Skip to main content
eScholarship
Open Access Publications from the University of California

Contributions of premotor nucleus RA for the spectral and temporal structure of song in the Bengalese Finch

  • Author(s): Cheung, Chung Yan Joanne
  • Advisor(s): Brainard, Michael
  • et al.
Abstract

Song is a learned motor behavior where each vocal element (syllable) has its own distinct acoustic structure. Syllable features, like pitch and amplitude, have been linked to the activity of the forebrain nucleus RA; but whether these features are directly controlled by the level of RA activity remains to be shown. Natural variation in RA activity may be produced by a dynamic interaction between RA projection neurons and interneurons; but a role for local inhibition in RA for shaping vocal output is also unknown. Here, we address whether changing RA activity by modulating the influence of RA inhibition has a direct effect on syllable structure. We find that increasing or decreasing RA inhibition drives a corresponding increase and decrease in syllable pitch and amplitude. These results provide evidence that local inhibition within RA actively contributes to song production and suggest that maintaining pitch and amplitude at a desired level requires a careful balance of excitation and inhibition acting within RA.

Song also has a specific temporal structure that is linked to activity in the forebrain nucleus HVC. Previous studies have found that manipulating the temporal dynamics of HVC activity drives changes to song timing and sequencing. This supports a hierarchical model of song control in which temporal and spectral properties of song (such as syllable sequencing and timing versus syllable pitch and amplitude) are separately controlled by HVC and RA. However, the downstream targets of RA in the brainstem also make ascending projections back to HVC, raising the possibility that the output of RA may recurrently influence HVC activity to shape song timing and sequencing. Here, we find that natural variation in RA activity is correlated with variation in song timing; specifically, higher firing in RA is associated with shorter durations of the gaps between syllables. We also find that increasing or decreasing the influence of RA inhibition drives bidirectional changes to song tempo and syllable repetitions. These high order song effects are not dependent on auditory feedback. Taken together, these findings indicate that RA activity contributes centrally to the temporal structure of song.

Main Content
Current View