This repo contains the data and code required to reproduce the main figures presented in Cortese et al., 2021. "Value signals guide abstraction during learning". eLife. Available at https://elifesciences.org/articles/68943
Summarised data and code to reproduce the main figures are available in the folder 'Figures'
Data and code for the reinforcement modelling -- both using simple feature RL and Abstract RL, as well as for the mixture-of-experts architecture, are available in the folder 'RL-modelling'
The stimuli, code and a sample task file are contained in the folder 'TASK'
The HBI toolbox (Piray et al., 2019) was used for hierarchical fitting of reinforcement learning algorithms. The toolbox is available at https://github.com/payampiray/cbm.
Reference:
Piray, Payam, Amir Dezfouli, Tom Heskes, Michael J. Frank, and Nathaniel D. Daw. 2019. “Hierarchical Bayesian Inference for Concurrent Model Fitting and Comparison for Group Studies.” PLoS Computational Biology 15 (6): e1007043.