Testosterone facilitates some conspecific song discriminations in castrated zebra finches (Taeniopygia guttata).

An experiment was designed to test for the influence of testosterone on song discriminations. We found that testosterone did have an effect, which interacted with practice and the nature of the stimuli. Fourteen adult castrated zebra finches (Taeniopygia guttata) were grouped into seven pairs. In each pair, one bird was implanted with a testosterone-filled silastic tube and the other was implanted with an empty silastic tube. They were then trained on a go/no-go operant task to discriminate between bird songs in six consecutive experiments. The songs to be discriminated were as follows: experiment 1, two canary song segments; experiment 2, the bird's own song and that of the other member of the pair; experiment 3, the same two songs as in experiment 2 but with reversed stimulus-response contingencies; experiment 4, two other zebra finch songs; experiment 5, another two zebra finch songs; and experiment 6, another two canary song segments. There were no reliable learning differences between birds treated with testosterone or with an empty silastic in experiments 1 and 3-6. However, in experiment 2, testosterone-treated birds mastered the discrimination between their own song and the song of the other member of the pair in fewer trials than birds treated with empty silastics. We suggest that a song's ability to control the behavior of male zebra finches is influenced by the nature of the song, prior experience with the training paradigm, and hormone levels.