Facial recognition: hemispheric asymmetry, processing strategy, importance of individual features and the influence of changing memory intervals
Date
1977
Authors
Van Mastrigt, Robert Leonardes
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Three studies were designed to investigate the relationship between hemispheric asymmetries and differential processing strategies in facial recognition, the role of individual facial features and the influence of variable retention intervals on the emergence of hemispheric asymmetries and processing strategies. The right hemisphere is thought to process visual information as a global unit, while the left hemisphere processes the same information in a sequential or serial fashion.
A total of 45 volunteer male subjects were pretested for right-handedness, equal and normal short distance visual acuity, and right ear superiority on a verbal dichotic listening test. Identi-Kit face pairs, composed of up to six features, were presented under visual half field tachistoscopic conditions in a forced-choice (same/ different) recognition paradigm. The dependent variables were percent recognition accuracy and recognition speed, defined as the reaction time in milliseconds from the onset of a comparison face to the initiation of a response. Hemisphere superiority was inferred from visual field superiority.
In Study I, 15 subjects were presented face pairs equally divided between right and left visual fields, same and different presentations and a variable number of features (4, 5 or 6). The prediction that reaction times would increase with the number of features presented to the right visual field but not to the left visual field was not upheld. There was guarded support for a global processing strategy in both hemispheres. There were no accuracy differences.
In Study II. 15 subjects were exposed to face pairs equally divided between the visual fields. same and different presentations and a variable number of differences between faces (1, 3 or 6). The prediction that reaction times decrease as the number of face differences increase for right visual field compared to left visual field presentations was not upheld. Reaction times decreased with increasing face differences in both visual fields indicating an overall serial process. The left visual field reaction time superiority, when faces differed on 3 or 6 features, changed to a right visual field superiority, when faces differed on only one feature. Implications for a dual processing model were discussed. Faces were more accurately recognized when presented to the left visual field compared to the right visual field. In addition, the prediction that faces would be better and faster recognized on the basis of the eyes, rather than the nose or mouth, was confirmed.
In Study III, 15 subjects were presented face pairs equally divided between the two visual half-fields, same and different presentations and three interstimulus retention intervals (50, 100 or 500 milliseconds). The prediction that hemispheric asymmetries emerge only after the brief visual trace of the first face has deteriorated (200-250 milliseconds) was confirmed. A left visual field superiority was evident with retention intervals of 500 milliseconds, but not with 50 or 100 milliseconds. There was evidence supporting an overall global processing strategy, whether the brief visual trace was active or not. As in Study I, there were no accuracy differences.
The results were discussed in terms of a dual coding strategy in which a material-specific hemisphere specialization interacts with a processing-specific hemisphere specialization on a continuum of discrimination difficulty. The relationship between accuracy and speed of recognition was also discussed.