BCI Toolbox¶
Bayesian Causal Inference for multisensory research
BCI Toolbox is a Python package and graphical workflow for fitting, simulating, visualizing, and exporting Bayesian causal inference models for behavioral data. It is designed for researchers who want a reproducible model pipeline without writing custom analysis code for every experiment.
What You Can Do¶
Fit behavioral datasets with discrete or continuous responses.
Compare model-averaging, model-selection, and probability-matching decision strategies.
Simulate one-dimensional and two-dimensional BCI predictions.
Use the graphical interface for import, fitting, plotting, and export.
Use the Python API for reproducible scripts and advanced workflows.
Recommended Reading Path¶
Start with Installation, then open Graphical User Interface for the standard GUI. If your experiment contains two task dimensions, such as numerosity and time, continue with 2D GUI Tutorial.
User Guide
Citation¶
If BCI Toolbox supports your work, please cite:
Zhu, H., Beierholm, U., & Shams, L. (2024). BCI Toolbox: An open-source python package for the Bayesian causal inference model. PLOS Computational Biology, 20(7), e1011791. https://doi.org/10.1371/journal.pcbi.1011791
For the 2D BCI module, please also cite:
Zhu, H., Zhang, Y., Beierholm, U., & Shams, L. (2026). Crossmodal interaction of flashes and beeps across time and number follows Bayesian causal inference. Psychonomic Bulletin & Review, 33, 58. https://doi.org/10.3758/s13423-026-02857-z
Contributors¶
Haocheng Zhu, Dr. Ulrik R. Beierholm, and Dr. Ladan Shams.
Questions and feedback are welcome at evanszhu2001@gmail.com.