Scientific Software
We have published several software packages.
Cannon Task
The Cannon Task is a MATLAB implementation of the canonical predictive inference paradigm widely used to study how humans learn and update beliefs under uncertainty. It provides a flexible framework for manipulating environmental changes, outcome noise, and feedback visibility, allowing researchers to probe distinct components of adaptive learning. This version is actively used in the DFG Research Unit 5389 to generate high-quality behavioral data for computational modeling. The codebase includes ready-to-run experiment scripts, trial-generation utilities, and clean output structures optimized for downstream analysis in Python or MATLAB. It is designed to be both experimenter-friendly and modeler-friendly, ensuring tight integration with Bayesian models of belief updating.
Reduced Bayesian Model (RBM) Python Package (rbmpy)
A Python library for simulating, fitting, and analyzing the Reduced Bayesian Model (RBM) of adaptive learning, with clean APIs for inference, likelihood computation, and regression-based parameter analysis.
All-In Python Package (allinpy)
Lightweight Python utilities for plotting, data handling, and model evaluation used across RBM and Gabor-Bandit analysis pipelines.
< Go back to the landing page to find out more about the group >