Lieu : centre Broca Nouvelle-Aquitaine
Invité par Bordeaux Neurocampus et la NBA
The ability to generate variable movements is essential for learning and adjusting complex behaviors. This variability has been linked to the temporal irregularity of neuronal activity in the central nervous system. However, how this neuronal irregularity actually translates into behavioral variability is unclear. In songbirds, a specialized portion of the basal ganglia-thalamo-cortical circuits generates song variability that underlies vocal experimentation during song acquisition in young birds. Relying on this model system, we combine modeling, electrophysiological and behavioral studies to explain the generation of motor variability and its role during trial-and-error learning. First, I will show how a model circuit comprising topographically organized and strongly recurrent neural networks can autonomously generate irregular motor behaviors. Then, I will present an analysis of simultaneous recordings of neurons in singing finches that reveals an increase in neural correlations across the circuit driving song variability, in agreement with the model predictions. Going beyond neural correlation predicted by our modeling study, I will show that correlations between neural activity and behavior in singing finches also increase along this pathway. Interestingly, this last result may challenge the recently proposed framework for the implementation of reinforcement learning in the song-related BG-thalamo-cortical loop. Finally, analyzing behavioral data, we find remarkable similarities in the babbling statistics of 5-6 month-old human infants and juveniles from three songbird species, and show that our model naturally accounts for these ‘universal’ statistics.