantages on the dichotic listening task have been employed as an indication of cerebral specialization for speech perception, manual asymmetries on movement tasks have been used as an indication of contralateral hemispheric dominance for movement organization. Manual asymmetries on tasks such as rhythmic finger and limb tapping [31], transfer of training [30], and accuracy and timing of rapid aiming movements [8] have all been used as indices of cerebral specialization for movement organization and production. The results of these studies indicate that, for the right-handed individual, the left hemisphere plays a dominant role in movement organization. Although the left hemisphere has been suggested to have this executive role in movement organization, it must be remembered that it is the motor areas of the contralateral hemisphere that deliver the final movement directives to the distal musculature of the limbs. The purpose of the present study was to test a methodology that combines a dichotic listening and manual asymmetry paradigm. It was developed to be sensitive to cerebral specialization for both speech perception and movement organization.

A dichotic methodology involving a manual response was used by Jancke and Steinmetz [16]. They required their participants to monitor dichotic consonant–vowel pairs and instructed them to press a button with a cued hand when they perceived a target syllable in either ear. Eighty-nine percent of the right-handed and 63% of the left-handed participants demonstrated a right ear/left hemisphere advantage in reaction time. These laterality percentages are similar to estimates of cerebral specialization for speech perception utilizing traditional dichotic methodologies [4]. However, like many of the earlier studies employing the dichotic listening procedure, the laterality effects using Jancke and Steinmetz’s [16] procedure are still susceptible to attentional influences. Specifically, the participants may have demonstrated a right side advantage because they typically focus their attention to their dominant, right side [12] that may make them more sensitive to information presented to the right ear. The methodology developed for the present study avoids the pitfalls associated with attentional biases and lateralized limb control by combining the selective dichotic listening paradigm [11] with a rapid two alternative aiming task. Specifically, the participants were presented verbal target information, either monaurally or dichotically, and were required to focus their attention on the information presented to one of their ears and then make a rapid aiming movement to the target cued in that ear. By using factorial combinations of ear and hand, specific predictions were made about ear and hand advantages based on the time taken for withinand between-hemisphere communication. These predictions were based on the assumption that communication between centres within the same hemisphere is faster and more efficient than communication between the two hemispheres. For example, Fisk and Goodale [10] showed that right- and left-handed reaction times are shorter when the visual information is presented in the same visual field as the reacting hand. For most right-handed individuals, the left cerebral hemisphere plays a special role in both speech perception [29] and the organization and control of voluntary movement [31]. Thus, if the target information is presented dichotically, and the participants are instructed to pay attention to the right ear while responding with the right hand, they should enjoy a reaction time, and perhaps a movement time, advantage as compared with conditions involving other ear-hand pairings. If Kimura’s [20] hypothesis about the occlusion of the ipsilateral auditory pathways is correct, this temporal advantage would result because the processing required to complete the task is restricted to the left cerebral hemisphere (see Fig. 1B). Assuming that the left hemisphere must be involved in the perception of speech and the organization of accuracy-demanding limb movements, the left ear–left hand dichotic condition should

T.N. Welsh, D. Elliott / Neuropsychologia 39 (2001) 25–35

27

Fig. 1. Schematic of predicted paths of intra- and interhemispheric information transfer. The conditions are (A) left ear presentation, right hand movement; (B) right ear presentation, right hand movement; (C) left ear presentation, left hand movement; and, (D) right ear presentation, left hand movement. Note: SP, area for speech perception; MO, area for movement organization; RCH, area for right hand control; LHC, area for left hand control; solid lines, direct links between areas; dashed line, transmission through corpus callosum.

lead to the longest reaction times and movement times. This temporal disadvantage would result because this condition involves the greatest between-hemisphere transfer of information [16]. That is, the verbal information specifying the target, as well as information required for the regulation of the movement may have to cross and recross the corpus callosum (see Fig. 1C). The left ear–right hand and right ear – left hand pairings should be intermediate because movement organization based on this type of verbal input may involve just a single between-hemisphere transfer of information (see Fig. 1A and D). The above-given predictions are based on a further assumption that the participant is using the proper information to plan and complete the movement. If, as it is in the present task, the speed of the response is emphasized, the participant may program the movement based on the target information first available to the center for movement organization. Thus, because

the areas for speech perception and movement organization are specialized to the same hemisphere, due to the advantage of intrahemispheric communication the pathway in Fig. 1B may be used on the majority of the trials and, subsequently, an overall difference reflecting favoring the right ear may be found. The temporal advantages predicted in processing time might also generalize to movement error if one assumes that interhemispheric transfer can result in degradation or loss of information [18]. Thus, a combination of the temporal disadvantage and degradation of the information may result in more movements to the target cued in the right ear even though the left ear was the focus of attention for that trial. Also of interest was the examination of possible gender differences in this verbal–motor task. Although it is generally assumed that women are less lateralized than men for speech perception [29], the evidence is inconsistent. For example, Lake and Bryden [22] and

28

T.N. Welsh, D. Elliott / Neuropsychologia 39 (2001) 25–35

Piazza [26] found that women had less of a right ear advantage to verbal stimuli than men, while Bryden et al., [5], Carr [6], and Jancke and Steinmetz [16] found no differences between the genders. Differences in the results of these studies may be due to the variations of the types of tasks used. In support of this idea, Jancke et al. [17] found little relation between the laterality indices based on seven different dichotic tasks. Thus, ear advantages found in a particular task may be influenced by things other than just the ‘neural hardware’; that is, individual ‘software’ differences related to strategy may affect performance. Thus, as Segalowitz and Bryden [29] suggest, gender differences in laterality for speech perception may reflect differences in the way men and women approach the task. Munro and Govier [24] and Wexler and Lipman [33] had men and women perform a dichotic listening task for extended periods of time and examined how their laterality scores changed over time. In both the studies, the male participants showed a stronger REA at the beginning of the session than at the end, while the women showed the opposite pattern. Although laterality for speech perception has a structural basis (see Segalowitz and Bryden [29]), these changes over time suggest that some strategic differences or adaptations also have an influence on ear differences in performance. Thus, the purpose of this study was two-fold. First, we wanted to determine whether advantages due to intrahemispheric communication could be employed to develop a task that is sensitive to laterality for speech perception and movement organization. Second, we wanted to examine whether the suggested differences between men and women in lateralization for speech perception were due to processes other than just cerebral function.

2. Methods

2.1. Participants Participants were 28 members of the McMaster University community (14 men and 14 women). The mean age of the group of men was 25.3 years (ranging from 19 to 35) and 23.0 years (ranging from 19 to 29) for the group of females. For inclusion in the study, each participant was required to meet the following criteria, (1) report that they ‘always’ use their right hand to write with a pen, eat soup with a spoon, and throw a ball (modified Bryden Handedness Questionnaire [3]); (2) report normal, or corrected-to-normal, vision; and (3) report no known hearing impairments.

2.2. Apparatus and task Participants performed aiming movements over the surface of a black metal box (30 cm wide × 43 cm long × 4 cm high). Embedded in the surface of the box were three colored buttons. The starting location was a 1.5 cm diameter yellow button located in the center of the board. The two targets were one blue and one green translucent plastic buttons (1.5 cm diameter) that were located 16 cm on either side of the start button at the midline. Participants were cued to the target locations through specially created audio files of a male voice speaking the color words of the target buttons (i.e. ‘blue’ and ‘green’). The software used to create the audio files (Soundscape: SSHDR1 — Version 1:18) allowed us to align the attack of the words to within less than 0.25 ms and ensure that the two words were equal in volume. There were three conditions of stimuli presentation (two control and one experimental) that differed with respect to how information was presented through the headphones. The first control condition, termed the Single Ear (SE) condition, involved information presented to one ear only (e.g. ‘blue’ in left ear, blank track in right ear). The second control condition, termed the dichotic-same (DS) condition, involved the simultaneous presentation of the same word to both left and right ears (e.g. ‘blue’ in both the left and right ear). The third, and experimental, condition was the dichotic-different (DD) condition and involved the simultaneous presentation of different words to each ear (e.g. ‘blue’ in right ear and ‘green’ in left ear). Consistent with the selective dichotic listening procedure [11], participants were instructed to attend to one ear and then complete a movement as quickly as possible based on the information presented to that ear. They were further instructed to ignore any sounds associated with the other ear. All stimuli originated from an external computer (Pentium-Compupartner with Sound Blaster AWE64) and were presented to the subject via headphones (Koss Pro/466). The presentation of the stimuli was controlled by the experimenter and was randomly delayed 1–3 s from a ‘Ready’ cue. The external computer and starting and target buttons were interfaced with a Lafayette Interval Timer (Model 63520) such that a timer started upon stimulus presentation. This first timer was stopped when the participant’s hand was lifted from the starting position, thus measuring reaction time. When the subject lifted their hand from the start button, a second timer was started and subsequently stopped when either of the target buttons was depressed, thus measuring movement time.

T.N. Welsh, D. Elliott / Neuropsychologia 39 (2001) 25–35

29

2.3. Procedure

2.4. Dependent measures and data analysis

The entire procedure was completed in one session that lasted approximately 30 – 40 min. At the beginning of the session, participants were asked to report handedness and hearing and visual impairments. Following the screening process, participants were shown the movement environment and the procedure was explained. All the participants were kept naı¨ve as to the purpose of the testing. The study consisted of 12 blocks of 12 trials. All the 144 trials were blocked factorially based on reacting hand and attended ear. Condition of presentation was not blocked factorially. SE presentations were blocked separately from the two-dichotic presentation conditions while both DS and DD presentations were randomly mixed into the same blocks. Thus, there were four blocks of SE trials and eight blocks in which participants were presented information in both the ears. The DS and DD conditions were randomly mixed into the same blocks in order to ensure that participants were maintaining their attention on the information presented to the cued ear1. At the beginning of each block, participants were told which ear to focus their attention on, with which hand to move, and whether information was going to be presented to only one ear or to both the ears. Any given trial began when the participant depressed the start button. The experimenter then gave a warning cue of ‘Ready’. One to 3 s after the warning cue, the audio file for that trial was played and the participant was required to move as quickly as possible to the signaled target. The next trial began when the participant depressed the start button again. In order to ensure that any ear effects were due to lateralization and not due to a difference in the strength of the signal between the channels, the earphones were switched (i.e. right ear phone placed on left ear) on the second block of each dichotic block. In terms of errors, the participants were instructed to react and move as quickly as they could. They were told that all the other participants made errors so not to worry if they did. Participants were never told, at the end of each trial, if they moved to the correct or incorrect location. Finally, the location of the target was randomized within each block and the order of the blocks was randomly assigned to the participants.

Reaction time (RT) and movement time (MT) were recorded from the interval timers. Total response time (TT) was calculated by adding the two times together. The final dependent measure was the number of movement errors (ER). Simply defined, an ER occurred every time the participant completed the movement to the incorrect target. Because in the two control conditions (SE and DS) participants made errors on less than 1% of the trials, only the number of ER made in the DD condition were analyzed. A 2 Gender (women/ men)× 2 Ear (left/right) × 2 Hand (left/right) mixed ANOVA with repeated measures on the last two factors was used for the ER analysis. Mean RT, MT, and TT were submitted to a 2 Gender ×2 Ear × 2 Hand ×3 Condition (SE/DS/DD) mixed ANOVA with repeated measures on the last three factors. Subsequent to this analysis, the errorful trials were removed from the data set and the same analysis was performed on the medians of the correct trials2. Tukey’s HSD procedure (PB 0.05) was used to post hoc any effects involving more than two means.

1

Pilot testing revealed that when the DD files were blocked together, the participants would tend to adopt a strategy in which, regardless of the ear they were instructed to attend to, they would always concentrate on the right ear. Thus, when instructed to attend to the left ear, they would concentrate on their right ear and then move to the opposite target. This allowed them to decrease error rates to nearly zero by ignoring the instructions. Thus, in order to ensure that the subjects were truly paying attention to the target ear, DS files were randomly mixed with the DD files.

3. Results

3.1. Intrusion errors As predicted, there was a main effect for Ear, F (1, 26)=40.8, PB 0.0001, with participants committing significantly fewer errors when they were concentrating on their right ear (M= 2.2) than when they concentrated on their left ear (M= 5.5). Interestingly, there was also a three-way interaction between Gender, Ear, and Hand, F (1, 26)= 8.08, PB 0.01. Post hoc analysis of this interaction revealed that, while women made the same number of errors in the left ear–right hand condition (M= 4.3) and as in the right ear–right hand condition (M=2.7), all other between-ear comparisons showed a REA (see Table 1 and Fig. 2).

3.2. Reaction time For the RT analysis involving all the trials, the only significant main effect was for Condition, F (2, 52)= 63.18, PB 0.0001; RTs to SE presentation (M =315 ms) were shorter than those to the DS presentation 2

Due to the large number of ERs some participants committed in certain conditions, the means in these cases would have represented the average across only a few trials. Thus, in order to decrease the effects of possible outliers on the means used in this analysis, the initial and reported analysis was performed on the medians. However, a secondary analysis was performed on the means and the pattern of results was identical.

T.N. Welsh, D. Elliott / Neuropsychologia 39 (2001) 25–35

30

(M =405 ms), which were in turn shorter than those to the DD presentation (M =503 ms). While the effect for Ear, F(1, 26) =3.04, PB 0.10, approached traditional levels of significance, Gender, F(1, 26) B 1, and Hand, F(1, 26) B1, effects were absent. There was a significant interaction between Gender, Ear, and Condition, F (2, 52)=4.18, PB0.05. As evident from Fig. 3, while the men demonstrated a REA in the DD condition only, the women displayed no ear advantage in any condition. The analysis of RT on the correct-only trials revealed that the REA was enhanced when the errorful trials were removed. In this analysis, along with the main effect for Condition, F(2, 52)=77.06, P B 0.0001, there was also a main effect for Ear, F(1, 26) = 4.55, PB 0.05. Post hoc analysis of the Ear×Condition interac-

tion, F(2, 52) = 3.62, P B 0.05, again revealed there was a REA present only in the DD condition. Finally, it is important to note that, with the errorful trials removed, the interaction between Gender, Ear, and Condition disappeared, F(2, 52)= 2.21, P= 0.12 (see Table 2). This interaction disappeared because of a decrease in the female participant’s average RT in the error-free right ear attention condition (500 ms) from the trials in which the errors are included in the data set (528 ms), while the RTs in the left ear attention condition had a smaller decrease (512 from 522 ms).

3.3. Mo6ement time The MT analysis involving all trials revealed a main effect for Gender, F (1, 26)= 9.20, PB 0.01, and Con-

Table 1 Mean reaction time (ms), movement time (ms), total response time (ms) and number of movement errors across all the trials as a function of Gender, Ear, Hand, and Condition Gender

Condition

Left Ear

Right Ear

Left Hand

Right Hand

Left Hand

Right Hand

Single Ear Dichotic same Dichotic different

307 414 518

295 416 499

297 395 449

322 407 463

Single Ear Dichotic same Dichotic different

326 401 525

326 413 520

325 394 527

323 401 529

Single Ear Dichotic same Dichotic different

159 184 246

152 181 261

169 183 223

155 169 262

Single Ear Dichotic same Dichotic different

223 238 329

208 249 350

229 250 293

207 236 279

Single Ear Dichotic same Dichotic different

465 598 763

447 595 760

466 579 671

475 577 726

Single Ear Dichotic same Dichotic different

543 638 854

534 662 870

553 643 820

528 638 807

Reaction time Males

Females

Mo6ement time Males

Females

Total response time Males

Females

Mo6ement errors Males Single Ear Dichotic same Dichotic different

0 0.1 5.8

0.2 0.2 6.8

0.1 0.1 2.4

0.1 0.1 1.7

Single Ear Dichotic same Dichotic different

0 0.1 5.3

0 0 4.3

0 0.1 2.1

0 0 2.7

Females

T.N. Welsh, D. Elliott / Neuropsychologia 39 (2001) 25–35

31

only was strengthened when the errorful trials were removed, F(2, 52) = 11.55, PB 0.0001. Notably, the increase in the REA in the DD condition was mainly due to a decrease in the MT when errorful trials were removed when the participants were concentrating on their right ear.

3.4. Total response time The analysis of TT revealed main effects for Ear, F(1, 26)= 4.37, PB0.05, and Condition, F(2, 52)= 96.64, PB 0.0001, which were superceded by a significant Ear × Condition interaction, F(2, 52) =4.03, PB 0.05. Post hoc analysis of the interaction revealed, as in RT and MT, a REA present in the DD condition only. Again, as in both RT and MT, the above effects were strengthened or equivalent when the errorful trials were removed: ME for Ear, F(1, 26)= 9.28, PB0.01; ME for Condition, F(2, 52) = 94.49, PB 0.0001; and, Ear× Condition, F(2, 52)= 12.23, PB 0.0001. Again, the interaction was strengthened due to a decrease in the TT when the participants were required to concentrate on their right ear.

Fig. 2. Movement errors as a function of Gender, Ear, and Hand. Solid bars represent movements made with the Left Hand and open bars represent movements made with the Right Hand.

dition, F (2, 52)=37.17, P B0.0001. It was found that the men had shorter MTs (M =195 ms) than the women (M=257 ms), and that MTs in the SE (M= 187 ms) and DS (M =211 ms) conditions were shorter than those in the DD condition (M = 280 ms). While there was no main effect for Ear, F (1, 26) = 3.84, P B0.06, or Hand, F (1, 26) B 1.0, there were significant two-way interactions between Ear and Condition, F (2, 52)=3.27, PB 0.05, and Hand and Condition, F (2, 52)=3.83, P B 0.05. As in RT, post hoc analysis of the Ear ×Condition interaction revealed that, in the DD condition only, participants moved fastest when focusing on the right ear. Analysis of the Hand × Condition interaction revealed that participants took less time to complete right-handed movements in the SE condition than the DS condition. Both these conditions were faster than the DD condition. For left-handed movements, there was no reliable difference between the SE and DS conditions. As with the analysis of RT, when the errorful trials were removed from the data set, effect sizes generally increased. Specifically, while there were main effects for Gender, F(1, 26)=9.85, P B 0.005, and Condition, F(2, 52)=24.57, PB 0.0001, there was also a main effect for Ear, F(1, 26)=10.35, P B0.005. Finally, while the Hand×Condition interaction for MT disappeared, F(2, 52)B1, the REA for MT in the DD condition

Fig. 3. Reaction time (ms) across all the trials as a function of Gender, Ear, and Condition. Solid bars represent movements in which the Left Ear was the ear of focus and the open bars represent movements in which the Right Ear was the ear of focus.

T.N. Welsh, D. Elliott / Neuropsychologia 39 (2001) 25–35

32

Table 2 Median reaction time (ms), movement time (ms) and total response time (ms) of the correct-only trials as a function of Gender, Ear, Hand, and Condition Gender

Condition

Left Ear

Right Ear

Left Hand

Right Hand

Left Hand

Right Hand

Single Ear Dichotic same Dichotic different

297 410 512

296 411 486

294 394 444

312 406 447

Single Ear Dichotic same Dichotic different

325 392 513

320 415 510

322 386 509

318 403 492

Single Ear Dichotic same Dichotic different

143 158 254

131 161 275

150 162 189

144 154 189

Single Ear Dichotic same Dichotic different

212 220 301

197 231 319

216 231 265

201 218 250

Single Ear Dichotic same Dichotic different

455 573 784

427 580 787

456 555 650

468 566 666

Single Ear Dichotic same Dichotic different

543 616 833

519 644 847

538 631 786

523 629 738

Reaction time Males

Females

Mo6ement time Males

Females

Total response time Males

Females

3.5. Indi6idual success rates Success rates were calculated by taking the number of correct responses made per ear as a proportion of the total number of opportunities (24). Overall, participants were 27.5% more successful when concentrating on their right ear (M= 81.4%) than on their left ear (M = 53.9%). However, during this analysis, it was recognized that one of the women demonstrated no lateral advantage and another woman was actually more successful while attending to her left ear (79.17%) than her right ear (50%). With these two participants removed, the remaining 26 participants displayed an average of 31.4% difference in performance between the two ears (left ear −51.3%, right ear − 82.7%).

4. Discussion On the basis of clinical evidence, it has been suggested that 95.5% of the right-handed population is left hemisphere specialized for speech perception [29]. Thus far, however, most studies utilizing traditional dichotic listening paradigms have produced results that tend to underestimate that percentage. These underestimations

could be the result of methodologies that may allow things other than functional differences between the hemispheres to mediate laterality effects. As 26 of the 28 (93%) of the participants displayed an REA for correct responses, with only one participant demonstrating a LEA and one other demonstrating no ear advantage, it seems that this new adaptation of the dichotic listening procedure may be more sensitive to lateralization for speech perception than more traditional dichotic techniques. This procedure may be more effective than the traditional dichotic listening procedures at identifying hemispheric specialization for speech perception for a couple of reasons. First, by instructing the participant to focus attention on one ear, attentional biases are reduced. Second, because the responses are immediate, any memory and report-order biases are eliminated. This is not to say that strategies do not affect outcomes, only that their effects are minimized compared with the other dichotic listening procedures. To elucidate, when auditory signals are presented dichotically, the information presented to a particular ear is assumed to have a direct access to the contralateral hemisphere. In the majority of the right-handed population, that means that the information presented

T.N. Welsh, D. Elliott / Neuropsychologia 39 (2001) 25–35

to the right ear has ‘direct’ access to the areas specialized for the processing of speech sounds [20]. Thus, because the information presented to the right ear was received and decoded first, this information has primacy for response programming. This primacy, in consort with the emphasis placed on the speed of the response, more often than not, resulted in the right ear information being the basis for the programming of the response, whether the right ear was the target ear or not. Further, because the right ear information need not be transmitted via the corpus callosum, there is a greater chance that the left ear information may be lost or degraded before being received by the processing centers in the left hemisphere. Thus, the information presented to the right ear not only has a temporal advantage, but it may also be less degraded than the information presented to the left ear. Metaphorically speaking, if there was a race between the two pieces of information for activation, the information presented to the left ear not only has longer to run, but, as a result of this less direct route, also has a greater opportunity to get lost or tripped along the route, thus giving the right ear information a significant advantage. This notion is supported by the RT and TT data. Specifically, the REA for RT and TT in the DD condition when errors were removed was 32 and 103 ms, respectively. Functional differences between the two hemispheres, and thus temporal advantages, can be mediated, however, through the use of particular strategies. The pattern of gender differences in the present study demonstrates how differences in strategy can affect laterality results obtained in dichotic listening procedures. Initially, an examination of the error data and the RT collapsed across all the trials revealed that the men had a larger REA for both the dependent variables than the women. This could suggest that the women in the study were not as lateralized as the men. The interaction between Gender, Ear, and Condition present in the all-trial analysis for RT, however, disappeared when the data on which movement errors were made were removed from the analysis. Since errors are more likely to occur when conflicting target information is concurrently available to the areas responsible for speech perception, errors made while attending to the right ear should occur with greater frequency on trials in which the participant took longer to respond. Thus, to delay the response in an attempt to increase success, although beneficial in the left ear focus conditions, would actually hamper performance in the right ear condition. The finding that the differences between the genders disappeared in the analysis performed only on the correct RTs was due mainly to a decrease in RT during the right ear condition, with little change when the participants were instructed to attend to the left ear, suggests that the women were attempting to trade-off speed for accuracy.

33

More specific evidence of this strategy being utilized is in the performance of three participants, one man and two women. At the initial examination of their error data, these three individuals were not strongly lateralized with REAs ranging from 4.2 to 8.3%. These small advantages, however, occurred in conjunction with success rates of 80% or better for both ears. A deeper investigation of their individual performance revealed that the average difference between RTs and TTs on the correctly responded trials and RTs and TTs in which they committed an error was, 4 and 141 ms for M8; 70 and 122 ms for W1; and 83 and 51 ms for W10. Thus, as Rabbitt [27] first suggested, these participants were attempting to find and stay in their fast, but safe response time zone by trading off speed for accuracy. Another type of strategy was observed. It was found that four of the participants, two of each gender, fell into a response set in the DD condition when responding to the information presented to their left ear. Specifically, regardless of the target for that trial in the DD condition, these participants almost always went to the same target. For example, one of these individuals, in the left ear DD condition moved to the ‘green’ target 22 of the 24 trials (correctly to the ‘blue’ target only twice) while making no incorrect movements in the right ear DD condition. This pattern of results may be indicative of participant uncertainty when attending to the left ear and thus employing a ‘playing the odds’ type of strategy. While the strategy or ‘software’ explanation reconciles the gender difference, it may also be that neural or ‘hardware’ differences account for the results. Ringo [28] suggested that individual neurons, and thus processing centers, in smaller brains are more interconnected than in larger brains and, because of this greater inconnectedness, smaller brains tend to be less specialized than larger brains. The idea being that processing centers in the smaller brain enjoy a greater degree of direct access to each other and, thus, work more cooperatively than independently to complete the required processing. Jancke et al. [15], in their MRI study of 79 men and 41 women, found that the women, on average, had smaller forebrain volumes and larger corpus collosum to forebrain volume ratios than the men. Following Ringo’s [28] reasoning, they postulated that the notion that women have greater bilateral representation of processing centers (less lateralized) than men is more related to differences in brain size than a biologically predisposed difference between men and women. While no anatomical measurements were made in the present study, it could be argued that the female participants in the present study had slightly different ear advantages than the male participants because each hemisphere was better able to perform the individual parts of the task. However, if it was a decrease in laterality that resulted in the gender differences found in the present

34

T.N. Welsh, D. Elliott / Neuropsychologia 39 (2001) 25–35

study, then one would have expected the women to have better and more similar performances in the conditions in which all processing could have been completed within the same hemisphere. In other words, the women would have performed better in the left ear–left hand condition than the right ear – left hand condition or the left ear–right hand condition, which was not the case. Although all the predictions for ear advantages were realized, interactions between Ear and Hand in the DD condition were not present in either the error or temporal data. These predicted interactions involving the motor response system were not found, perhaps, because of the nature of the motor response. The low emphasis on accuracy [8] and a degree of uncertainty about the spatial location of the movement [7], may have washed out any advantages the right hand/left hemisphere may have enjoyed. In sum, the results of the present study suggest that this new adaptation of the dichotic listening paradigm is very sensitive to cerebral specialization for speech perception. One advantage of this new paradigm is that it reduces the impact of memory and attentional biases. The temporal measures of performance associated with the paradigm allow us to move from a nominal scale of measurement (i.e. right or wrong) to a more powerful ratio scale of measurement. By examining the error patterns and temporal patterns in tandem, we are better able to identify the performance strategies adopted by individual participants, and perhaps, groups of participants.

[5]

[6] [7]

[8]

[9]

[10]

[11] [12]

[13]

[14]

[15]

[16] [17]

Acknowledgements This research was co-funded by the National Down Syndrome Society (NDSS) and the National Institute of Child Health and Human Development (NICHD), NIH c 1 R01 HD37448-01. The authors would like to express their gratitude to Lorne Tulk for the software and technical assistance in creating the audio files, to John Moroz for developing the apparatus used in this study, and to Dr Dominic Simon for comments made on an earlier copy of the paper.

[18]

[19] [20] [21] [22] [23]

[24]

References [1] Broadbent DE. The role of auditory localization in attention and memory span. Journal of Experimental Psychology 1954;47: 191 – 6. [2] Bryden MP. Ear preference in auditory perception. Journal of Experimental Psychology 1963;65:103–5. [3] Bryden MP. Measuring handedness with questionnaires. Neuropsychologia 1977;15:617–24. [4] Bryden MP. An overview of the dichotic listening procedure and its relation to cerebral organization. In: Hugdahl K, editor.

[25]

[26]

[27]

Handbook of dichotic listening: Theory, methods and research. Toronto: Wiley, 1988:1 – 43. Bryden MP, Munhall K, Allard F. Attentional biases and the right-ear effect in dichotic listening. Brain and Language 1983;18:236 – 48. Carr BM. Ear effect variables and order of report in dichotic listening. Cortex 1969;5:63 – 8. Carson RG, Chua R, Goodman D, Byblow WD, Elliott D. The preparation of aiming movements. Brain and Cognition 1995;28:133 – 54. Elliott D, Chua R. Manual asymmetries in goal-directed movement. In: Elliott D, Roy EA, editors. Manual Asymmetries In Motor Performance. Boca Raton, FL: CRC Press, 1996:143–58. Fiez JA, Raichle ME, Balota DA, Tallal P, Petersen SE. PET activation of posterior temporal regions during auditory word presentation and verb generation. Cerebral Cortex 1996;6:1–10. Fisk JD, Goodale MA. The organization of eye and limb movements during unrestricted reaching to targets in contralateral and ipsilateral visual space. Experimental Brain Research 1985;60:159 – 78. Hiscock M, Kinsbourne M. Selective listening asymmetry in preschool children. Developmental Psychology 1977;13:217–24. Honda H. Rightward superiority of eye movements in a bimanual aiming task. Quarterly Journal of Experimental Psychology 1982;34A:499 – 513. Hugdahl K, Bronnick K, Kyllingabæk S, Law I, Gade A, Paulson O B. Brain activation during dichotic presentations of consonant-vowel and musical instrument stimuli: a 15O-PET study. Neuropsychologia 37:431 – 40. Jancke L, Mirzazade S, Shah NJ. Attention modulates activity in the primary and the seconday auditory cortex: a functional magnetic resonance imaging study in human subjects. Neuroscience Letters 1999;266:125 – 8. Jancke L, Staiger JF, Schlaug G, Huang Y, Steinmetz H. The relationship between corpus callosum size and forebrain volume. Cerebral Cortex 1997;7:48 – 56. Jancke L, Steinmetz H. Interhemispheric transfer and corpus callosum size. Neuroreport 1994;5:2385 – 8. Jancke L, Steinmetz H, Volkmann J. Dichotic listening: what does it measure? Neuropsychologia 1992;30:941 – 50. Jones R, Elliott D. Intra- and interhemispheric integration of tactual and visual spatial information. Bulletin of the Psychonomic Society 1988;26:229 – 31. Kimura D. Cerebral dominance and the perception of verbal stimuli. Canadian Journal of Psychology 1961;15:166 – 71. Kimura D. Functional asymmetry of the brain in dichotic listening. Cortex 1967;3:163 – 78. Kinsbourne M. The cerebral basis of lateral asymmetries in attention. Acta Psychologica 1970;33:193 – 201. Lake DA, Bryden MP. Handedness and sex differences in hemispheric asymmetry. Brain and Language 1976;3:266 – 82. Mazoyer B, Dehaene S, Tzourio N, Frak V, Cohen L, Murayama N, Levrier O, Salamon G, Mehler J. The cortical representation of speech. Journal of Cognitive Neuroscience 1993;5:467 – 79. Munro P, Govier E. Dynamic gender-related differences in dichotic listening performance. Neuropsychologia 1993;31:347–53. O’Leary DS, Anderson NC, Hurtig RR, Hichwa RD, Watkins GL, Boles Ponto LL, Rodgers M, Kirchner PT. A positron emission tomography study of binaurally and dichotically presented stimuli: effects of level of language and directed attention. Brain and Language 1996;53:20 – 39. Piazza D. The influence of sex and handedness in the hemispheric specialization of verbal and nonverbal tasks. Neuropsychologia 1980;18:163 – 76. Rabbitt P. How old and young subjects monitor and control responses for accuracy and speed. British Journal of Psychology 1979;70:305 – 11.

T.N. Welsh, D. Elliott / Neuropsychologia 39 (2001) 25–35 [28] Ringo JL. Neuronal interconnectedness as a function of brain size. Brain Behavior and Evolution 1991;38:1–6. [29] Segalowitz SJ, Bryden MP. Individual differences in hemispheric representation of language. In: Segalowitz SJ, editor. Language functions and brain organization. Toronto: Academic Press, 1983:341 – 72. [30] Taylor HG, Heilman KM. Left hemisphere motor dominance in right-handers. Cortex 1980;16:587–603.

.

35

[31] Todor JI, Kyprie PM, Price HL. Lateral asymmetries in arm, wrist, and finger movements. Cortex 1982;18:515 – 23. [32] Tzourio N, El Massioui F, Crivello F, Joliot M, Renault B, Mazoyer B. Functional anatomy of human auditory attention studied with PET. Neuroimage 1997;5:63 – 77. [33] Wexler BE, Lipman AJ. Sex differences in change over time in perceptual asymmetry. Neuropsychologia 1988;26:943– 6.