Robustness to two-object images in human visual cortex



Figure W10: Intracranial field potentials and receptive fields in human temporal cortex





We argue in the text that the main results (robustness to the presence of a second object in the image) are unlikely to be accounted for by a small IFP receptive field that is centered only on the preferred object / object category. While more detailed measurements of receptive field sizes would be welcome, it seems that most of the visually selective electrodes responded to objects presented at both positions (black squares in this scheme). We therefore represent the IFP receptive field by the gray circle. We emphasize that our study is a far cry from a detailed mapping of the IFP receptive fields. To our knowledge, there is no detailed electrophysiological mapping of IFP receptive fields in human temporal cortex in the literature.

We note in the text that the IFP receptive field could include the activity of many neurons with smaller receptive fields (dashed circles in this figure). Many studies have examined the receptive fields of neurons in the macaque inferior temporal cortex (and other parts of ventral visual cortex). The receptive field estimates in ITC vary significantly across studies ranging from ~2 degrees (e.g. DiCarlo&Maunsell 2003) to several tens of degrees (e.g. Rolls 1991). Assuming that IFP receptive fields are larger than single-neuron receptive fields (e.g. Yoshor&Maunsell 2007) and given the wide ranges in the macaque literature, our coarse estimates seem to be approximately consistent with the macaque neurophysiology studies.