La perception menant à une prise de décision implique de multiples aires corticales. Il a été proposé que l'information sensorielle se propage des aires sensorielles primaires, codant principalement la nature du stimulus, aux aires de haut-niveau - plus frontales - codant d'avantage la valence du stimulus ou la décision. Pour mieux comprendre l'intégration corticale des signaux sensoriels, nous avons enregistré les réponses sensorielles évoquées (RSE) simultanément dans différentes aires corticales, tandis que des souris apprenaient une tâche de détection sensorielle. Chez les souris ayant appris la tâche, une RSE est observée dans toutes les aires enregistrées suivant la stimulation de la vibrisse, avec des latences croissantes des aires somatosensorielles primaire (vS1) et secondaire (vS2), vers le cortex moteur primaire des vibrisses (vM1), le cortex pariétal associatif (PtA), l'hippocampe dorsal (dCA1) et enfin le cortex préfrontal médian (mPFC). Nous avons constaté une réduction des RSEs lors des échecs par rapport aux essais réussis dans toutes les aires, sauf vS1. Toutefois, seule l'inactivation de vS1, vS2 ou mPFC affecte significativement la performance des souris. Pendant l'apprentissage de la tâche, une augmentation sélective de la RSE est observée dans le mPFC en corrélation avec la performance. Des enregistrements unitaires dans le mPFC démontrent la nature excitatrice de la réponse sensorielle chez les souris entrainées. Nos résultats confirment ainsi que la réponse sensorielle dans le mPFC reflète l'importance comportementale du stimulus et corrèle avec la prise de décision, tandis que la réponse des aires sensorielles reflète plutôt la nature du stimulus / Sensory perception leading to goal-directed behavior involves multiple, spatially-distributed cortical areas. It has been hypothesized that sensory information flows from primary sensory areas encoding mainly the nature of the stimulus, to higher-order, more frontal, areas encoding the valence of the stimulus or the decision. To further understand the cortical integration of sensory signals, we recorded sensory evoked potentials (SEPs) simultaneously from different areas while mice learned a whisker-based sensory detection task. In mice that have learned the task, the whisker stimulus evoked SEP in all recorded areas with latencies increasing from the whisker primary (wS1) to the secondary somatosensory area (wS2), the whisker motor area (wM1), the parietal area (PtA), the dorsal hippocampus (dCA1) and the medial prefrontal cortex (mPFC). We found a reduction of SEPs during Miss trials compared with Hit trials in all areas except wS1. However, only the local inactivation of either wS1, wS2 or mPFC significantly impaired the mice performance. During training to the detection task, we observed a selective increase of the SEPs in mPFC that correlated with performance. Finally, using high-density extracellular recordings in mPFC, we found that whisker stimulation in trained mice evoked an early increase in the firing rate of putative excitatory neurons (regular spiking units) that was positively correlated with behavioral outcome. Our results support the idea that mPFC could signal the relevance of a sensory stimulus in the context of a well-defined behavior, whereas sensory areas would be more constrained by the nature of the stimulus

In the study of perceptual decision making, it has been widely assumed that random fluctuations of motion stimuli are irrelevant for a participant’s choice. Recently, evidence was presented that these random fluctuations have a measurable effect on the relationship between neuronal and behavioral variability, the so-called choice probability. Here, we test, in a behavioral experiment, whether stochastic motion stimuli influence the choices of human participants. Our results show that for specific stochastic motion stimuli, participants indeed make biased choices, where the bias is consistent over participants. Using a computational model, we show that this consistent choice bias is caused by subtle motion information contained in the motion noise. We discuss the implications of this finding for future studies of perceptual decision making. Specifically, we suggest that future experiments should be complemented with a stimulus-informed modeling approach to control for the effects of apparent decision evidence in random stimuli.

Perceptual decision making can be described as a process of accumulating evidence to a bound which has been formalized within drift-diffusion models (DDMs). Recently, an equivalent Bayesian model has been proposed. In contrast to standard DDMs, this Bayesian model directly links information in the stimulus to the decision process. Here, we extend this Bayesian model further and allow inter-trial variability of two parameters following the extended version of the DDM. We derive parameter distributions for the Bayesian model and show that they lead to predictions that are qualitatively equivalent to those made by the extended drift-diffusion model (eDDM). Further, we demonstrate the usefulness of the extended Bayesian model (eBM) for the analysis of concrete behavioral data. Specifically, using Bayesian model selection, we find evidence that including additional inter-trial parameter variability provides for a better model, when the model is constrained by trial-wise stimulus features. This result is remarkable because it was derived using just 200 trials per condition, which is typically thought to be insufficient for identifying variability parameters in DDMs. In sum, we present a Bayesian analysis, which provides for a novel and promising analysis of perceptual decision making experiments.

Perceptual decision making can be described as a process of accumulating evidence to a bound which has been formalized within drift-diffusion models (DDMs). Recently, an equivalent Bayesian model has been proposed. In contrast to standard DDMs, this Bayesian model directly links information in the stimulus to the decision process. Here, we extend this Bayesian model further and allow inter-trial variability of two parameters following the extended version of the DDM. We derive parameter distributions for the Bayesian model and show that they lead to predictions that are qualitatively equivalent to those made by the extended drift-diffusion model (eDDM). Further, we demonstrate the usefulness of the extended Bayesian model (eBM) for the analysis of concrete behavioral data. Specifically, using Bayesian model selection, we find evidence that including additional inter-trial parameter variability provides for a better model, when the model is constrained by trial-wise stimulus features. This result is remarkable because it was derived using just 200 trials per condition, which is typically thought to be insufficient for identifying variability parameters in DDMs. In sum, we present a Bayesian analysis, which provides for a novel and promising analysis of perceptual decision making experiments.