The birdsong system has become a paradigmatic example of biological learning, and one in which there is hope that classic learning algorithms may be mapped to the underlying neural circuits and biophysics. In this talk we will discuss how detailed biological responses can help this system to implement reinforcement learning. In particular, we will discuss the potential role of a newly identified excitatory neural signal in basal ganglia and how it may help to modulate basal ganglia synchrony and the variability required for learning.