socialSim

Simulate and Analyse Social Interaction Models

The socialSim R package provides tools to simulate and analyse datasets of social interactions between individuals using hierarchical Bayesian models implemented in Stan.

It enables users to generate realistic social interaction data, where individual phenotypes influence and respond to those of their partners. You can simulate a sampling design by adjusting the number of individuals, partners, and repeated dyads. The simulation framework allows control over variation in mean trait values, social responsiveness, and social impact and correlation, making it suitable for research on direct and indirect genetic effects (DGEs and IGEs) and interacting phenotypes. See ?simulate_data for a full list of adjustable parameters.

The package also provides analysis functions to evaluate model performance in terms of bias and dispersion, using both established and novel approaches to modelling social effects, including impact–responsiveness, variance–partitioning, and trait-based models.

🧭 Installation

You can install the development version from GitHub using:

# install.packages("remotes")
remotes::install_github("RoriWijnhorst/socialSim")
# Then load the package:
library(socialSim)

⚙️ Example workflow:

library(socialSim)

# 1. Simulate data. See ?simulate_data for all adjustable parameters
sim <- simulate_data(
  ind = 1200,           # number of unique focal individuals
  partners = 4,         # number of social partners per individual
  repeats = 1,          # number of repeats of dyads   
  iterations = 10,      # number of datasets created    
  B_0 = 1,              # population intercept
  psi = 0.3,            # population-level response
  Valpha = 0.2,         # variance in direct effects
  Vepsilon = 0.1        # variance in residual partner effects
)

# 2. Fit a Stan model. For the analyses, cmdstanr or rstan needs to be installed.
# using cmdstanr will be faster, since it runs model in parallel on different cores of your PC.
res <- run_model(sim, model = "Trait.stan", iter=2000, cores = 6)

# 3. Summarise results
summary <- summarise_results(res)
print(summary)

🧪 Available IGE models