Biology subjects

Come, M.

Publications and source records attributed to Come, M..

4 recordsLinked to original sources

bioRxiv2024

Dopamine reveals adaptive learning of actions representation

Flexible decision-making requires not only updating values, but redefining which features constitute an action in a given context. We recorded nucleus accumbens (NAc) dopamine release while mice navigated a three-target intracranial self-stimulation foraging task in which outcomes were evaluated under three distinct reward delivery rules. Despite a constant motor repertoire, dopamine transients reorganized across contingencies and generalized linear models revealed context-dependent dopamine signal reflecting action direction, recent outcome-history, or target identity. Reinforcement-learning model comparison showed that these signatures are best explained by distinct reward prediction errors (RPEs) defined over different state-action representations, rather than a single fixed model-free scheme. A single deep reinforcement-learning agent trained by temporal-difference learning, recapitulated both the rule-specific policies and the corresponding dopamine signature. These results identify NAc dopamine as a dynamic readout of representation learning, remapping prediction errors onto the task features that define successful action as contingencies change.

neuroscience↗

bioRxiv2022

Self-generation of goal-directed choices in a distributed dopaminergic andprefrontal circuit

Goal-directed choices that are not triggered by external cues arise from internal representations of the outcomes. The use of a stimulus to specify when to act, which option to take, or whether to explore, has led to consider the reward circuit as a feedforward set of modules carrying independent computations. Here, we develop an uncued task in which mice self-determine the initiation, direction, vigor and pace of their actions based on their knowledge of the outcomes. Using electrophysiological recordings, pharmacology and optogenetics, we identify a sequence of oscillations and firing in the ventral tegmental area (VTA), orbitofrontal (OFC) and prefrontal cortices (PFC) that co-encodes and co-determines self-initiation and choices. This sequence appeared with learning as an unguided realignment of spontaneous dynamics. The interactions between the structures depended on the reward context, in particular regarding the uncertainty associated with the different options. We suggest that self-generated choices arise from a distributed circuit based on an OFC-VTA core setting whether to wait or to initiate actions, while the PFC is specifically engaged by reward uncertainty to participate in both the selection and pace of actions. HighlightsO_LISelf-paced actions arise from contextual reorganization of mesocortical dynamics. C_LIO_LIVTA, PFC and OFC complementarily encode predictions and errors about outcomes. C_LIO_LIDistributed firing-then-oscillations dynamics set the goal, initiation and pace of actions. C_LIO_LIVTA and PFC antagonistically promote and inhibit motivation by reward uncertainty. C_LI

neuroscience↗

bioRxiv2021

Chronic nicotine increases midbrain dopamine neuron activity and biases individual strategies towards reduced exploration in a foraging task.

Long-term exposure to nicotine alters brain circuits and induces profound changes in decision-making strategies, affecting behaviors both related and unrelated to drug seeking and consumption. Using an intracranial self-stimulation reward-based foraging task, we investigated the impact of chronic nicotine on the trade-off between exploitation and exploration, and the role of ventral tegmental area (VTA) dopamine (DA) neuron activity in decision-making unrelated to nicotine-seeking. Model-based and archetypal analysis revealed a substantial inter-individual variability in decision-making strategies, with mice passively exposed to chronic nicotine visiting more frequently options associated with higher reward probability and therefore shifting toward a more exploitative profile compared to non-exposed animals. We then mimicked the effect of chronic nicotine on the tonic activity of VTA DA neurons using optogenetics, and found that photo-stimulated mice had a behavioral phenotype very close to that of mice exposed to nicotine, suggesting that the dopaminergic control of the exploration/exploitation balance is altered under nicotine exposure. Our results thus reveal a key role of tonic midbrain DA in the exploration/exploitation trade-off and highlight a potential mechanism by which nicotine affects decision-making.

neuroscience↗

bioRxiv2019

Mice adaptively generate choice variability in a deterministic task

Can our choices just be driven by chance? To investigate this question, we designed a deterministic setting in which mice reinforce non-repetitive choice sequences, and modeled it using reinforcement learning. Mice progressively increased their choice variability using a memory-free, pseudo-random selection, rather than by learning complex sequences. Our results demonstrate that a decision-making process can self-generate variability and randomness even when the rules governing reward delivery are not stochastic.

neuroscience↗