Abstract
Introduction. The detectability of transitions between dichoptic correlation states is highly anisotropic, with the detection of uncorrelation (U) or anticorrelation (A) transitions from a correlation (C) state being one of the most rapid perceptual events, while the detection of dichoptic correlation from uncorrelated or anticorrelated states requires up to several seconds of observation (Tyler & Julesz, 1979). Functional MRI was used to explore the dynamics of these neurontropy processes, together with binocular/monocular (B/M) transitions, throughout the human occipital lobe.
Methods. FMRI response were measured with a GE Signa 3T scanner, 3 s TR and block design to dynamic noise stimuli switching among stereoscopic structure, dichoptic correlated, dichoptic uncorrelated and purely monocular fields.
Results. In retinotopic cortex, presentation of disparity structure activated retinotopically appropriate regions in V1, V2, V3, V3A, ventral area VMO, foveal V7 and the lateral depth structure area ODS/KO, but not for hV4, hMT+ or other lateral cortical areas. Most of the activated areas showed a continuous signal to the depth structure, but had a different dynamic for the binocular/monocular alternation, responding equally to the B[[gt]]M and M[[gt]]B transitions. For the dichoptic correlation conditions there was a strong transition anisotropy, with most areas responding with a delayed suppression to the C[[gt]]U transition and no response to the U[[gt]]C transitions. Foveal V7, however, showed a rapid positive response to the C[[gt]]U transition only. There was minimal response in hMT+ to changes between any of these stimulus types.
Conclusion. FMRI can reveal striking differences in the cortical response dynamics, both across cortical regions and among different types of dichoptic stimulation. These dynamics suggest that the C[[gt]]U / U[[gt]]C transition anisotropy is a result of processing in the higher dorsal retinotopic areas.
Tyler, C.W., Julesz, B. (1976). The neural transfer characteristic (neurontropy) for binocular stochastic stimulation. Biol. Cybernetics 23, 33-37.
Supported by the Pacific Vision Foundation.