Abstract
The validity of using the macaque brain as a model of the human brain assumes that the two brains share relevant functional characteristics. One challenge is that the methods used in each species can have substantial differences in spatial and temporal resolution, precluding a direct comparison. To bridge the gaps imposed by this limitation, several groups have collected data in both species using the same technique (fMRI) and similar experimental paradigms. This has allowed researchers to establish qualitative similarities in brain organization showing that the topographic maps for early visual regions (V1, V2, V3, and MT) are consistent with results from invasive techniques used in monkeys (e.g., tract tracing and micro-electrode recording). However, the extent to which "higher-order" cortical regions (e.g., inferior temporal cortex, prefrontal cortex) are similar has been more difficult to determine. Here, we describe an across species fMRI study that combines multiple experimental paradigms (pRF mapping, category localizers, movie watching, and hyperalignment) to better understand similarities and differences in the organization of inferior temporal cortex. PRF-mapping allowed us to identify visual cortical areas such as V1. The category localizer tasks allowed us to identify regions sensitive to faces, bodies, objects, scenes, color, and word-forms. Hyperalignment allowed us to abstract beyond idiosyncratic voxel patterns to identify representational structure shared across participants. Congruent with previous findings, individual categories of the localizer task were associated with multiple regions of cortex. Comparing their responses to rich naturalistic stimuli, we were able to better understand the computations performed by these regions by exploring similarities and differences in their rich representational spaces; both within and across species.