John Reynolds, Ph.D.
Fiona & Sanjay Jha Chair in Neuroscience
The Salk Institute for Biological Studies
“Studies of Attention and Perception in New and Old World Monkeys”
Studies of visual attention in the non-human primate have found that endogenously-generated attention signals modulate neuronal responses in ways that closely parallel changes that are observed with increases in salience. These include changes in response gain, modulation of center-surround interactions, and reductions in correlated neuronal response variability. This suggests that changes in attention and salience may both tap into some of the same neural circuitry. I will present data supporting this conclusion, within the framework of the Normalization model of Attention. This work suggests that evolution may have co-opted circuits that originally evolved to regulate sensitivity in response to changes the strength of sensory input, but now also mediate attention, where they adjust sensitivity to meet changing task demands. I will then present unpublished work showing a form of neural response fluctuation that improves perception. Using recently developed analytical techniques, we are able to detect ubiquitous spatiotemporal patterns of neural activity in Area MT of the common marmoset. These patterns take the form of traveling waves that propagate across the cortical surface. They occur spontaneously without need for a triggering visual stimulus, and even occur in darkness. Traveling waves form naturally in a sparsely connected random E/I balanced network model. The model predicts that when stimuli coincide with the depolarized phase of the wave, their evoked responses will be increased in gain, and that this will result in a reduction in perceptual threshold. We find clear support for both model predictions. Propagating waves thus appear to play a positive role in perception, providing brief windows of improved sensitivity to incoming sensory stimuli.