Darwin was the first to propose that female mating preferences can result in selection on male morphology and behaviour. Since then, particularly over the last two decades, many studies have confirmed that individuals prefer some opposite-sex trait variants over others and that potencial mates with preferred traits have enhanced mating success .
Sexual selection by mate choice can therefore be important in the evolution of the opposite-sex secondary sexual traits. More recently, attention has focused on the factors that influence the evolution of these preferences.
Three general experimental approaches must be adopted to study the evolution of mating preferences:
1) A comparative approach can be used to examine historical associations between phenothypic traits, opposite-sex preferences, and environmental factors. For example, statistical associations across populations between preferences and traits have been used to evaluate whether female preferences and male traits have co-evolved.
2) The benefits of an average preference of a population can be examined. For example, in populations where individuals tend to prefer one trait variant over another, the potential benefits of this preference can be evaluated by examining the fitness consequence of mating with mates with different trait values.
3) Selection on mating preferences can be measured directly by assessing the relationship between the strength of a preference and fecundity and survivorship.
For some questions about the evolution of female preferences, a comparative approach is the most appropriate method . When examining the adaptive significance of female preferences, however, historical associations between female preferences and other variables are merely suggestive; such correlations can arise because of causal relationships or because the preference and the other variable of interest are both correlated with some other variable.
In contrast, examinations of the adaptive significance of population-level preferences can help to determine whether selection currently acts on mating preferences. There are, however, at least two important drawbacks with this approach.
a) First, there may be adaptive variation in preferences; some females might benefit from having preferences different from the population average. Under such circumstances, examining the benefits of population-level preferences may lead to a misleading view of the factors influencing the evolution of preferences.
b) Second, selection on preferences is likely to be multifactorial; more than one direct source of selection is likely to act on preference functions, and indirect selection can act on preferences due to phenotypic correlations with other traits under selection. As a result, the examination of a single benefit of a population-level preference may provide an incomplete view of the factors influencing the evolution of preferences.
An alternative approach is to measure how selection acts on mating preferences within natural populations. Such studies would allow us to assess the fitness consequences of variation in preferences, the relative importance of different sources of selection, and whether correlations with other traits result in indirect selection on preferences.
For example, we could assess the relative importance of direct selection on preferences due to variation in survivorship and fecundity and indirect selection on preferences due to variation in offspring viability.
No studies have attempted to assess how selection acts on mating preference functions in humans within natural populations.
To be continue.