Decomposing social interactions

Groot, Corné de (2025): Decomposing social interactions: quantifying individual variation in social foraging behaviours. Dissertation, LMU München: Faculty of Biology

Abstract

Social interactions are an important driver for phenotypic variation, especially in traits that mediate the outcome of social interactions. This leads to social evolution when such traits affect the fitness of both the focal individual and its social partner. Social evolution is often studied in the framework of quantitative genetics as indirect genetic effects (IGEs). Here the genetic variance of a given trait in a social partner modifies the phenotypic expression in the focal individual, and vice versa. IGEs can modify the total heritable variation on which selection can act, and consequently either accelerate or slow down the rate of evolution. Classic modelling approaches used to estimate IGEs generally neglect the biological reality that social interactions are two-sided and constitute of senders and receivers. Therefore, this dissertation proposes an expanded statistical framework to estimate directionality in indirect social effects in congruence. Here “social impact” describes the mechanism where a focal individual modifies the phenotype expressed by its social partner. In contrast, “social responsiveness” describes the mechanism where a focal individual adjusts its own phenotype in response to the phenotype(s) of its social partner(s). Social impact and responsiveness represent different components of social interactions, which contribute to evolutionary responses in different manners. Social impact and responsiveness may be integrated together with the average-level behaviour of an individual as a multivariate phenotype or strategy, which can constrain the response to selection. Thus, acknowledging the distinction between social impact and responsiveness will affect the short-term evolutionary responses to selection, representing a relevant phenomenon. As a worked example, we conducted behavioural experiments in wild Norwegian house sparrows (Passer domesticus) in the winters of 2022 and 2023. These experiments were set up to estimate social impact and responsiveness in producer-scrounger interactions. Producing and scrounging are mutually exclusive prosocial or exploitative tactics to obtain resources in a social context. The producer-scrounger game is commonly modelled within evolutionary game theory, and provides us with assumptions and long-term evolutionary predictions to verify our predictions about social impact and responsiveness. We repeatedly assayed individuals in variable social groups to estimate the patterns of (co)variation of social impact and responsiveness within and across producing and scrounging behaviour. First, we estimated social impact in isolation according to a classic “variance partitioning approach”, and estimated whether indirect social effects were stable across time and whether they depress the total phenotypic variation. Second, we estimated social impact and responsiveness according to the expanded contemporary statistical framework, and estimated the patterns of (co)variation within and across traits. Our findings showed that individuals vary in their average-level behaviour, and their social impact in both producing and scrounging behaviour. Moreover, we found that individuals were consistent in these traits across years. Social impact decreased the total phenotypic variation in both producing and scrounging, slowing down the rate of evolution. According to the contemporary framework approach, we found individual variation in social impact and social responsiveness in producing behaviour, but no variation in social responsiveness for scrounging behaviours. Sparrows behaved in accordance negative frequency-dependence and invested more in the rarer tactic as a function of their social environment. We provided empirical evidence that social impact and responsiveness are differential traits, based on their patterns of (co)variance. Individuals that were more socially responsive in their producing elicited more scrounging behaviour in others. This implies that social impact and responsiveness can co-evolve as a strategy to deal with variable social environments. Furthermore, such multivariate trait structure can pose constraints on evolutionary responses for social impact and responsiveness. Overall, this dissertation provides an expanded statistical framework to decompose indirect social effects. We supplement this by providing empirical evidence that social impact and responsiveness are two distinct underlying components that govern IGEs. This underlies the complex nature of social interactions, and comes with new predictions regarding evolution in response to selection, thus greatly improving our understanding of social evolution.

• Item Type: Theses (Dissertation, LMU Munich)
• Keywords: Behaviour, Behavioural ecology, Social interactions, Social evolution, Quantitative genetics, Game theory, Interacting phenotypes, Indirect genetic effects, Producer-scrounger, Sparrow, Phenotypic variation
• Subjects: 500 Natural sciences and mathematics; 500 Natural sciences and mathematics > 570 Life sciences
• Faculties: Faculty of Biology
• Language: English
• Date of oral examination: 23. June 2025
• 1. Referee: Dingemanse, Niels
• Signature of the printed copy: 0001/UMC 32052