So the argument that social partners varied less in attractiveness than extra-pair males cannot explain why the attractiveness of extra-pair males to other females affected the occurrence of extra-pair copulations, while the mean attractiveness of the social partner did not.
In addition, I test whether male song rate, which has been interpreted as reflecting genetic quality [11], [22], affects male attractiveness in extra-pair encounters where females are thought to seek good-gene benefits.
Interestingly, the occurrence of extra-pair copulations depended on both the individual preferences of a female and the mean attractiveness of the extra-pair male towards the seven other females of the group.
Intrinsic variation in female readiness to copulate as well as variation in the attractiveness of the extra-pair male but not the social partner decided the outcome of extra-pair encounters.
Here I address these questions by quantifying intrinsic female variation in readiness to copulate, and the relative importance of the attractiveness of the extra-pair male vs. that of the social partner in the occurrence of extra-pair copulations.
Within the individual female, this propensity seems to vary with the attractiveness of available extra-pair males, but surprisingly not with the attractiveness of her social partner.
However, the study also shows that the attractiveness of the extra-pair male has a large effect on female behaviour, which makes the female-intrinsic effect harder to detect.
Given that patterns of paternity in the wild will be influenced by many other factors such as variation in male attractiveness, availability of extra-pair males, male mate guarding behaviour, or competitiveness of male sperm [1] [4], demonstrating an effect of female personality has remained elusive.
It is possible that the threshold differences found in the current study are due to an extra step in processing the health and attractiveness judgments.
The effect of the extra-pair male's phenotype: Does a male's attractiveness (measured in the choice chamber) affect his success in obtaining extra-pair copulations, and if so, which component of his phenotype makes a male attractive?
Given the low repeatability of female responsiveness in extra-pair trials and the large effect of male attractiveness (see above), plus the fact that males did not court females at all in 15% of the extra-pair trials, clearly more than four extra-pair encounters per social pairing would have been required to obtain a repeatable copulation record.
