High repeatability of egg rejection in response to experimental brood parasitism in the American robin (Turdus migratorius)

Abstract

Repeatability is a measure of the amount of variation in a phenotype that is attributable to differences between individuals. This concept is important for any study of behaviour, as all traits of evolutionary interest must be repeatable in order to respond to selection. We investigated the repeatability of behavioural responses to experimental brood parasitism in American robins, a robust (100%) rejecter of parasitic brown-headed cowbird eggs. Because tests of repeatability require variation between individuals, we parasitized the same robin nests twice successively with model eggs dyed with colours known to elicit rejection at intermediate rates (58–70%). We calculated the repeatability of responses to parasitism, and used a generalized linear mixed model to also test for potentially confounding effects of ordinal date, presentation order, and clutch size. We found that repeatability in response to brood parasitism in this host species is high, and the best model predicting responses to sequential artificial parasitism includes only nest identity. This result is consistent with a critical assumption about egg rejection in this cowbird host as an evolved adaptation in response to brood parasitism.