Distinct cortical networks subserve spatio-temporal sampling in vision through different oscillatory rhythms

Abstract

Although visual input arrives continuously, sensory information is segmented into discrete events. Here, the neural correlates of spatiotemporal binding in male/female human subjects were investigated with MEG using two tasks where separate flashes were presented on each trial but were perceived, in a bi-stable way, as either a single, or two separate, events. The first task (two-flash fusion: TFF) involved judging one versus two flashes while in the second task (apparent motion: AM) participants judged coherent motion versus two stationary flashes. Results indicate two different functional networks underlying two unique aspects of visual temporal binding. In the TFF task, involving an integration window of {approx}50 ms, evoked responses differed as a function of perceptual interpretation by {approx}25 ms after stimuli presentation. Multivariate decoding of subjective perception based on prestimulus oscillatory phase was significant for alpha-band activity in the right medial temporal (MT) area, with the strength of pre-stimulus connectivity between early visual areas and MT being predictive of performance. In contrast, the longer integration window ({approx}130 ms) for AM showed evoked field differences only {approx}250 ms after stimuli onset. Phase decoding of the perceptual outcome in the AM task was strongest for theta-band activity, localized to a right intra-parietal sulcus (IPS) source. Pre-stimulus connectivity between MT and IPS seeds in the theta band best predicted perceptual outcome. Overall, these results show a strong relationship between specific spatiotemporal binding windows and specific oscillations, linked to the information flow between different areas of the "where" and "when" visual processing pathways. Significance StatementMultiple neural rhythms seem relevant for sampling visual information across space and time, but the cortical networks underlying these fundamental computational principles of the visual system remain unexplored. We filled this gap by employing source-level multivariate decoding and connectivity analyses of magnetoencephalographic data recorded during an integration/segregation task of temporal and spatio-temporal events. We identified a first and faster network involving early visual areas (V2 to MT/V5) that determines the basic temporal resolution of visual perception at the speed of the alpha rhythm, and a second slower network involving parietal regions (IPS) that had a key role in the integration of more complex spatiotemporal events at a theta speed. These findings elucidate the neural mechanisms that transfer sensory information into temporal sequences.