Hormonal Responses to Noncontact Aggression in Convict Cichlid Fish

This study explored whether convict cichlid fish mount a hormonal response to aggressive encounters where dominance status remains unresolved. Hormone samples were collected at two time points before an aggressive interaction to obtain confinement-induced and baseline measures, and at one time point following a contest across a clear partition (experimental) or exposure to an opaque partition with an opponent on the opposite side (control). There was no overall significant effect of treatment (control vs. experimental) on hormone release rates but there were trends for cortisol and testosterone (T). A priori linear contrasts showed that individuals that engaged in aggressive interactions had lower postfight cortisol and T release rates than controls, suggesting that aggression, in this context, might attenuate the synthesis of both hormones. Cortisol decreased significantly between initial confinement and baseline, indicating that individuals habituate to the water-borne hormone collection procedure. Contrary to expectation, individuals with higher baseline T and 11-ketotestosterone (KT) release rates took longer to initiate conflict. None of the other measures of behavior were predicted by baseline hormone release rates, and contest behavior did not predict postfight hormone release rates. There was a significant positive relationship between KT and T at all time points. As with studies that employ mirror image stimulation, we found no hormonal response to unresolved contests despite high levels of aggressive behavior. Our study is unique because we demonstrate that animals engaged in conflict with live opponents also do not mount a significant hormonal response when clear dominance relationships are not established.