The role of trained winning in a broodmate dominance hierarchy

Abstract

We tested the hypothesis that stability of a linear dominance hierarchy is due to each individual being trained to a different level of general aggressiveness. This mechanism, combined with assessment of aggressiveness, is particularly likely when individual recognition is absent, as may be the case in young broods of altricial chicks. Dominance relationships in broods of blue-footed boobies involve trained winning and trained losing, and broods of three chicks form stable linear dominance hierarchies. To test the prediction that chicks of intermediate rank in broods of three express an intermediate level of general aggressiveness, we pitted intermediates against similar sized bottom rank chicks from broods of two (equivalent to bottom rank chicks from broods of three). Contrary to expectation, intermediates failed to outperform bottom rank chicks by attacking first, attacking more, or submitting less. This failure suggests that general aggressiveness of intermediate chicks is not enhanced by the experience of attacking their younger broodmates and does not determine their intermediate rank in the brood hierarchy. Differential training of general aggressiveness may well be important in linear hierarchies of broods, but in dyads that do not differ in their training effects, stable dominance may depend on responsiveness to relative size.