Title: A Web-Based Study of Cerebral Asymmetry for Perception of Emotion.
Author: Linda Rueckert; Northeastern Illinois University; L-Rueckert@neiu.edu
Web site: http://www.neiu.edu/~lruecker
Corresponding author: Dr. Linda Rueckert
Dept. of Psychology
Northeastern Illinois University
5500 N. St. Louis Ave.
Chicago, IL 60625
773-334-0769
Abstract:
Many studies have examined cerebral asymmetry via lateralized presentation of stimuli using specialized software. Because these studies typically require extensive control over variables such as stimulus exposure duration, size and location, etc., they are not easily transferred to the web. The present study replicated a free-vision lateralization task, the Levy chimeric faces task, via the internet, using html and Javascript. The results replicated those of the original paper and pencil study.
Summary:
Many studies have examined cerebral asymmetry via lateralized presentation of stimuli using computers running specialized software. Because these studies typically require a great deal of control over variables such as stimulus size and location, exposure duration, etc., they are not as easy to translate into web-based experiments as are simpler surveys.
The purpose of the present study was to replicate on the internet one of the simpler and well-studied lateralization tasks, the Levy chimeric faces task (Levy, Heller, Banich, & Burton,1983). In this task subjects view two chimeric faces (faces with one half smiling and the other half showing a neutral expression). The two faces are mirror-images of each other, and are aligned vertically on a single page. The subject is asked to judge which face, the top one or the bottom one, appears happier. Numerous studies have shown that right-handed subjects are significantly more likely to choose the chimera with the smile to their (the viewer’s) left. This is thought to be due to the role of the right cerebral hemisphere (RH) in the perception of emotion in faces. When asked to do the task, the subject’s RH is activated, which causes a bias of attention to the left side of space, thereby making the chimera with the smile to the left appear happier than the one with the smile to the right. (For an extensive discussion of the design and rationale behind the study, see Levy et al., 1983). Left-handers show a smaller leftward bias than right-handers. The significant leftward bias can be obtained even when stimuli are presented in free vision, without control over viewing distance or exposure duration. The results are highly reliable, and have been found whether the stimuli are presented on a page or via a slide projector. For these reasons, this lateralization task might be particularly well-suited to the internet.
Method
The Levy chimeric faces task was replicated in a web page using html and Javascript. The orignal 36 stimuli were scanned and saved as .gif files. The presentation program was roughly based on a program described by Baron and Siepmann (2000). After reading the Informed Consent and instructions, the stimuli were presented. A stimulus consisted of a new window that filled the entire screen. (Actually, depending on the subjects’ operating system and browser, there could be a title bar and / or menus on the screen. Ramifications of this will be discussed later.) Within the window the two chimeras were vertically aligned, centered, and sized to fill 70% of the height of the screen. Below the chimeras was the question, “Which face is happier?”. Below that were radio buttons for “top” and “bottom”. Below that was a button that said “next trial”. All of these were centered on the screen. The entire stimulus fit within the window, so no scroll bars were necessary.
Subjects were asked to indicate their choice for “top” or “bottom”. If they clicked on “next trial” without making a choice, an alert appeared and they could not advance to the next trial. When utilizing a T1 or cable connection, the entire stimulus usually appeared within 1 sec. However, when utilizing a modem it could take up to 1 minute for it to load. If the subject attempted to make a choice before the picture had fully loaded, an alert appeared.
After going through all 36 stimuli, a demographics form was presented asking questions about gender, handedness, age, history of neurological disorders, etc. In an attempt to eliminate multiple submissions, it also asked if the subject had ever participated in this, or a similar, experiment previously. Subjects could also optionally give their name and email address.
Upon clicking on the “submit” button, the data were submitted to a .cgi program on the NEIU server. This program put all data into a readable data file. If the subject gave their email address, a receipt was sent to that address. (Some subjects are motivated to get extra credit in psychology classes, and require a receipt confirming their participation). The following environmental variables were also automatically recorded: screen size, IP address, operating system, navigator name and version, date, start time (when the subject clicked on “start experiment”), and end time (when the subject clicked on “submit”).
After submitting the data, the subject was told their number of smile-right choices, smile-left choices, and laterality quotient (LQ) ((right-left) / (right+left)). A negative LQ indicates that the subject chose the face with the smile to their left more often. Subjects were then linked to a debriefing page that explained the purpose of the experiment.
Results
After 2 months of data collection there were 164 total responses. Twenty-one of these were obvious multiple submissions (two or more cases with the same IP address submitted within a few minutes of each other), and were deleted. Also deleted were subjects who reported they had participated before, did not have 20/20 vision, had a neurological disorder, or left the handedness question blank. This resulted in 103 usable subjects (90 right handed, 13 left handed).
For right-handers, the mean LQ was -.428, which is significantly less than 0, indicating the expected leftward bias. The mean for left-handers (-.393) was also significantly less than 0. These scores did not differ significantly from those found by Levy et al. (1983) using the paper and pencil test (their mean LQ was -.303 for right-handers, -.134 for left-handers). LQ was not affected by screen size, browser, or operating system.
Discussion
These results suggest that at least some types of lateralization tasks can be effectively used in internet research. The lack of difference between operating systems in results is particularly striking, given the fact that it is only possible to achieve a perfectly symmetric stimulus using Internet Explorer for Windows (the menus are constantly visible in the upper left corner of the screen on MacIntosh computers). Although our results did not differ significantly from the original paper and pencil study, there was a trend toward greater leftward bias in the internet study, suggesting that further investigation may be necessary.
References
Baron, J., & Siepmann, M. (2000). Techniques for creating and using web questionnaires in research and teaching. In M. H. Birnbaum (Ed.) Psychological experiments on the internet. New York: Academic Press, pp. 235-263.
Levy, J., Heller, W., Banich, M. & Burton, L. (1983). Asymmetry of perception in free viewing of chimeric faces. Brain and Cognition, 2, 404-419.