Four European starlings (Sturnus vulgaris) discriminated a 300-ms segment of starling song from a 300-ms segment of budgerigar song in three contexts in a two-choice key-peck operant discrimination task. In the starling-song context, on each trial, one of the song segments was presented in the context of starling song; in the budgerigar-song context, the segments were presented in the context of budgerigar song. In the no-song context, the song segments were presented outside of a song context. On occasional unreinforced probe trials, the song segments were replaced by either white noise or silence. On noise trials in the two song contexts, but not on silent trials in the song contexts or noise trials in the no-song context, the starlings were significantly more likely to respond to the key associated with the starling song segment than to the key associated with the budgerigar song segment. This effect was especially pronounced in the starling-song context. The results indicate that noise induces perceptual restoration of missing birdsong segments in songbirds.

No abstract