Running head: LEFT-RIGHT ASYMMETRY: AN EXAMINATION OF FACTORS


 

Left-right Asymmetry:An Examination of Factors That May Influence Leftward Bias in the Viewing of Emotive Faces

H. G. McFadden

Northeastern Illinois University


Abstract

Due to the lateralization of face processing functions within the human brain,right-handers show a marked hyperattentiveness to facial information observed within the left visual field.This study examined the possible influence of other factors on this leftward bias through the manipulation of instruction detail prior to viewing chimeric faces.Half of the subjects received traditional instructions which emphasized an intuitive approach, and the other half received more detailed instructions in which an evaluative strategy emphasizing care and attention to particular facial characteristics was suggested.While well-established findings of a leftward bias in facial preference were replicated, analysis established no significant relationship between level of instruction detail and facial preference.


Left-right Asymmetry:An Examination of Factors That May Influence Leftward Bias in the Viewing of Emotive Faces

Cerebral hemispheric lateralization of function is a very well researched area of inquiry. Evidence suggests that there are function-specific brain mechanisms which govern tasks such as facial recognition, interpretation of emotion, and language skills. These phenomena have been examined in regard to many brain structures and within a wide variety of contexts.It is the hemispheric lateralization and mechanistic specialization of functions relating to faces and emotive facial expression as reflected by perceptual bias that this paper will examine.

In general, a great body of previous research has demonstrated that the right hemisphere of the brain is dominant in tasks involving face processing and the evaluation of emotional expression, while the left hemisphere shows superiority in functions involving language and object component processing.Beyond this broad cloth, however,there remain questions.We do not entirely understand the role played by specific neuroanatomical structures in processing faces and emotional facial expression, or how they relate to the cognitive mechanisms which underlie these processes and which must be distributed over numerous anatomical structures.Observation of populations that display atypical hemilateral specialization patterns may contribute to our understanding of these issues.Individuals with Williams Syndrome ( WMS ), for example, show very different patterns of brain activity from normal persons when viewing faces ( Mills, et al., 2000 ).In their 1998 study, Mandal and Asthana showed that patients with damage to only the left hemisphere of the brain demonstrated marked superiority over right hemisphere damaged patients in matching performance for line drawings of faces.Schizophrenic persons also show significant differences in perceptual bias patterns when viewing faces in comparison to nonschizophrenic controls( Phillips & David, 1997 ).Another group in which differences from the general population may be observed, but who lack the structural brain damage of the above examples, is nonimpaired left-handed adults.

There are differences in the degree of hemispheric lateralization of brain function between dextrals and sinistrals. Previous research has demonstrated that when viewing a pair of chimeric faces, right-handers show a significant bias toward evaluating the face with the smile to the viewer’s left as the happier of the two ( Levy, Heller, Banich, & Burton, 1983 ).While also manifesting this leftward bias, left-handers do so to a significantly lesser extent than do right-handers ( Levy, et al., 1983 ).Similar findings have been reported in regard to left hemispheric dominance of language functions.In work with bilingual students who had been forced to switch their preferred hand from left to right during childhood, Hoosain ( 1991 ) showed that those subjects who had experienced forced handedness switching performed significantly more similar to sinistrals than to dextrals on tasks which measured accuracy of word identification.This suggests that the dissimilarity of lateralization patterns between right and left-handers is of genetic rather than developmental origin.

While it is well established that the right cerebral hemisphere of right-handers is primarily responsible for evaluating faces, there is also evidence to suggest that the brain’s face processing system may have certain inherently bilateral facets.Parkin and Williamson ( 1987 ) demonstrated that the left-hemisphere may be dominant in the discrimination of individual facial features.Thus the question of holistic vs. componential processing mechanisms is raised, which may in turn bear on developmental facets of lateral specialization.Alvarez, Mills, and Neville ( as cited in Mills, et al.,2000 ), found that adults utilize different brain systems to mediate processing of upright and inverted faces.In comparison, children appear to utilize a similar mechanism when viewing both upright and inverted faces.Furthermore, “ … the mature pattern of right-greater-than-left asymmetry to upright faces is not evident until the late teens”( Alvarez & Neville, as cited in Mills, et al. ).The juxtaposition of this developmental neurologic event with the typical onset time of schizophrenia is intriguing.

It appears that the processes of facial recognition and the interpretation of emotional expression may be more broadly distributed and bilaterally interactive than we understand them to be.Such systems are inherently difficult to study, largely confined as we are to the observation of the emergent properties of these complex mechanisms.This study attempts to add a piece to the puzzle by purposefully stimulating such interactivity and observing its effect on leftward bias.It was hypothesized that by giving subjects instructions which directed them to pay careful and considered attention to individual facial characteristics ( and thereby stimulating activation of left hemisphere processes ), the typical leftward bias should be significantly reduced in comparison to subjects who received traditional, more intuitively focused instructions.

Method

Participants

Participants in this study were 27 undergraduate university students ( 18 female,9 male ) aged 21 to 39 years inclusive.Of the participant pool, only two were found to be left-handers.Due to this low number, the left-handers’ scores were excluded from analysis, leaving 25 subjects ( 17 female, 8 male ) aged 21 to 39 years inclusive.Subjects were enrolled in the undergraduate psychology course Statistics and Research Methods II at Northeastern Illinois University and participated voluntarily.Two groups of students participated: those enrolled in the morning section of the class, and those enrolled in the evening section of the class.No incentive was offered for participation.

Materials

A packet was given to each participant before viewing began.Each packet contained a subject data questionnaire, one of two instruction sets ( traditional ordetailed ), and a scoring sheet.The questionnaire provided space for each participant to report their age, gender, seating position ( the right side or the left side of the room ),whether they were right or left handed, and whether they attended the morning or the evening section of the class.This survey provided space for the participants to later rate the ease with which they were able to complete the experimental tasks on a continuous scale of 1-5, with 1 indicating the subject found the task “easy,” to 5 indicating the subject found the task “difficult.”Additionally, space was provided for subjects to briefly describe the strategy they employed to make their decisions.Both instruction sets revealed that the subjects would be presented with several pairs of funny-looking faces, each of which would have one face at the top and one at the bottom.Participants were further told that their task would be to decide which face looked happiest, and indicate their choice on the provided answer sheet. The traditional instruction set then concluded with the statement “This task is easiest of you go with your first impression, and don’t think too much about it.” The detailed instruction set instead concluded with “ Be sure to be very careful in making your decision.Look closely at the faces and examine every detail.Pay special attention to the eyes and mouth.”Packets containing traditional and detailed instructions were alternated in the distribution phase to allocate evenly to experimental conditions.The scoring sheet consisted of 36 numbered spaces arranged in two columns upon which the subjects indicated their responses.Chimeric faces constructed of two face halves ( one half displaying a positive emotional expression, and the other half depicting a neutral expression ) of the same poser were used to elicit face preference response.These were constructed by bisecting two photographs of the same person and combining the halves with disparate expressions, thus resulting in a pair of faces arranged vertically.The chimeric faces used were those devised for similar purposes by Jerre Levy at the University of Chicago ( Levy, et al., 1983 ).This set of 36 chimeric faces was presented to subjects via an overhead projection display system.Each overhead projector transparency delivered one chimeric face pair to the subjects.Participants viewed the chimeric faces while seated around the outside perimeter of a U-shaped table.The overhead projector was located within the perimeter of the table at the center vertex position and displayed the faces on a screen arrayed on a wall approximately eight feet from the enclosing table perimeter.

Procedure

Participants gathered at their regularly scheduled class time and instruction packets were distributed.The experimenter indicated a point at the base position of the U-shaped table around which the participants were arrayed, and informed the subjects that all those seated from one side of the indicated point to the end of that arm of the table would be considered to be on the right side of the room and should indicate such on their questionnaire.The remaining portion of the subjects ( approximately one-half ) were to be considered to be on the left side of the room.Participants provided the requested personal data on the subject survey, and were then directed to read the instructions contained in their packet.Subjects were at that time informed that there should be no discussion of their impressions of the experimental materials with their fellows.Presentation of the chimeric faces then began.The experimenter verbally indicated the number ( from 1-36 ascending ) of each transparency containing a chimeric face pair as it was placed onto the overhead projector for viewing by the subjects.Each chimeric face pair was viewed for approximately three seconds, after which it was removed and replaced by the next in the set.Subjects indicated on their individual response sheet which face ( the top face or the bottom face ) of each pair looked happiest, presumably utilizing the evaluative strategy indicated in the instruction set they received.After 18 transparencies had been viewed, the experimenter directed the subjects to briefly review the instruction set provided in their packet.The presentation of chimeric faces then resumed as before and continued until the full set of 36 had been evaluated by the participants.Subjects were then verbally directed to indicate on their respective data sheets the ease with which they completed the facial evaluations, using the aforementioned 1-5 scale.Packets were then returned to the researcher.This procedure was conducted in identical fashion with both the morning and evening sections of the class.

Results

To measure the expected leftward bias, a Laterality Quotient ( LQ ) was determined.The LQ for each subject was calculated by subtracting the number of preferred faces in which the positive emotional expression was present on the left side from the number of preferred faces in which it was on the right side, and dividing by the total number of chimeric face pairs viewed ( 36 ).The result is a number from –1.000 to 1.000 inclusive, with a negative number indicating a preference for faces with a smile on the left side.As anticipated, the mean returned LQ for all subjects ( -.276 ) demonstrated the expected leftward bias,t ( 24 ) = -3.312, p < .01.Although a trend in the expected direction was present as Figure 1 shows, analysis did not reveal instruction detail to have a significant effect upon LQ, t ( 23 ) = -.718, p > .05.

-------------------------------------------------------------------------------------------------------------------------------------------------

Insert Figure 1 here.

-------------------------------------------------------------------------------------------------------------------------------------------------

While a strong trend was exhibited ( p = .051 ), no significant difference was found between the LQ of male and female subjects, t ( 23 ) = 2.062, p > .05.Further analysis also indicated that no relation existed between age and LQ, r ( 25 ) = -.021, p > .05.Interestingly, a significant positive correlation was shown between LQ and the difficulty with which participants reported making their decisions, r ( 23 ) = .396, p = .05.As illustrated in Figure 2, as leftward bias decreased, subjects rated the task more difficult.

-------------------------------------------------------------------------------------------------------------------------------------------------

Insert Figure 2 here.

-------------------------------------------------------------------------------------------------------------------------------------------------

Discussion

While a trend in the anticipated direction was noted, the results of our study failed to demonstrate that instruction detail can influence leftward bias among right-handed persons.There are a number of possible explanations for this result.Firstly, while instructed to utilize an evaluative strategy consisting of care and attention to detail, subjects may still respond intuitively, which would skew results toward the normal leftward bias.Secondly, in order to preserve identicality of stimulus delivery, subjects who received detailed instructions could only observe the chimeric faces for three seconds, which may not have been long enough for them to adequately utilize the directed evaluative strategy.This may have led to an abandonment of the incomplete detailed approach when the stimulus was removed and a reversion to an intuitively derived response.

The significant correlation between decreasing leftward bias and increasing perception of task difficulty may be indicative of an increase in bilateral cognitive processing among those subjects.That is to say, those subjects who more fully utilized the evaluative strategy outlined in the detailed instructions ( and thereby increased bilateral processing ) found the task more arousing ( i.e. more difficult ).Although the conducted analyses do not provide adequate evidence to conclusively support such an assertion, the data is tantalizingly suggestive.

There are a number of ways in which this experiment might be partially redesigned in future research to more fully explore the various phenomena involved.There was, in actuality, very little difference in the instruction set delivered to both groups of subjects.Given that the independent variable was instruction difference, it would appear that in future trials the detailed instruction set could be much more strongly worded to stimulate greater strategy compliance among subjects without compromising internal validity.Observation time might also be moderately increased across both experimental conditions.Delivery of stimulus and response via an apparatus which could measure response time is recommended, as presumably a longer response time would indicate greater care and attention to detail.

Given the volume of research in this area, new combinations of previously utilized methodologies would appear to be in order.This could provide data for futuremeta-analysis which may help clarify more specifically how the lateral specialization of 

component processes of face recognition, visual hemifield attention, and the evaluation of emotional expression interact and how these factors may differ in left and right-handers.Moreover, if a set of conditions may be developed under which dextrals would reliably return responses similar to those of sinistrals rather than dextral controls, we might more readily understand the mechanistic differences of hemilateral specialization between right and left-handers.

References

Alvarez, T. D., Mills, D., Neville, H. J.,( 1999 ).Different neural mechanisms for upright and inverted face recognition: Electrophysiological evidence ( Technical Rep. No. 9905 ).La Jolla: University of California at San Diego, Center for Research in Language.

Hoosain, R.,( 1991 ).Cerebral lateralization of bilingual functions after handedness switch in childhood.Journal of Genetic Psychology, 152, 263-268.

Levy, J., Heller, W., Banich, M., & Burton, L.( 1983 ).Asymmetry of perception in free viewing of chimeric faces.Brain and Cognition, 2, 404-419.

Mandal, M. K., & Asthana, H. S.,( 1998 ).Right hemisphere damage impairs the ability to process emotional expressions of unusual faces.Behavior Modification, 22, 167-177.

Mills, D. L., Alvarez, T. D., St. George, M., Appelbaum, L. G., Bellugi, U., & Neville, H.( 2000 ).Electrophysiological studies of face processing in Williams Syndrome.Journal of Cognitive Neuroscience, 12 ( Suppl. 2000 ), 47-64.

Parkin, A. J., and Williamson, P.( 1987 ).Cerebral lateralization at different stages of facial processing.Cortex, 23, 99-110.

Phillips, M. L., & David, A. S.,( 1997 ).Viewing strategies for simple and chimeric faces:An investigation of perceptual bias in normal and schizophrenic patients using visual scan paths.Brain and Cognition, 35, 225-238.

Figure Captions

Figure 1.Mean Laterality Quotient for traditional instruction set ( n = 12 ), and detailed instruction set ( n = 13 ).

Figure 2.As subject perception of task difficulty increases, mean laterality quotient decreases ( n = 25 ).