Visual imagery of familiar people and places in category selective cortex

Visual mental imagery is a dynamic process that involves a network of multiple brain regions. We used an electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) fusion approach to ask how the neural dynamics of category selective mental imagery in EEG related to activity within medial parietal, ventral temporal, and primary visual cortex (V1). Subjects attended separate EEG and fMRI sessions where they were asked to recall personally familiar people and place stimuli. The fMRI contrast of places versus people successfully replicated previous findings of category-selectivity in the medial parietal cortex during visual recall (Silson et al., 2019), as well as other regions including the ventral and lateral place memory areas, the fusiform face area, and frontal eye fields. Using multivariate decoding analysis in pre-defined fMRI regions of interest (ROIs), we tested the hypothesis that we would be able to decode individual stimuli within the preferred category for each region. This was largely the case in the ventral temporal ROIs, but a more complex pattern emerged in the medial parietal cortex; these regions represented information during imagery that was not restricted to their preferred category. EEG-fMRI fusion indicated that the timing of both medial parietal and ventral temporal involvement peaked early on during recall but did not clearly differ from each other. However, place-selective regions generally peaked earlier than people-selective regions, suggesting that representations of place stimuli evolved more quickly for these subjects. This mirrors the results from EEG stimuli decoding, where individual places were decodable earlier in time than people. In contrast, fusion correlations in V1 occurred later during the recall period, possibly reflecting the top-down progression of mental imagery from category-selective regions to primary visual cortex.