Males reversed their initial preference for larger females in the presence of a conspecific audience male because they recognized the audience male as a competitor and tried to deceive that male about their real mating preference

Test of the Deception Hypothesis in Atlantic Mollies Poecilia mexicana—Does the Audience Copy a Pretended Mate Choice of Others? Klaudia Witte et al. MDPI.com, http://www.mdpi.com/2079-7737/7/3/40

Abstract: Animals often use public information for mate-choice decisions by observing conspecifics as they choose their mates and then copying this witnessed decision.  When the copier, however, is detected by the choosing individual, the latter often alters its behavior and spends more time with the previously non-preferred mate. This behavioral change is called the audience effect. The deception hypothesis states that the choosing individual changes its behavior to distract the audience from the preferred mate. The deception hypothesis, however, only applies if the audience indeed copies the pretended mate choice of the observed individual. So far, this necessary prerequisite has never been tested. We investigated in Atlantic molly males and females whether, first, focal fish show an audience effect, i.e., alter their mate choices in the presence of an audience fish, and second, whether audience fish copy the mate choice of the focal fish they had just witnessed. We found evidence that male and female Atlantic mollies copy the pretended mate choice of same-sex focal fish. Therefore, a necessary requirement of the deception hypothesis is fulfilled. Our results show that public information use in the context of mate choice can be costly.

Keywords: sexual selection; public information; male mate choice; female mate choice; audience effect; mate-choice copying; social learning; eavesdropping; Atlantic molly; Poecilia mexicana