Contents |
Matching categorical object representations in inferotemporal cortex of man and monkey
N. KRIEGESKORTE(1), M. MUR(1), D. RUFF(1), R. KIANI(2), J. BODURKA(1), P. BANDETTINI(1) 1 Section Functional Imaging Methods, NIH/NIMH Lab. Brain & Cognition, Bethesda, MD; 2 HHMI, Dept. of Neurobiology and Behavior, Univ. of Washington, Seattle, WA
Introduction
Single-cell recordings have shown that primate inferotemporal (IT) neurons respond selectively to visual features occurring in natural images as parts of objects. Neuroimaging research has demonstrated that information on object membership in human conventional categories is present in focal activations as well as in widely distributed response patterns. Neuroimaging, however, has focused on analyzing category-average responses, begging the question if an inherent categorical structure falls out of the representations and, if so, what the inherent categories are. A recent single-cell recording study demonstrates that monkey-IT response patterns in fact cluster according to natural categories (Kiani R et al 2007, J Neurophysiol). Here we ask, for a range of human regions from early visual to anterior temporal, if representations are inherently categorical, and, if so, what the main categories are. In addition we relate these human representations to that found in monkey IT in the cited study.
Measurements
We measured the human IT (hIT) response pattern elicited by each of 96 photos of natural objects from the Kiani set with hires fMRI (3T, SENSE, voxels: 1.95x1.95x2 mm3) while subjects detected color changes occurring at fixation during image presentation (stimulus duration: 300 ms).
Results
The hIT and mIT dissimilarity matrices (96x96) match closely (see figure). IT response patterns cluster in natural categories, with the animate-inanimate distinction explaining most variance and faces forming a subcluster. Early visual responses show no such categorical clustering. Our results suggests that the extraction of information on membership in these behaviorally crucial categories constitutes a fundamental function of primate IT across species. The close match also provides some hope that data from single-cell recording and fMRI, for all their differences, may similarly reveal neuronal representations when subjected to massively multivariate analyses of response-pattern information.
