Brain space for a learned task: strong intraspecific evidence for neural correlates of singing behavior in songbirds.

There is a controversial issue in neuroscience whether the expansion of neural network space permits the development of more complex behavior. One of the best-known model systems for studying the relationship between brain space and behavior is song production and the associated song control system in songbirds. Although the neuroanatomical background of song production is well established, the direct link between song nuclei volumes and song traits remains puzzling. Analyses within species have provided conflicting results regarding the association between song nuclei volumes and measures of song complexity and song length. Based on a meta-analysis, we present here the results of the first synthetic review, in which we test for overall intraspecific patterns in relation to bird song and the size of associated neural tissues. We found significant positive relationships between the volume of two important song nuclei (HVC and RA) and repertoire size and song length. We assessed the importance of absolute and relative volumes, and found that a control for the covariation with the telencephalon may be important. By estimating the adequate sample size that would be needed to reach sufficient statistical power in particular studies, we conclude that previous studies finding non-significant associations between song and volumes of brain nuclei were of weak power. When we factored out the covariation between song length and repertoire size, we found that these traits may explain independently significant amount of variations in the relative volume of HVC, but not of RA. The link between the volumes of song nuclei and song features has important theoretical implications with regard to the neurobiology and evolution of bird song.