1/5 We introduce a behavioural task in mice that requires integration of reward value and sensory confidence. Using RL modelling we formalise trial-by-trial learning.
-
-
Prikaži ovu nit
-
2/5 Many mPFC neurons integrate confidence and reward value prior to outcome, encoding the value of chosen option.
Prikaži ovu nit -
3/5 Dopamine neurons integrate confidence and reward value both before and after the outcome, encoding the value of chosen option, and reward prediction error.
Prikaži ovu nit -
4/5 Optogenetic manipulation of mPFC pre-outcome (but not post-outcome) activity drives learning. In contrast, optogenetic manipulation of dopamine post-outcome (but not pre-outcome) activity drives learning.
Prikaži ovu nit -
5/5 Thus, both medial frontal and dopamine responses are important for learning but not for choice computation: the first one by signalling reward prediction and the second one by evaluating the prediction.
Prikaži ovu nit -
and here is the link to the paper:https://www.sciencedirect.com/science/article/pii/S0896627319309821 …
Prikaži ovu nit
Kraj razgovora
Novi razgovor -
-
-
Congrats Armin et al! Great work!
-
Thanks Nick!
Kraj razgovora
Novi razgovor -
-
-
Congratulations Armin! Very nice to see this paper out.
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
Amazing! Well done.
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.