Why care? Inferring the evolution of complex social behaviour

birds; character evolution; parental care; parental investment; phylogenetic comparative methods; sex role. Phylogenetic comparative analyses of complex traits often reduce the traits of interests into a single (or a few) component variables. Here, we show that this may be an over-simplification, because components of a complex trait may evolve independently from each other. Using eight components of parental care in 400 bird species from 89 avian families that represent the relative contribution of male vs. female to a particular type of care, we show that some components evolve in a highly correlated manner, whereas others exhibit low (or no) phylogenetic correlation. Correlations were stronger within types of parental activity (brooding, feeding, guarding) than within stages of the breeding cycle (incubation, prefledging care, post-fledging care). A phylogenetically corrected cluster analysis identified two groups of parental care components that evolved in a correlated fashion: one group included incubation and brooding, whereas the other group comprised of the remaining components. The two groups of components provide working hypotheses for follow-up studies to test the underlying genetic, develop-mental and ecological co-evolutionary mechanism between male and female care. Furthermore, the components within each group are expected to respond consistently to different ambient and social environments