Purchase this article with an account.
Sangita Dandekar, Justin M. Ales, Thom Carney, Stanley A. Klein; Cortical folding as a sparseness criterion for identifying vep sources. Journal of Vision 2005;5(8):893. doi: https://doi.org/10.1167/5.8.893.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Background. Topographic mapping of the multifocal VEP can reveal temporal processing and interactions between closely spaced sources in early visual areas, such as V1 and V2. To identify closely spaced sources (the inverse problem) a sparseness condition is needed. The present study asks whether sparseness of cortical folding at certain time points might be exhibited in the changing topographic maps in multielectrode, multifocal VEP data.
Methods. The VEP stimuli consisted of a dartboard pattern containing 96 checkerboard patches. The patches were arranged in 4 rings of 24 patches each. Each patch was modulated according to a binary m-sequence and was cortically scaled to activate about 30 mm2 of primary cortex. Each subject viewed the stimulus wearing a cap with 64 or 96 recording electrodes. The checkerboard reversal response for each patch at each electrode was calculated by cross-correlation (1st cut of 2nd order kernel) and topographic maps of scalp activation for each stimulus patch were evaluated. When the area of activation moves around a cortical fold in response to a neighboring stimulus patch the scalp topography is expected to change dramatically. The number of distinct response topographies was estimated for each time period of strong activation.
Results. We found that in the densely sampled hemifields (48 patches per hemifield) there were between 5 and 10 distinct cortical maps indicative of sites of activation having moved around a cortical fold or a dorsal/ventral jump. At some time points the distinct map count was much smaller as expected when the response of one cortical area passes through a zero point and is effectively silenced.
Discussion. We expect the reduction of distinct folds at particular time points provides the sparseness condition needed to resolve the rotation ambiguity that has stymied previous attempts to disambiguate closely spaced sources in early visual areas.
This PDF is available to Subscribers Only