Value-guided remapping of sensory cortex by lateral orbitofrontal cortex

Abstract

Adaptive behaviour crucially depends on flexible decision-making, which in mammals relies on the frontal cortex, specifically the orbitofrontal cortex (OFC) 1−9 . How OFC encodes decision variables and instructs sensory areas to guide adaptive behaviour are key open questions. Here we developed a reversal learning task for head-fixed mice, monitored the activity of neurons of the lateral OFC using twophoton calcium imaging and investigated how OFC dynamically interacts with primary somatosensory cortex (S1). Mice learned to discriminate 'go' from 'no-go' tactile stimuli 10,11 and adapt their behaviour upon reversal of stimulus-reward contingency ('rule switch'). Imaging individual neurons longitudinally across all behavioural phases revealed a distinct engagement of S1 and lateral OFC, with S1 neural activity reflecting initial task learning, whereas lateral OFC neurons responded saliently and transiently to the rule switch. We identified direct long-range projections from lateral OFC to S1 that can feed this activity back to S1 as value prediction error. This top-down signal updated sensory representations in S1 by functionally remapping responses in a subpopulation of neurons that was sensitive to reward history. Functional remapping crucially depended on top-down feedback as chemogenetic silencing of lateral OFC neurons disrupted reversal learning, as well as plasticity in S1. The dynamic interaction of lateral OFC with sensory cortex thus implements computations critical for value prediction that are history dependent and error based, providing plasticity essential for flexible decision-making.

Is the core function of orbitofrontal cortex to signal values or make predictions?

“…And correlative evidence supporting this role continues to accumulate; in various choice tasks, neuronal activity in the OFC correlates with value in fMRI imaging, electrophysiological, and calcium imaging studies [13][14][15][16][17][18][19][20]. These value representations are adaptive to outcome and context changes [15,21,22], modulated by gaze fixation [23] and covert shift of attention [24], and supported by a mixture of temporally dynamic and stable coding schemes [25]. OFC neurons projecting to striatum predominantly encode integrated value among several other decision variables [26], and even theta oscillations in the OFC are reported to track the values of new reward-predictive cues during learning [27].…”

A cortical circuit mechanism for coding and updating task structural knowledge in inference-based decision-making

“…Recent studies began to elucidate the functions of specific projections from OFC to certain cortical and subcortical regions [8][9][10] . However, the sensory discrimination learning 8 or the conventional reversal learning based on the go/no-go task 9 provides only a single prolonged learning curve, does not manifest model-based algorithm, and cannot distinguish rapid behavioral adaptation via model-based inference 2,3,11 from slow model-free reinforcement learning 1,12 . Therefore, it remains an open question how the structured knowledge is represented in the brain, and what type of circuit architecture and operation mechanism is responsible for using and updating of the internal models during flexible decision-making.…”

Value Shapes Abstraction During Learning

“…The facilitating effect of value on abstraction can be thus directly linked to changes in the early processing stage of visual information. In line with this interpretation, recent work in mice has elegantly reported how value governs a functional remapping in the sensory cortex by direct lateral OFC projections carrying RPE information 46 , as well as by modulating the gain of neurons to irrelevant stimuli 47 . While these considerations clearly point to a central role of the vmPFC and valuation in abstraction - by controlling sensory representations, it remains to be investigated whether this effect results in more efficient construction of abstract representations, or in better selection of internal abstraction-based RL algorithms.…”

“…a top-down (attentional) effect, to tag sensory information 45 . Work in rodents has reported strong top-down modulation of sensory cortices by OFC neurons implicated in value computations 46,47 . We thus hypothesized abstraction could result from a direct effect of value in the VC.…”