Linking behavioural variance to environmental heterogeneity: Variance decomposition in surface versus cave isopod populations

Abstract A longstanding question in evolutionary biology is how within‐population phenotypic variation is maintained under natural selection. The fluctuating selection hypothesis suggests that genetic and phenotypic variation scales with fluctuations in selection over space and time. This implies that phenotypic variation might be greater in populations with fluctuating than stable environmental conditions. However, this aspect has rarely been investigated, likely because habitats with minimal fluctuations are rare. We addressed this hypothesis by comparing surface and cave populations of the freshwater isopod Asellus aquaticus species complex. Surface environments are spatiotemporally more variable than cave environments, grounding the prediction that the surface ecotype is phenotypically more variable than the cave ecotype. We conducted a longitudinal behavioural study on individuals from four surface and four cave populations, measuring their movement activity and risk‐taking in two light conditions. To account for differences in the natural light regimes between surface and cave populations, half of the individuals were first acclimated under a diurnal light regime and the other half in complete darkness. Initially (acclimation tests), their behaviour was recorded in bright and dark light conditions, respectively. Next, each individual was evaluated six times in alternating light conditions—three times in each light condition. In addition to assessing average differences, we estimated among‐ and within‐individual variation, as well as repeatability and light‐induced plasticity in behaviours, enabling a comparison of these parameters between the two ecotypes. In the dark, surface individuals were on average more active and less risk‐taking than cave individuals. As predicted, the surface compared with the cave ecotype displayed greater among‐ and within‐individual variance in movement activity, but not in risk‐taking. Repeatability was not significantly different between ecotypes. Both ecotypes showed significant among‐individual variation for light‐induced plasticity in both behaviours, however, plasticity did not differ between ecotypes. Our results also suggest that more active or risk‐taking individuals exhibited greater plasticity. Our findings support the hypothesis that fluctuating selection plays a role in maintaining variation for movement activity but not for risk‐taking. Read the free Plain Language Summary for this article on the Journal blog.