Purchase this article with an account.
Michael Anes, Lindsey Short, Jennifer Storer; Hemispheric specialization for face processing revealed by use of Thatcherized and feature distorted faces. Journal of Vision 2008;8(6):900. doi: https://doi.org/10.1167/8.6.900.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Despite striking homogeneity in facial features and structure, humans effortlessly recognize faces across variable exposure conditions. Inversion makes face processing difficult; one explanation is that face perception involves encoding configural information (relationships among features) rather than features per se, and that configuration is unrecoverable from inverted faces. Functional imaging studies and observations of brain-damaged patients show the brain's right hemisphere is particularly sensitive to face configuration. We investigated right hemispheric sensitivity to face configuration using the Thatcher Effect, whereby faces with inverted eyes and mouths are perceived as grotesque when upright but not when inverted. One eye or one eye and the mouth were Thatcherized in faces cropped to ovals to remove hair and clothing features. Display duration was 120 ms. Participants rated bizarreness on each trial. We predicted higher ratings for left eye than right eye manipulations because of the contralateral projections of the visual fields to the brain, but only for upright and not inverted faces. This side by orientation interaction was robust and moderated by age; 8–10 year olds showed the smallest side difference in rated bizarreness for upright displays. Performance of 11–13 year olds was the same as college students. Stronger right-handedness was associated with greater rating bizarreness for upright left-sided than right-sided manipulations in the youngest group. Other results are forthcoming: We obtain bizarreness ratings after manipulating features (iris and teeth) on sides of the face. We present results from strongly left-handed participants. Also, we manipulate side and information type (configural, featural) independently in a same-different task with pairs of sequentially presented faces; information in the second face is manipulated for 20 ms at different times during the display. By examining lengthened correct same responses, this latter task can reveal relative time courses in utilizing featural and configural information by the hemispheres.
This PDF is available to Subscribers Only