Purchase this article with an account.
Patrick J. Bennett, Masayoshi Nagai, Allison B. Sekuler; Perceptual completion is not better within than across hemispheres. Journal of Vision 2004;4(8):718. doi: 10.1167/4.8.718.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Pillow and Rubin (2002) found that shape judgments of figures that could contain illusory contours were poorer when the illusory contours crossed the vertical meridian than when it crossed the horizontal meridian. These results were interpreted as meaning perceptual integration is more difficult across hemispheres than within a hemisphere. However, because no baseline measure of perceptual completion (PC) was included, it is unclear whether their results were due to completion per se. Here, we re-examine the issue, comparing results when completion was expected (illusory contour conditions) and when PC was not possible (fragmented control condition). In all experiments, we asked observers to make shape judgments and varied the orientation of local inducing elements until performance reached our threshold criterion (staircase method). Expt 1 included both illusory and fragmented control figures, and stimulus duration was approximately 140 ms, about 45 ms longer than that of Pillow and Rubin. We found no consistent hemisphere effect, and only a few participants showed evidence of PC. In Expt 2, we changed the instructions to emphasize the presence of illusory contours. More participants showed evidence of PC, but we still found no consistent hemisphere effect. In Expt 3, we eliminated the control condition, and reduced the stimulus duration to one closer to Pillow and Rubin's. However, we still found no consistent hemisphere effect. Finally, we reduced the stimulus diameter from 14 to 6 deg, and confirmed that PC could be measured with our procedure, but we still found no consistent hemisphere effect. Overall, unlike Pillow and Rubin, we find no evidence of a hemisphere effect, and no evidence for PC with the large stimuli like those used in the original Pillow and Rubin study. Thus, our results do not support the idea that PC-related integration processes are impaired across hemispheres.
This PDF is available to Subscribers Only