Distinct Roles of the Cortical Layers of Area V1 in Figure-Ground Segregation

What roles do the different cortical layers play in visual processing? We recorded simultaneously from all layers of the primary visual cortex while monkeys performed a figure-ground segregation task. This task can be divided into different subprocesses that are thought to engage feedforward, horizontal, and feedback processes at different time points. These different connection types have different patterns of laminar terminations in V1 and can therefore be distinguished with laminar recordings. We found that the visual response started 40 ms after stimulus presentation in layers 4 and 6, which are targets of feedforward connections from the lateral geniculate nucleus and distribute activity to the other layers. Boundary detection started shortly after the visual response. In this phase, boundaries of the figure induced synaptic currents and stronger neuronal responses in upper layer 4 and the superficial layers ∼70 ms after stimulus onset, consistent with the hypothesis that they are detected by horizontal connections. In the next phase, ∼30 ms later, synaptic inputs arrived in layers 1, 2, and 5 that receive feedback from higher visual areas, which caused the filling in of the representation of the entire figure with enhanced neuronal activity. The present results reveal unique contributions of the different cortical layers to the formation of a visual percept. This new blueprint of laminar processing may generalize to other tasks and to other areas of the cerebral cortex, where the layers are likely to have roles similar to those in area V1. [Display omitted] •Feedforward, horizontal, and feedback processes activate different laminar circuits•Boundary detection occurs early in upper layer 4 and the superficial layers of V1•Labeling of figural regions with enhanced activity occurs in superficial and deep layers•Region filling is caused by currents targeted to feedback recipient layers 1, 2, and 5