| dc.description.abstract | Social learning allows individuals to acquire beneficial information through observing or interacting with others. A consequence of social learning for group-living animals is that it can facilitate the spread of novel behaviours, individual-to-individual, across the group (social transmission). Following the development of sophisticated statistical techniques numerous studies have shown behaviours to spread via social transmission across numerous species. The speed and pattern of spread appears to be contingent on multiple factors including individual attributes of the model and the learner, the social structure and dynamics of the group and the physical environment that the group inhabits. However, the majority of these studies investigated behavioural spread across single, natural wild groups making it hard to disentangle the relative contributions of each factor. My thesis has taken an experimental, replicated approach to explore how factors of the individual, social environment and environmental conditions can affect social learning and transmission of behaviours. I do so in a way that independently manipulates one aspect of the physical or social environment whilst controlling for other additional factors. I used groups of both pheasant chicks (Phasianus colchicus) and domestic fowl chicks (Gallus gallus domesticus) to explore the effects of these factors on social learning and behavioural transmission. Firstly, I looked at factors of the individual - I found individuals to vary greatly in social learning performance on a novel foraging task but their sex, mass, dominance ranking and social position did not explain this individual variation (Chapter 2). I then explored factors of the social environment through manipulating group size (Chapter 3), group structure (Chapter 4) and social foraging dynamics (Chapter 6) to determine their effect on social transmission. Increasing group size during interaction with a novel foraging task did not result in faster behavioural spread as predicted and thus indicates that total numbers of connections to informed individuals may not always best describe the likelihood of social learning (Chapter 3). Similarly, increasing the modularity of a population’s network (how clustered it is) did not limit behavioural spread as expected or lead to the establishment of specific behavioural variants within clusters (Chapter 4). These two results indicate that social transmission may follow more complex rules that incorporate proportions of connections to informed individuals, social reinforcement and forgetting rates that influence social learning and transmission of behaviours. By manipulating the social environment, I found scrounging opportunity to greatly facilitate the spread of a novel foraging behaviour (Chapter 5). I reason that this is likely due to the particular social learning mechanism (local/stimulus enhancement) deployed to acquire the behaviour. Lastly, making the group’s environment less predictable did not (contrary to predictions) affect how individuals used social information in a separate context (Chapter 6), suggesting that reliance on social information is not generalised across contexts. Taken together this thesis challenges the assumptions and predictions of several current theoretical models about the use of and processes underlying social transmission of behaviours, and clearly reveals a number of factors that appear critical in shaping the use and outcomes of social learning. | en_GB |
