Happy to share our new paper https://t.co/NGtvz15Y71 in Nature Communications with @FindlingCharles, Margaux Romand and @vasilisa_skv: How neural variability/noise in the prefrontal cortex contributes to efficient adaptive behavior without facing computational complexity issues!
Glad to share our new paper https://t.co/c8ZZkdVvmg in @NatureHumBehav entitled "Imprecise neural computations as a source of adaptive behaviour in volatile environments" by @FindlingCharles, Nicolas Chopin and @KoechlinE. The Weber Law in inferring stable state beliefs is key!
Our new paper https://t.co/fFM7GAa4Bj by @PhilippeDOMENE2, Sylvain Rheims and @KoechlinE describing with model-based intracranial EEG how the prefrontal cortex resolves exploitation-exploration dilemmas through predictive coding between the vmPFC and dmPFC. @ScienceMagazine
Showing how noisy computations conforming to the Psychophysical Weber Law dispense humans to make complex inferences about environment volatility to exhibit efficient behavior in volatile environnements!!!
Our last paper https://t.co/tz5hw57VCe just released entitled “imprecise neural computations as source of human adaptive behavior in volatile environments” by @FindlingCharles, Nicolas Chopin & @KoechlinE.
New @biorxivpreprint on agency, instrumental learning and counterfactual simulation: "Believing in one's power: a counterfactual heuristic for goal-directed control", with Héloise Théro, @FindlingCharles and @KoechlinE#preprint#christmasreading
1/8
https://t.co/W03ueuBq14
Complement our last study https://t.co/Ss5edt2oAR in Cerebral Cortex by Sandrine Duverne et Etienne Koechlin about ACC-DLPFC interaction combining rewards and cognitive control. All open access!
our new paper https://t.co/z0K6qkSWO0 in Nature Comm "prefrontal mechanisms integrating rewards and beliefs in human decision-making" by Marion Rouault, Jan drugowitch @jdrugowitsch and Etienne Koechlin @KoechlinE. It is about the neural sources of subjective probabilities
We show that subjective probabilities reflect that proper beliefs encoded in the vmPFC are treated as affective signals that add up to rewards values in the ACC rather than as probabilities underlying expected value computations