Synthesis of higher order feature codes through stimulus-specific supra-linear summation

• DOI: 10.1101/2020.06.24.169359
• Publication Date: 2020-06-25T11:45:15Z
• Authors: Evan H. Lyall; Daniel P. Mossing; Scott R. Pluta; Amir Dudai; Hillel Adesnik
• Keywords: Neural coding; Stimulus (psychology); Stimulus modality

Abstract How cortical circuits build representations of complex objects is poorly understood. The massive dimensional expansion from the thalamus to primary sensory cortex may enable sparse, comprehensive higher order features facilitate object identification. To generate such a code, neurons must integrate broadly over space, yet simultaneously obtain sharp tuning specific stimulus features. logic integration that synthesize high code for not known. address this question, we probed and population coding stimuli in somatosensory visual cortices awake mice using two-photon calcium imaging across layers. We found sum highly combinations inputs supra-linearly, but other sub-linearly, leading selective responses This integrative process generates earliest stages processing. These results multiple modalities imply input-specific supra-linear summation represent widespread mechanism synthesis feature codes. new explain how brain exploits thalamocortical dimensionality encode arbitrary stimuli.