[CTN] correction & one more Laurier talk

[Matthijs van der Meer]

Apologies for the lack of a date for this one: it's February 28.

Dr. Guillermo Esber
Post-doctoral fellow, Johns Hopkins University / NIDA IRP
Friday, February 28

11:00 AM - 12:30 PM

N2005, Science Bldg, WLU

Title: Basolateral amygdala, prediction errors, and attention

Abstract: In Pavlovian learning, a prediction error (PE) is said to
occur when there is a mismatch between the reinforcer that the animal
predicts on the basis of antecedent cues and the reinforcer that
actually occurs. The dominant view within Learning Theory posits that
the primary effect of PEs is to instigate learning, thereby adjusting
the animal's predictions and promoting adaptive behavior. However, there
is good evidence to suggest that a secondary effect of PEs is to
heighten the salience of antecedent cues, rendering them better able to
capture attention when they are next encountered. Interestingly, these
two effects of PEs, salience acquisition and learning, may be uncoupled
both at the behavioral and neural level. The experiments I will describe
show that the basolateral nucleus of the amygdala (BLA) is critical for
enhancing cue salience but not for learning per se. First, lesions of
the BLA abolish the incremental effects of PEs on cue salience while
sparing simple associative learning. Second, BLA neurons track
theoretical changes in acquired salience as predicted by the Pearce-Hall
model, an attentional theory of associative learning. This neural
signal, moreover, seems to depend on midbrain dopaminergic input. Taken
together, the evidence identifies the BLA as part of a circuit that
modulates the salience of environmental stimuli to help animals allocate
their attention efficiently as they navigate towards their goals.