Abstract
When an identical stimulus is presented repeatedly, the activity of sensory cortical neurons varies from trial to trial, dubbed 'neuronal noise'. Recent electrophysiological and imaging studies reported that ‘noise’ is not just a random and independent deviation from ‘signal’ and reflects correlated activity among local cortical sites. Here we investigated the structure of correlated noises in early human visual areas by monitoring moment-to-moment fluctuations in fMRI responses from V1, V2 and V3 during visual stimulation or during a resting state. A signal correlation for a pair of voxels was estimated from a correlation in across-trial-averaged time series of responses to periodic traveling waves of orientation or spatial frequency-modulated stimuli. A noise correlation was estimated from a correlation between two voxels in time series of deviations from their own averages. Individual voxel pairs were also characterized in terms of visuotopic distance via fine-grained population receptive mapping and in terms of distance along cortical surface via segmentation of high-resolution anatomical images. Noise correlations were high for the voxel pairs with high signal correlation (high orientation or spatial frequency tuning similarity), short visuotopic or cortical distance. The analysis of partial correlation, where physical distance factor was held constant, confirmed that the observed functional structure of noise correlation cannot be explained by trivial factors such as spatially correlated MR noises or head motions. Resting-state brain imaging further revealed a significant correlation between noise correlations and correlations in spontaneous BOLD fluctuation, no matter eyes closed or open with fixation on the center of a gray-screen display. Resting-state correlations were highly dependent on signal correlations as the same manner exhibited by noise correlations. Finally, the functional architecture of noise/resting-state correlations was held same both within and between visual areas.
Meeting abstract presented at VSS 2012