Traces of top-down modulations of sensory responses in behaving primate by Nicolas Priebe

Early sensory cortical areas receive a large number of feedback projections from higher cortical areas, yet the role of these connections in mediating behavior is not well understood. Such top-down signals could optimize cortical representations for task demands, yet these modulations are likely to be subtle in order to minimize the interference with sensory signals that are useful for multiple concurrent perceptual goals. Consistent with this possibility, top-down modulations are typically small or absent in single unit measurements in macaque V1. It is possible, however, that important top-down modulations are present at the subthreshold level. Indeed, previous (Vm) studies using voltage-sensitive dye imaging, which measures the pooled membrane potential over large populations of neurons, revealed robust top-down modulations in macaque V1 (Chen & Seidemann, 2012), but the effects of these modulations at the single neuron level are unknown. The goal of the current study was to use whole cell recording to measure, for the first time, the effects of top-down modulations at the level of single V1 neurons. Two macaque monkeys were required to report whether a small, low contrast Gabor target appeared at one of two possible locations by making a saccade toward the target location. At a random interval after the animal established fixation, it received a temporal cue 300 ms prior to target onset (tone + fixation point dimming). In a subset of the trials, no temporal cue was given and no visual target appeared. On those ‘blank’ trials the animal was required to maintain fixation. As expected, when the target appeared in V1 neuron’s receptive field, we observed clear contrast-dependent visual response. However, when we compared Vm in detection trials in which no target appeared in the receptive field to Vm in blank trials, we observed reliable depolarization that started shortly prior to stimulus onset.  V1 neurons exhibited significant Vm depolarization in the interval between target onset and saccade initiation (Monkey T, mean depolarization =1.84 +/- 0.15 mV, N=11, monkey A, mean depolarization = 1.24 +/- 0.19, N=11). Of the 22 recorded neurons, 18 showed significant Vm depolarization. This small but reliable depolarization has a weak effect on the spiking activity of single neurons, but is likely to have a robust effect at the level of neural population due to the highly interconnected nature of cortical circuits. These results demonstrate that primary visual cortex is not a purely sensory area, but instead contextual information modulates visual responses in a way that is likely to improve performance in demanding perceptual tasks.

Nicolas Priebe is Professor at Department of Neuroscience University of Texas, Austin USA and is invited professor in the team of Cendra Agulhon in june 2019.