Abstract
We investigated the effect of varying the size, speed and eccentricity of moving random textured patterns (RTP) on the motion processing cortical network in the awake macaque monkey. We used contrast-enhanced fMRI (Vanduffel et al., 2001) experiments as well as the 2-DG technique. Both techniques map functional activity in the whole brain, the former allows us to investigate a wide range of stimulus parameters, while the latter yields excellent spatial resolution.
After an i.v. injection of a contrast agent (MION, 7–11 mg/kg), we acquired GE-EPI whole brain volumes in a 1.5T MR scanner in 2 awake macaques. Stimuli consisted of moving or static RTP's varying in size (3 , 7 , 14 and 28 diam.), speed (1 , 2 , 4 , 8 , 16 /sec) or eccentricity (central 4 or annulus 14–28 ). In the 2-DG experiment, metabolic activity evoked by a moving RTP (30 , 4 /sec) was labeled with 3H-DG and that related to a static RTP (control condition) with 14C-DG.
In agreement with earlier results (Vanduffel et al., 2001), we found motion responses in visual areas V1, V2, V3, MT/V5, FST, VIP and FEF. In the STS however, we found an additional motion responsive site which was located on the posterior bank of the middle STS, 11 mm anterior to MT/V5. Furthermore, area MSTd showed a clear increase in motion response with increasing stimulus size, while the opposite was true for area VIP. Both latter areas showed a preference for higher speeds, as did FST.
These results indicate that the STS contains more motion sensitive regions than classically described. The results obtained by the two functional mapping techniques were highly comparable. This suggests that local hemodynamic signals as measured by fMRI and local cerebral glucose consumption as measured by 2-DG are colocalized. Since DG uptake is primarily linked to pre-synaptic activity (Sokoloff et al., '77), this result is in agreement with the finding by Logothetis et al. (2001) that fMRI signals primarily reflect ‘pre-synaptic’ activity.