Hands in motion: Characterization of upper-limb selective regions in the occipito-temporal cortex.
56.579, Tuesday, 20-May, 2:45 pm - 6:45 pm, Pavilion
Tanya Orlov1, Yuval Porat1, Tamar Makin3, Ehud Zohary1,2; 1Neurobiology Department, Life Sciences Institute, Hebrew University of Jerusalem, Jerusalem 91904, Israel, 2Interdisciplinary Center for Neural Computation and ELSC, Hebrew University of Jerusalem, Jerusalem 91904, Israel, 3FMRIB Centre, Nuffield Department of Clinical Neuroscience, University of Oxford, Oxford OX39DU, UK
We report a region (termed here extrastriate arm area, EAA), in the occipito-temporal cortex (OTC), which shows clear preference to static images of human upper-limbs when compared to other body-parts or to other object categories. Such selectivity was previously attributed to shape aspects, which presumably vary across image categories. However, functional selectivity for upper-limbs may also be driven by their unique motion features. Here we show that in EAA selectivity to static upper-limb images and motion kinematics go hand in hand. Using resting-state and task-based functional MRI, we demonstrate that OTC voxels' preference to different image categories can be predicted, to a significant extent, from their patterns of functional connectivity. We find that the degree of OTC voxels' selectivity to rigid, typically inert objects is positively associated with their strength of functional connectivity with mid-level shape-selective areas (i.e. V4 / LO-1). Thus, representations of these objects within the OTC are likely to obey the accepted visual hierarchy scheme. Contrary to this scheme, we show that greater preference of OTC voxels to static pictures of upper-limbs coincides with their stronger functional connectivity with hMT+ (but not hV4 / LO-1). This suggests a tight link between upper-limb selectivity and motion processing. To corroborate this working hypothesis we created a set of natural arm movement videos, in which kinematic patterns were parametrically manipulated, while keeping shape parameters constant. Using multivariate pattern analysis, we show that the degree of (dis)similarity in arm velocity-profiles predicts, to a significant extent, the degree of (dis)similarity in multivoxel activation patterns, in both EAA and hMT+. Together, these results suggest that the functional specificity of EAA is at least partly determined by articulated visual motion. We propose that selectivity to static upper-limb images in the OTC may result from experience-dependent association between shape elements, which characterize upper-limbs, and upper-limb-specific motion patterns.