The proposed work will advance understanding of the neural mechanisms of sensory-motor learning by providing a quantitative description of the relationship between multiple physiological variables and vocal performance during development in a songbird, the zebra finch. Young male zebra finches learn their songs from adult models through a process of imitation that resembles human speech acquisition. Learned song, consists of a stereotyped sequence of distinct sounds produced by coordinated action of the two brain hemispheres, via right and left sound generators in the bird's syrinx (vocal organ), on patterned respiratory pressure pulses. Behavioral methods have been developed that trigger the vocal imitation process by presenting a model song, then measure the emerging structure of vocal output on a moment-to-moment basis. The bird rapidly transforms its limited, unstable vocal repertoire into a pattern of distinct syllable types defined by temporal and acoustic parameters - a process of differentiation and stabilization. This application proposes an interdisciplinary collaboration to study this learning process at the behavioral, articulatory, and brain levels in laboratories with expertise in physiological, behavioral and computational approaches to motol earning. Using the standardized behavioral paradigm, vocal output and physiological data will be recorded from different levels of the system at critical times during learning. The emerging structure of song will be elated to: 1) articulatory gestures recorded as respiratory pressure patterns and syringeal airflow; 2) premotor activity in subregions of vocal nucleus RA involved in respiratory vs. syringeal control; and 3) the role of synchronizing inputs to vocal nucleus HVC in song initiation and coordination. Quantitative analysis of the behavior and multi-level physiological data will yield a detailed account of how neural activity is structured into a learned temporal pattern during development and how multiple patterns, controlling different vocal effectors, combine to produce the song that is heard.
