Sexual selection can both increase and decrease extinction probability: reconciling demographic and evolutionary factors

1. Previous theoretical models of the effect of sexual selection on average individual fitness in a population have mostly predicted that sexually selected populations should adapt faster and clear deleterious mutations more quickly than populations where sexual selection is not operating. 2. While some laboratory studies have supported these predictions, others have not and studies of field systems have tended to find negative effects of sexual selection, or no effect. The negative effects of sexual selection found in field and other studies are usually ascribed to the costs associated with strong sexual selection acting on the population. 3. Here, using an individual-based model that allows feedback between demographic and evolutionary processes, we find that sexual selection can lead to both increases and decreases in population-level fitness measures such as extinction probability and adaptation rate. Whether fitness increases or decreases depends on a variety of environmental and demographic factors including the nature of environmental change, the carrying capacity of the environment, the average fecundity of the population in question and the strength of condition dependence. 4. In many cases, our model predicts that sexual selection leads to higher extinction probability in small populations because of an increased risk of demographic stochasticity, but lower extinction probability in larger populations because of faster adaptation rates. This is consistent with field studies that have mostly focussed on very small populations such as recently introduced birds, and tend to find negative effects, and also with laboratory studies that tend to use larger populations and have tended to find positive effects. 5. These findings go at least some way towards an understanding of the apparent contradictions between theoretical predictions, laboratory studies and field data.