Ontogeny shapes individual specialization

Abstract

Individual dietary specialization, where individuals occupy a subset of a population’s wider dietary niche, is of key importance for species’ resilience against environmental change. However, the ontogeny of individual specialization, as well as associated underlying social learning, genetic, and environmental drivers remain poorly understood. Using a multigenerational dataset of female European brown bears (Ursus arctos) followed since birth, we discerned the relative contributions of social learning, genetic predisposition, environmental forcings, and maternal effects to individual dietary specialization. Individual specialization varied from omnivorous to carnivorous diets spanning half a trophic position. The main determinants of this dietary specialization were maternal learning during rearing (13%), environmental similarity (12%), maternal effects (11%), and permanent individual effects (8%), whereas the contribution of genetic heritability was negligible. Importantly, the offspring’s trophic position closely resembled the trophic position of their mothers during the first 3-4 years after separation from the mother, but this relationship ceased with increasing time since separation. Our study reveals that social learning and maternal effects are as important for individual dietary specialization as environmental forcings. We propose a tighter integration of social effects into future studies of range expansion and habitat selection under global change that, to date, are mostly explained by environmental drivers.

Variation in personality shaped by evolutionary history, genotype and developmental plasticity in response to feeding modalities in the Arctic charr

“…We calculated marginal (R2m), conditional (R2c) and inclusive (IR2) R 2 -values for model.0, describing the amount of variance in boldness scores explained by the fixed predictors, the whole model and each predictor, respectively (method adapted from [67], details in accompanying R script [63]). We applied the functions emmeans and contrast ( package emmeans v. 1.7.8 [68]) under Tukey linear hypothesis to model.0, to perform pairwise post hoc comparisons of boldness scores between morphs and between treatments within morphs.…”