Interregional connections across early visual areas in contour processing
53.512, Tuesday, May 14, 8:30 am - 12:30 pm, Vista Ballroom
Cheng Qiu1, Philip Burton2, David Do3, Daniel Kersten1, Cheryl Olman1; 1Psychology, University of Minnesota, 2Center for Magnetic Resonance Research, University of Minnesota, 3School of Medicine, Johns Hopkins University
It has been shown that many early visual areas are involved in collinear contour processing. However, the interregional coordination that yields coherent structural percepts has not been fully studied, nor is it clear how background clutter may influence these early visual processes. To answer these questions, we used fMRI to measure activity in early visual cortex while subjects (N = 8) performed a contour detection task. In half of trials, a ring of Gabor elements at 2 degrees eccentricity were aligned to create a perceived contour, and on half they were randomly oriented. Presence or absence of background clutter was also manipulated, with clutter absent on half the trials and present on half. Six regions of interest (two ROIs, containing cortex representing either target Gabors or background Gabors, in each of V1, V2 and V3) were predefined using separate target versus background localizer scans. The first analysis used a GLM to estimate the magnitude of the response for each condition in each ROI. When background clutter was absent, responses in V2 target ROIs were slightly suppressed by aligned contours compared with unaligned, while in the presence of background clutter, responses were significantly stronger to aligned contours than unaligned. The effect of clutter on V2 target aligned/unaligned contrast was significant. In addition, a functional connectivity analysis (psychophysiological interactions or PPI) was run using V2 target ROIs as seeds. In the case with background clutter, there was an increase in V1-V2 task-dependent correlations when perceiving aligned versus when perceiving unaligned contours. Without background clutter, however, V2 showed no interaction with V1. Both the average response magnitude and the PPI analysis suggest different mechanisms may support contour processing without or with background distracters. Additionally, V2 may play a major role in coherent structure perception, especially with complex scene organization.