Non-linearity of spatial integration varies across layers of primary visual cortex

Area of Science:

• Neuroscience
• Computational Neuroscience
• Visual Processing

Background:

• Receptive fields (RFs) of visual neurons are dynamic and context-dependent.
• Classical RF mapping misses subthreshold synaptic inputs.
• Cortical circuit architecture influences RF properties.

Purpose of the Study:

• Investigate spatial integration in different cortical layers.
• Focus on the linearity of spatial summation in visual cortical neurons.
• Explore mechanisms behind layer-specific integration properties.

Main Methods:

• Intracellular recordings in cat primary visual cortex (V1).
• Analysis of spatial summation linearity in different cell types and layers.
• Large-scale recurrent spiking model of cat V1.

Main Results:

• Supragranular complex cells show highly sublinear spatial summation.
• Infragranular complex cells and simple cells exhibit near-linear summation.
• Layer-specific integration properties may stem from differential horizontal connections.

Conclusions:

• Receptive fields are shaped by feedforward, horizontal, and feedback inputs.
• Spatial summation linearity varies across cortical layers and cell types.
• Challenging conventional notions of fixed RF boundaries in early visual processing.