Tutorials#
The tutorials are a learning path through the core workflow of computational modelling with cpm. Work through them in order: each one builds on the previous. If you have never used cpm before, do the cpm in five minutes first.
Beginner 30 min
Build a reinforcement learning model of a two-armed bandit task from cpm components, fit it to data from many participants, and simulate new behaviour from the fitted parameters.
Beginner 20 min
Check whether the parameters of a model can be estimated reliably: simulate data from known parameters, fit the model, and compare.
Intermediate 30 min
Check whether competing models can be told apart, compare them with penalised likelihoods, and map the landscape of their predictions.
Advanced 40 min
Estimate group-level priors from the data and use them to regularise the parameter estimates of each participant.
Advanced 40 min
Estimate the group distribution of the parameters together with its uncertainty, and test group-level means.