This code accompanies the following paper:

Delos Reyes, R., Lyons Keenan, H., & Zachreson, C. (2025). A microsimulation model of behaviour change calibrated to reversal learning data. Journal of Artificial Societies & Social Simulation, 28(3). https://doi.org/10.18564/jasss.5637

The experimental data used for calibration and validation can be found in the data folder:

• data/target_data_single.csv: preprocessed single reversal learning data of large-brained guppies from Buechel et al. (2018)
• data/target_data_series.csv: preprocessed serial reversal learning data of large-brained guppies from Boussard et al. (2020)

The simulation results reported in the paper can be found in the experiments folder:

• experiments/run/single/output_1_20240621_000000: contains the data for the model dynamics shown in Figure 3 (see plot_figure3.ipynb for plotting code)
• experiments/calibrate/single/output_1_20240621_000000: contains the data for the single reversal learning result shown in Figure 5 (see plot_figure4.ipynb for plotting code)
• experiments/calibrate/serial/output_1_20240621_000000: contains the data for the serial reversal learning result shown in Figure 6 (see plot_figure5.ipynb for plotting code)

Alternatively, to run your own experiments, please see the commands in the run.sh file. The configuration files in the config folder are identical to those used for our experiments with the exception of the ABC_SAMPLE_COUNT parameter, which is set to 10 rather than 1000 so that the experiment is faster to run.

Cite as:

@article{delosreyesMicrosimulationModelBehaviour2025,
  title = {A {{Microsimulation Model}} of {{Behaviour Change Calibrated}} to {{Reversal Learning Data}}},
  author = {Delos Reyes, Roben and Lyons Keenan, Hugo and Zachreson, Cameron},
  year = {2025},
  journal = {Journal of Artificial Societies and Social Simulation},
  volume = {28},
  number = {3},
  pages = {8},
  issn = {1460-7425},
  doi = {10.18564/jasss.5637}
}