This repository is for code analyzing and simulating reinforcement learning models for a cognitive psychology task called the 2-step task. The shcripts folder contains scripts to parallelize the analyses. shcripts/run_all_jobs.sh creates an array of jobs to a cluster, with each job calling shcripts/run_one_job.sh to implement that job by hooking into py_scripts/run_wrapper.py. This python script receives a number as an input from the shell script and retrieves the corresponding row number in a csv of parameters, which is generated by prep_params_for_cluster.py. py_scripts/run_wrapper.py then inputs the row's parameters into py_scripts/wrapper.py, which in turn calls py_scripts/simulate_and_model.py to carry out the reinforcement learning simulations and model fitting.