Quantitative assessment of song-selectivity in the zebra finch “high vocal center”

Abstract1.Auditory responses in the zebra finch (Taenopygia guttata) song-system nucleus HVc were assessed at 54 recording sites by 3 different methods: discriminated action potentials; excitatory summed responses; and excitatory minus inhibitory summed responses. Four standard stimuli were presented at each site: the bird's own song; this song reversed; a conspecific song; and a noise burst. Responses were quantified by calculating a relative response index that partitioned the response, to provide a response profile, across the stimuli.2.Regardless of analysis method, the strongest response was most often to the bird's own song (78–82%, depending on method). The predominant rank order of response strength across the remaining three stimuli was conspecific song > reversed song > noise.3.The distribution of relative response magnitude was sensitive to analysis method. Discriminated spikes captured the heterogeneity of HVc neurons, whereas the excitatory summed responses reflected the overall trends more consistently. When inhibition was subtracted from excitation in the summed responses, the variance of the relative responses increased, but this method presented some problems for statistical analysis.4.A small sample of neurons in other forebrain auditory areas was used for comparative analyses. At these recording sites, the bird's own song did not consistently elicit the best response and there were generally smaller differences in the relative responses to the four stimuli. The smaller degree of stimulus selectivity among these cells resulted in less sensitivity to differences in the assessment methods.