Cultural behaviour, i.e., that which is transmitted socially among conspecifics, is found in a variety of taxa, including cetaceans. Different methods have been used to detect social learning in animal populations. ‘Network-based diffusion analysis’ (NBDA), for example, provides a statistical frame-work with which the importance of social learning on the spread of a behaviour can be quantified. It infers social learning if the diffusion of behaviour follows the social network and therefore relies on accurate association data among individuals. Incomplete association data can lead to uncertainty over the strengths of connections among individuals. Restricting analyses to only include individuals above a certain threshold of sightings can minimize such uncertainty, but at the same time reduce power of NBDA to detect learning when linking individuals are removed from the network. Following my General Introduction, Chapter 2 of this thesis therefore provides a tool for researchers to select an appropriate threshold for the inclusion of individuals that maximizes the power of NBDA to detect social learning. In the study of the rise and spread of cultural behaviour, ecology and genetics are potentially confounding factors as they too can drive behavioural variation between individuals, communities and populations. I use a multi-network version of NBDA, which can account for these potential confounds by including networks reflecting association patterns, genetic relatedness and habitat use, in Chapters 3 and 4 to investigate the spread of two foraging strategies, ‘shelling’ and ‘sponging’, in a population of Indo-Pacific bottlenose dolphins (Tursiops aduncus) in the western gulf of Shark Bay, Western Australia, between 2007 and 2018. Shelling (Chapter 3) appears to spread horizontally among associated individuals, which stands in stark con-trast to the predominantly vertically transmitted foraging strategies, from mother to offspring, in Shark Bay dolphins and indeed toothed whales in general. My study provides the first quantitative evidence of horizontal transmission in any toothed whale species and suggests similarities in the cultural nature of cetaceans and great apes, which rely extensively on both vertical and horizontal social learning. Conversely, the findings presented in Chapter 4 suggest vertical social transmission of sponging from mother to primarily female offspring, confirming the results of previous research using different methods. Chapters 3 and 4 illustrate how long-term data sets on individual associations, habitat use and genetics, in combination with new statistical tools like NBDA, provide an ideal framework to assess the spread of behaviour in free-ranging animal populations. In Chapter 5, I investigate the impacts of a marine heatwave, which led to catastrophic losses of habitat-forming seagrass beds and mass mortalities of fish and invertebrates in Shark Bay, on the vital rates of the resident dolphin population. Long-term demographic data and capture-recapture analyses on data collected before and after the heatwave indicate immediate and on-going reductions in both survival and reproductive rates within the dolphin population, presumably due to the cascading effects of the heatwave on lower-trophic organisms combined with a lack of ecosystem recovery. Remarkably, survival rates of sponging dolphins appear less adversely impacted compared to those of non-spongers, suggesting that their foraging niche may have buffered them against more negative impacts. Whether or not culturally different communities within a population may respond differently to environmental change remains an exciting avenue of research in the future. Finally, I discuss the broader ramifications of this thesis in the General Discussion and suggest further directions in the study of cultural behaviour in bottlenose dolphins.
