Delphine De Moor, Lauren J. N. Brent, Matthew Silk , Josefine Brask

Animal social systems are remarkably diverse. Linking this diversity to its ecological and evolutionary drivers and consequences has been a major focus of biological research. Initial efforts have been done within groups, populations, and species. Equipped with this information, researchers are now turning to investigations of social structure that are comparative in nature. However, comparing social networks remains a considerable logistical and analytical challenge. Here we present the ‘layers of latency framework’, a conceptual framework that helps researchers to uncover and study the latent social structures that are of interest to them. We then use this conceptual framework to examine how we can tackle some of the central challenges to comparing animal social networks, focusing on differences between networks in behaviour type, sampling type, sampling effort, sampling scale and network size. For each of these focus points, we describe why and how they create challenges for comparative analyses, and we suggest potential directions for solutions. The layers of latency framework can help researchers to identify networks and features they can (or cannot) compare. In doing so, this framework facilitates advances in cross-species social network studies with the potential to generate new and important insights into the ecological and evolutionary drivers of variation in social structure across the animal kingdom.