Purchase this article with an account.
Xilin Zhang, Fang Fang; Cortical magnification factor and population receptive field size in human V1 predict the bottom-up saliency map. Journal of Vision 2014;14(10):537. doi: https://doi.org/10.1167/14.10.537.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
In this study, we explored the relationship between the bottom-up saliency map and functional properties of human V1. For twenty-six subjects, we first performed a psychophysical experiment to measure their bottom-up saliency map with the method developed by Zhang and colleague (Neuron, 2012). The method used the Posner cueing paradigm to measure the attentional attraction (i.e. the saliency strength) of a foreground region in the stimulus that was backward masked and was invisible to subjects. Then, we performed fMRI experiments to estimate cortical magnification factors (CMF) and population receptive fields (pRF) in the subjects' visual cortical areas responding to the foreground region with the method proposed by Dumoulin and Wandell (NeuroImage, 2008). Along the visual hierarchy (V1, V2, V3, V4 and IPS), the CMF decreased and the pRF size increased. Significantly, across individual subjects, the saliency strength of the foreground region positively correlated with the V1 CMF, but negatively with the V1 pRF size. No significant correlation was found in other cortical areas. We speculate that the higher saliency may result from reduced lateral interactions between V1 neurons due to smaller pRFs and larger CMFs, and the intrinsic properties of V1 may provide critical constrains for generating the bottom-up saliency map.
Meeting abstract presented at VSS 2014
This PDF is available to Subscribers Only