Repeatability of mate choice in the zebra finch: consistency within and between females

Numerous studies have measured the mating preferences of female zebra finches, Taeniopygia guttata, using choice-chamber experiments, but no study has focused on how consistent individual females are in their choices and the extent to which females agree on their preferences, although these questions are of great conceptual importance. We conducted a large number of mate choice trials involving unmanipulated stimulus males, and found low but significant consistency (repeatability of time allocation by a female tested twice with the same set of males R = 0.29), and very low but significant between-female agreement (different females tested with the same set of males: R = 0.11). Although low individual consistency indicates that preferences were relatively weak or hard to measure, we found significant repeatability of individual preference functions with regard to beak colour, song rate and male aggressiveness when individual females were tested twice with different sets of males. This means that some females consistently preferred red-beaked males whereas others preferred orange-beaked males, some preferred high and others low song rates, and some preferred aggressive and others less aggressive males. Of these male traits, only song rate was positively related to average male attractiveness. Low between-female agreement did not seem to result from assortative mating for quality, because there was no repeatability of an individual female's preference for attractive or unattractive males. It is possible that disagreement follows from choice for genetic compatibility, but much of it could also result from weak preferences and little variation in male quality. (C) 2004 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.