Cortical Localization of the Sensory-Motor Transformation in a Whisker Detection Task in Mice

Abstract

Responding to a stimulus requires transforming an internal sensory representation into an internal motor representation. Where and how this sensory-motor transformation occurs is a matter of vigorous debate. Here, we trained male and female mice in a whisker detection go/no-go task in which they learned to respond (lick) following a transient whisker deflection. Using single unit recordings, we quantified sensory-related, motor-related, and choice-related activities in whisker primary somatosensory cortex (S1), whisker region of primary motor cortex (wMC), and anterior lateral motor cortex (ALM), three regions that have been proposed to be critical for the sensory-motor transformation in whisker detection. We observed strong sensory encoding in S1 and wMC, with enhanced encoding in wMC, and a lack of sensory encoding in ALM. We observed strong motor encoding in all three regions, yet largest in wMC and ALM. We observed the earliest choice probability in wMC, despite earliest sensory responses in S1. Based on the criteria of having both strong sensory and motor representations and early choice probability, we identify whisker motor cortex as the cortical region most directly related to the sensory-motor transformation. Our data support a model of sensory encoding originating in S1, sensory amplification and sensory-motor transformation occurring within wMC, and motor signals emerging in ALM after the sensory-motor transformation.

Global, Low Amplitude Cortical State Predicts Response Outcomes in a Selective Detection Task

“…The experimental datasets in this study were previously published, including the whisker monitoring, widefield GCaMP6 imaging (Aruljothi et al, 2020) and single unit recordings (Zareian et al, 2021). Below, we summarize these experimental methods and describe the new analyses used in this study.…”

“…Sensory and motor encoding of single units was performed as previously reported (Zareian et al, 2021).…”

“…Single unit recordings during expert task performance were conducted as previously reported (Zareian et al, 2021). The dataset consists of 32 behavioral/recording sessions, recorded from 22 mice, yielding a total of 936 single units from three cortical regions (target-aligned whisker region of primary somatosensory cortex [S1], whisker region of motor cortex [wMC], and anterior lateral motor cortex [ALM]).…”

“…Choice probability (%) was quantified as the separation of prestimulus spiking in Hits versus Miss trials. ROC and AUC were calculated from the distributions of Hits and Miss across trials, 500 ms to 0 ms before stimulus onset, averaged over 50 ms nonoverlapping intervals, as previously reported (Zareian et al, 2021)..…”

“…We address these questions in the context of a selective whisker detection task in mice. We trained mice to respond (lick) to deflections on one whisker field (target) and ignore deflections in the contralateral whisker field (distractor) (Aruljothi et al, 2020;Zareian et al, 2021). Using widefield Ca 2+ imaging, we previously identified the cortical regions that are highly active post-stimulus and pre-response, and therefore may contribute to stimulus detection: the whisker region of primary somatosensory cortex (S1), the whisker region of primary motor cortex (wMC), and the pre-motor licking region anterior lateral motor cortex (ALM) (Aruljothi et al, 2020).…”