Prenatal exposure to incubation calls affects song learning in the zebra finch

Abstract

Songbirds are important models for understanding the mechanisms and fitness consequences of imitative vocal learning. Although the effects of early-life environmental and social conditions on song learning are well-established, the impact of early sound exposure has received surprisingly little attention. Yet recent evidence hints at auditory sensitivity in songbird embryos, including in the zebra finch (Taeniopygia guttata), a classic model species for song learning. Here, we tested whether prenatal exposure to incubation calls-highly rhythmic parental vocalisations produced on the nest-affected song learning in zebra finches. Embryos were exposed in the egg to either incubation (treatment) or contact (control) calls, and after hatching were reared in a large colony. The playback treatment did not affect song complexity nor the accuracy of song copying from the social father, but instead increased learning of non-paternal song syllables. This, in turn, improved males' mounting success in mating trials. These effects may be attributable to changes in juvenile social behaviours, as playback also influenced male behaviour during mating trials. Our study provides the first experimental evidence that prenatal acoustic environment affects song learning and courtship behaviour in songbirds, thereby raising interesting questions on the role of innate versus acquired biases for vocal learning.

Figure 1

The effects of prenatal acoustic playback on song syllable learning in male zebra finches. Boxplots (with individual data overlaid) showing the effects of treatment (incubation calls, n = 13) and control (contact calls, n = 18) playback on the number of song syllables that males (a) copied from their father, (b) failed to copy from their father, and (c) copied from non-father males or improvised. Double asterisks (**) indicate significant difference between treatment and control males (GLMM: z = 2.78, p = 0.006).

Figure 2

The effects of prenatal acoustic playback and song syllable learning on males’ mounting success during no-choice mating trials. (a) Boxplot (with individual data overlaid) showing no effect of prenatal acoustic playback on mounting success during mating trials (treatment: incubation calls, n = 15; control: contact calls, n = 24). Mounting success was calculated as the probability of mounting at least once per trial, across 4 or 5 trials per individual. (b) Males whose songs contained more non-paternal syllables were more likely to mount a female during mating trials (n = 31 males; GLMM: z = 1.99, p = 0.047). Larger dots represent multiple males in the same coordinates.