Abstract
To obtain a precise demarcation of human visual areas, it is necessary to use functional definitions. Retinotopic mapping routinely identifies many areas, but V6 is usually not among them. A previous study has shown that contrasting responses to egomotion-consistent and -inconsistent stimuli identifies visual areas in the intraparietal sulcus (putative VIP) and the cingulate sulcus visual area (CSv) (Wall and Smith, 2008). We report that this method also reliably identifies V6/V6A, and two parietal areas that may correspond to visual-vestibular regions.
We performed an event-related 3T fMRI experiment in which 10 subjects were scanned using a posterior array coil while looking at egomotion-consistent or -inconsistent stimuli. The display consisted of moving dots arranged into either a single 50-deg patch (egomotion-consistent; EC) or a 3×3 array of nine patches (egomotion-inconsistent; EI) of time-varying optic flow. Each stimulus was presented for 3 sec and inter-trial intervals varied between 2–10 sec.
Single-subject analysis was conducted by fitting a GLM with regressors representing the two stimulus categories and six movement parameters. The statistical contrast [EC [[gt]] EI] revealed consistent activations across subjects in three regions in addition to VIP and CSv. One, evident in 19 hemispheres, is at the dorsal margin of the postcentral sulcus, within or close to somatosensory cortex. The second, evident in 18 hemispheres, is in the medial parietal operculum, described by Antal et al. 2008 as PO/PT, and possibly corresponding to PIVC. The third, readily seen in all hemispheres, is a region of the parieto-occipital sulcus, medial to V3A and consistent with the description of human V6 (Pitzalis et al, 2006). Differential activation was sometimes also observed in the MT complex.
We show that these stimuli provide a simple and reliable localiser for the V6 complex and we suggest that V6 may be involved in the computation of egomotion.
Funded by the Wellcome Trust.