Social stimuli, testosterone, and aggression in gull chicks: support for the challenge hypothesis.

We tested the challenge hypothesis for the hormonal regulation of aggression in chicks of the black-headed gull, Larus ridibundus. Chicks of this species are highly aggressive toward conspecifics, but never to peers that hatched from the same clutch (modal clutch size is three). Therefore, in the first experiment small families were housed together in large groups (challenged condition) and compared to families kept isolated (nonchallenged condition). As expected, in the challenged condition during the initial stage of territory establishment basal levels of testosterone (T) were clearly higher than those in the nonchallenged condition. In the second experiment we tested the effect of a short social challenge on short-term T-fluctuations. The design was based on an earlier experiment, showing that after temporary T-treatment chicks become very sensitive to social challenges while having low basal T-levels. We now show that these social challenges induce brief elevations in plasma T-levels. These peaks are similar to those in previously untreated chicks but untreated chicks do not respond with aggression to a challenge. Therefore, we conclude that the initial exposure to elevated T-levels increases the sensitivity to brief changes in T induced by social challenges. In this way exposure to T, that may be detrimental for development, is minimized while birds remain able to defend territories. This is the first report showing that the challenge hypothesis as established for adult birds, is also applicable for aggressive behavior in young birds outside the sexual context. Furthermore we suggest that a phase of priming with T is necessary to obtain the high behavioral responsiveness to a challenge.