1. Although social learning capabilities are taxonomically widespread, demonstrating that freely interacting animals (whether wild or captive) rely on social learning has proved remarkably challenging.

2. Network-based diffusion analysis (NBDA) offers a means for detecting social learning using observational data on freely interacting groups. Its core assumption is that if a target behaviour is socially transmitted, then its spread should follow the connections in a social network that reflects social learning opportunities.

3. Here, we provide a comprehensive guide for using NBDA. We first introduce its underlying mathematical framework and present the types of questions that NBDA can address. We then guide researchers through the process of: selecting an appropriate social network for their research question; determining which NBDA variant should be used; and incorporating other variables that may impact asocial and social learning. Finally, we discuss how to interpret an NBDA model’s output and provide practical recommendations for model selection.

4. Throughout, we highlight extensions to the basic NBDA framework, including incorporation of dynamic networks to capture changes in social relationships during a diffusion and using a multi-network NBDA to estimate information flow across multiple types of social relationship.

5. Alongside this information, we provide worked examples and tutorials demonstrating how to perform analyses using the newly developed NBDA package written in the R programming language.

We provide tutorials guiding users through several examples illustrating how to carry out network-based diffusion analysis using the NBDA package (https://github.com/whoppitt/NBDA). These tutorials make use of simulated data in the form of social networks and individual-level data (e.g. sex and age), which are provided here

In addition, the NBDA code and data necessary to replicate the results presented in Box 3 in the main text are also included. These data were collected as part of a larger study examining the relative importance of different social network types in guiding honeybees to novel foraging locations. Two cohorts of honeybees originating from a single colony were simultaneously trained to separate artificial sugar water feeders 100 m from the hive. During the trial, one of these feeders was left empty, while the other continued to provide sucrose. The order in which individuals trained to the former feeder discovered the latter feeder (which they had never previously visited) was recorded. At the same time, all interactions in the hive between honeybees visiting the active feeder and those that had been trained to the now-empty feeder (but had yet to discover the active feeder) were filmed and recorded. For each dance-following interaction, the number of waggle runs an individual followed for the active feeder was recorded. For trophallaxis and antennation, the duration of each interaction was recorded in seconds.

From these interaction records, dynamic and static social networks were constructed. Static networks for each interaction type aggregated all interactions that occurred between each pair across the entire 2 hr trial. In contrast, dynamic networks updated throughout the trial. For each successful recruitment event, networks updated when that individual left the hive. In one instance, a recruit left the hive, but did not discover the feeder until after another recruit had discovered it; to prevent networks from "rewinding", the update time for the latter individual was used for the former. There were 16 recruitment events in total, meaning networks were updated a total of 15 times. The index numbers indicate the successive updates.

Annotated tutorials and example code are provided describing the use of these data.