Testing the long-term stability of a neural code in songbirds

Jeffrey Markowitz (Boston University)

Memories, habits, and motor skills may persist for a lifetime but little is known about the stability of neural activity underlying this persistence of memory. In an effort to study the limits of stability in a stereotyped neural code, we have continuously monitored neural activity from the pre-motor nucleus HVC in singing zebra finches. To perform long-term recordings, we developed minimally invasive carbon fiber electrodes designed to minimize immune rejection.  

In these recordings we find detailed spiking pattern patterns that persist for months. Furthermore, after nerve damage to the syrinx (vocal organ), we find evidence that central motor patterns remain unchanged for a minimum of one month, despite disruption of the acoustic form of the song.  These findings suggest that learned motor sequences, produced by cortical-thalamic loops, can be remarkably stable, even in the absence of normal proprioceptive and sensory feedback.

We discuss these results in the context of current knowledge of motor stability in other vertebrate behaviors, and consider prospects for studying the ontogeny of song motor sequences as young birds learn to sing.


Preferred presentation format: Poster
Topic: Electrophysiology

Document Actions