Abstract
A visual neuron’s response to stimulation of its classical receptive field (CRF) may be suppressed by stimuli presented outside the CRF. This is called “surround suppression” and observed at many levels of the visual system, including the primary visual cortex (V1) and middle temporal complex (hMT+). Here, we used a combination of psychophysical methods, fMRI, and computational modeling to investigate how increasing the extent of spatial attention affects surround suppression in V1 and hMT+ of human participants (N=10, 8 female). The stimulus was a drifting central grating (diameter: 1.5°) with 98% Michelson contrast presented either alone or surrounded by a drifting annular grating (width: 9.2°) with a 1.3° gap between them. Importantly, the drift directions of the central and annular gratings could be the same or opposite. Under two attention conditions, participants were instructed to either limit their attention to the central grating (Narrow Attention, NA) or to both gratings (Wide Attention, WA). First, in a behavioral experiment, we found that surround suppression was significantly stronger under the WA condition compared to the NA condition for the same-direction trials. Next, we conducted a mixed-design fMRI experiment and found that neural suppression in the hMT+, but not in V1, significantly increased as the extent of spatial attention increased in the same-direction trials. Further, we found that hMT+ activity better captures the behavioral results. Finally, we show that incorporating smaller vs. larger multiplicative attentional gain in the normalization model (Reynolds & Heeger, 2009) can successfully predict neural activity patterns and associated behavioral outcomes. Taken together, our results show that spatial attention has a critical role on the behavioral consequences of surround suppression, in which hMT+ plays an important role.