Vocal imitation in parrots allows addressing of specific individuals in a dynamic communication network

Abstract

Parrots in captivity are known for their ability to vocally imitate humans and recently it has been shown that wild-living orange-fronted conures are able to immediately imitate other individuals' contact calls. The function of this exceptional ability to imitate remains unclear. However, orange-fronted conures live in fission-fusion flocks where they encounter many different individuals every day, and it is possible that their vocal imitation ability is a flexible means to address a specific individual within a flock. We tested this via playback to short-term captive wild conures. Test birds were placed together in pairs in outdoor aviaries to form simple flocks. To simulate imitation of a specific individual these pairs received playback of contact calls that primarily imitate one of the two birds. Overall, individuals that received simulated vocal imitations of its calls responded more frequently and faster than the other individual. This suggests that orange-fronted conures can use imitations of contact calls to address specific individuals of a flock. In the discussion we argue that the fission-fusion flock dynamics of many parrot species has been an important factor in evolving conures' and other parrots' exceptional ability to imitate.

Figure 1. Spectrograms (A–C) of orange-fronted conure contact calls from three different playback trials to different flocks.

In each spectrogram the first call is the playback call, the second call is a solo contact call of the target bird (imitated bird) and the third call is a solo call of the non-imitated bird. The three playback calls came from three different flocks. The numbers above the solo contact calls are their cross-correlation similarity relative to the playback call. In A and B the playback call came from a male and in C a female. The test birds in A and C were all females and in B the test birds were both males. The three parts of the contact call are marked below the first call in spectrogram C. The spectrogram has been prepared in Avisoft (FFT = 256, Overlap = 75%, Blackman window).

Figure 2. Playbacks successfully imitated the targeted birds.

Playback calls (LS mean ± SE) had a higher similarity to the solo calls of imitated than non-imitated test birds. The similarity between the individuals solo contact call and the playback was quantified by spectrographic cross-correlation.

Figure 3. Imitation affected the vocal response.

Imitated birds responded with higher call rates and shorter response latency than non-imitated birds. Contact call rates (a) and response latency (b) (LS mean ± SE) of imitated and non-imitated test birds during playback.

Figure 4. Sex-specific responses to imitation.

Males gave higher call rates (LS means ± SE) when imitated than when not imitated, whereas females did not differ in their response to imitation versus non-imitation.