Laminar Organization of Attentional Modulation in Macaque Visual Area V4

Abstract

SUMMARY Attention is critical to perception, serving to select behaviorally relevant information for privileged processing. To understand the neural mechanisms of attention, we must discern how attentional modulation varies by cell-type and across cortical layers. Here we test whether attention acts non-selectively across cortical layers or does it engage the laminar circuit in specific and selective ways. We find layer- and cell-class specific differences in several different forms of attentional modulation in area V4. Broad-spiking neurons in the superficial layers exhibit attention mediated increases in firing rate and decreases in variability. Spike-count correlations are highest in the input layer and attention serves to reduce these correlations. Superficial and input layer neurons exhibit attention dependent decreases in low-frequency (<10Hz) coherence, but deep layer neurons exhibit increases in coherence in the beta and gamma frequency ranges. Our study provides a template for attention-mediated laminar information processing that might be applicable across sensory modalities.

Paying Attention to the Cortical Layers

“…In this issue of Neuron, Nandy et al (2017) reveal a number of important new insights into the neural mechanisms that are responsible for attentional selection.…”

“…In this issue of Neuron, Nandy et al (2017) reveal a number of important new insights into the neural mechanisms that are responsible for attentional selection. Many of the insights about the neuronal processing underlying attention shifts have come from studies in the visual cortex of monkeys that were trained to direct their attention to some visual items while ignoring others.…”

“…These proposed mechanisms are not mutually exclusive and the question of how much each of them contributes to attentional selection has not been fully resolved. Nandy et al (2017) addressed these questions by investigating neuronal activity in the different cortical layers, which have distinct roles in visual analysis and attentional processing (Figure 1A). The bottom-up driving visual input from lower brain regions arrives in layer 4, whereas the top-down inputs associated with attention shifts tend to avoid layer 4.…”

“…The authors took advantage of this division of labor between the layers in area V4, a mid-tier visual area where activity is strongly modulated by attention. Nandy et al (2017) also examined the contribution of two functionally defined subclasses of cells: broad-spiking (putative excitatory) and narrow-spiking (putative inhibitory) cells as previous studies suggested that inhibitory neurons contribute strongly to attention shifts (Mitchell et al, 2007). Nandy et al (2017) had to overcome a number of technical challenges to identify the different layers in V4.…”

“…Nandy et al (2017) also examined the contribution of two functionally defined subclasses of cells: broad-spiking (putative excitatory) and narrow-spiking (putative inhibitory) cells as previous studies suggested that inhibitory neurons contribute strongly to attention shifts (Mitchell et al, 2007). Nandy et al (2017) had to overcome a number of technical challenges to identify the different layers in V4. They used laminar electrodes with multiple contacts along a single shaft that needed to be positioned perpendicular to the cortical layers, which is challenging in area V4 because it has a complex shape.…”