Sexual competition hinges on the ability to impress other conspecifics, to drive them away or attract them. In such cases, the selective environment may be hedonic or affective in nature, as it consists of the evaluations of the individuals making the decisions. This may contribute to the power of sexual selection because evaluations may range from positive to negative rather than simply from positive to neutral. Selection due to mate choice may therefore be stronger than currently appreciated. Further, change in preferred mate types can occur simply by changes (“flips”) in the evaluation of similar display features, adding to the dynamism of sexual selection as well as its strength. We tested the hypothesis of positive-to-negative behavioural responses in mate choice with a playback experiment using two treefrog species with "mirror image" structures in their advertisement and aggressive calls. Female treefrog responses ranged from approach to evasion, and the presence of an aversive stimulus tainted evaluation of an attractive stimulus. Further, females in the two species showed flips in approach/evasion of stimuli with comparable signal structure. These results suggest that hedonic evaluation may have an important role in mate choice, and showcase how mechanistic analysis can help understand evolutionary processes.

We conducted single-speaker acoustic playback trials that presented female treefrogs singly with their conspecific advertisement call and their conspecific aggressive calls, in random order. We observed the reaction of the females to the playbacks, and scored their behavior on the 5-point scale ranging from attraction to avoidance (1, 0.5, 0, -0.5, -1).

To ask whether the presence of an aggressive call influences the evaluation of an advertisement call, we conducted two-choice trials that presented the conspecific advertisement call alongside the conspecific aggressive call. We scored the reaction of the females on the 5-point scale ranging from attraction to avoidance (1, 0.5, 0, -0.5, -1). We conducted a total of four two-choice trials for each species. In one trial, the stimuli had the mean features of the advertisement and aggressive call of the respective species. In three additional trials the stimuli were the mean advertisement call of each species against modified aggressive calls, making them either longer, with a faster call rate, or with a lower dominant frequency than the mean aggressive call for each species (which also made the modified aggressive calls longer, or faster or lower in frequency than the mean advertisement calls for each species).

see TREEFROGRESPONSESCOREreadme.txt