The Center For Advanced Mathematical Sciences invites you to a seminar entitled "Intrinsic neuronal properties of cortical neurons measure error in vocal learning" by Arij Daou, AUB.
Production of learned vocalizations requires precise comparison of output with feedback, evolving at the same rate as vocal production. The plastic mechanisms supporting such rapid dynamics are unknown. Recording from the basal ganglia–projecting “HVCX” forebrain neurons in adult zebra finch brain slices and mathematical modeling of these cortical neurons, we show that within each bird, the neurons share very similar somatic electrophysiological intrinsic properties (IP). The IP vary among birds in lawful relation to acoustic similarity of the birds’ songs. This arises from dynamic processes: HVCX within juveniles learning to sing show variable IP, and the IP uniformity within a given adult bird rapidly degrades, within hours if not quicker, during singing while exposed to abnormal auditory feedback. This represents the first example of a long-sought physiological signal that varies rapidly with changes in auditory feedback during singing. This signal arises from unanticipated network interactions with non-synaptic mechanisms that affect information storage and processing.