The mammalian brain can form associations between behaviourally relevant stimuli in an animal’s Environment. While such learning is thought to primarily involve high-order association cortex, even Primary sensory areas receive long-range connections carrying information that could contribute to High-level representations. Here, we imaged layer 1apical dendrites in the barrel cortex of mice performing a whisker-based operant behaviour. In addition to sensory-motor events, calcium signals in apical dendrites of layers 2/3 and 5 neurons and in layer 2/3 stomata track the delivery of rewards, both choice related and randomly administered. Reward-related tuft-wide dendritic spikes emerge gradually with training and are task specific. Learning recruits cells whose intrinsic activity coincides with the time of reinforcement. Layer 4 largely lacked reward-related signals, suggesting a source other than the primary thalamus. Our results demonstrate that a sensory cortex can acquire a set of associations outside its immediate sensory modality and linked to salient behavioral events.

Apical dendrites are a common feature of pyramidal neurons throughout the mammalian neocortex, suggesting a general function in cortical computation. Pyramidal neurons in cortical layers 2/3 and 5 send apical dendrites to the surface of the cortex, where they arborize in layer 1. Layer 1 is composed almost entirely of these apical dendrites and axons from both local and distant sources. Being devoid of somata except for a sparse population of inhibitory cells, layer 1 has been largely inaccessible by electrophysiology during behavior. Consequently, the role of apical dendrites in cortical processing remains mysterious.

Demonstrate that learned associations reinforced by reward can become potent additional drivers of apical dendrite activity, suggesting that apical dendrites could be a major conduit for assimilating disparate non-modalityspecific, contextual information into a sensory representation. We recorded calcium signals in apical dendrites and somata of cortical neurons using 2-photon imaging of sensors genetically targeted to specific layers of the barrel cortex while mice performed a tactile detection task. Reward-related signals were prominent in the apicals of layers 2/3 and 5 pyramidal neurons within layer 1 but not in the somata of layer 4 neurons, indicative of a non-afferent origin. These reward-reinforced associations emerged with learning and were task specific, in that such signals in the barrel cortex required training on a whisker-based task. Our results suggest that modalityspecific reinforcement recruits layer 1 apical dendrites of primary sensory cortex into new representations that extend beyond their normal repertoire of environmental sensory stimuli.

Apical activity during rewards could result from motor inputs into layer 1. Reward consumption inherently involves licking, but isolated spontaneous licking bouts during inter-trial intervals in the absence of water did not increase calcium to the level seen with random rewards  Another possibility is that reward delivery could arouse a mouse and induce additional whisking. Whisking did not appear to consistently drive calcium influx, but a detailed analysis revealed a weak correlation of whisking and calcium this correlation, however, fell to nearly 0 in those epochs in which rewards were administered (Figures S3B and S3D). As with licking, calcium responses to the onset of isolated whisking bouts differed markedly from responses to unexpected isolated rewards in the same

Sessions, suggesting that motor input cannot explain our results. Average responses to trial and random rewards are similar in amplitude; they could represent the activity of a single population of reward-sensitive dendrites or, alternatively, activity in distinct populations.

This could afford a powerful generalized mechanism for encoding task-relevant information to any given cortical area, including associations with multiple sensory modalities and motor behaviors, as well as predictions about upcoming inputs.

You can submit your research work through online at https://bit.ly/3k7QUiv

Media contact:

Christina

Editorial Assistant

Mail Id: cellsignaling@emedicalsci.com

Journal of Cell Signaling.

Whatsapp number: + 1-504-608-2390