To do this we first trained rats to use a light cue to predict food pellets. Across time, rats learn this association and we know that after this learning has occurred, DA neurons will show a PE at cue onset. 2/n
-
-
Prikaži ovu nit
-
We then tested if the increase in DA at cue onset would support new learning. We trained rats that another cue (e.g. a tone) predicted the light. Normally, rats show an increase in appetitive responding to the tone after this training because it predicts to food-paired light. 3/n
Prikaži ovu nit -
However, in half the rats we shunted DA firing at the light onset, preventing the PE from occurring during tone-light pairings. We found that this reduced appetitive responding to the tone, demonstrating the PE at light onset functions to support new learning. 4/n
Prikaži ovu nit -
Now, we asked if the PE at light onset also functions to tell the brain what is about to happen (i.e. the prediction for food). Here, we used a blocking procedure. After rats learnt the light predicts food, we taught them that a compound (e.g. light-click) also predicts food. 5/n
Prikaži ovu nit -
Usually, rats don’t learn about the click b/c it doesn’t predict anything new, just the food pellets. However, again we shunted DA firing at light onset. If the PE at light onset functions to signal the prediction, rats won’t know food is coming and will learn about the click 6/n
Prikaži ovu nit -
But this isn’t what happened, preventing DA from firing at light onset did not prevent the light from blocking learning about the click. This demonstrates that the DA PE functions exactly as that, a PE, and does not carry the prediction for the upcoming reward. 7/7
Prikaži ovu nit
Kraj razgovora
Novi razgovor -
-
-
Very neat!
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.
