Aggressiveness and size: a model and two tests

Individual variation in aggressive behavior in animals might be caused by adaptive covariation with body size. We developed a model that predicts the benefits of aggressiveness as a function of body size. The model indicated that individuals of intermediate sizes would derive the greatest benefits from being aggressive. If we assume that the cost of aggression is approximately uniform with respect to body size, selection should favor higher aggression in intermediate-sized individuals than in large or small individuals. This prediction was tested by stimulating male Madagascar hissing cockroaches, Gromphadorhina portentosa, with disembodied antennae and recording the males' aggressive responses. Antennae from larger males evoked weaker responses in subjects, suggesting that males obtained information about their opponents' size from the opponents' antennae alone. After accounting for this effect, we found support for the key prediction of our model: aggressiveness peaked at intermediate sizes. Data from actual male-male interactions validated that the antenna assay accurately measured aggressiveness. Analysis of an independent data set generated by staging male-male interactions also supported the prediction that intermediate-sized males were most aggressive. We conclude that adaptive covariation between body size and aggressiveness explains some interindividual variation in aggressiveness.