**All paper figures**
* *PAPER_FIGURES.m*
* Requires folder 'funs' and files in task folders
****
**Run models**
* *HBayes_RunModelsParallel.R* – script to run all models.
* needs an hBayesDM model file (‘*./BanditTask/Anahit_Hbayes_lapseS1.R*’ = example)
* needs a text file with data (e.g., '*All_Bandit.txt*' in ./BanditTask
* *RUN_testretest_embedded_corr_models_all.R* - runs all hierarchical Bayesian joint models with embedded correlation matrix
* *./BanditTask/bandit4arm_lapse_testretest.stan* - bandit model
* *./GambleTask/ra_prospect_testretest.stan* - gamble model
* *MixedModels_all.R* - runs all mixed models and saves individual predicted values:
* ./BanditTask/Bandit_mixedRegression_S1_S2_data.csv
* ./GambleTask/Gambling_mixedRegression_S1_S2_data
**Other files**
* *preprocessSIMULATEDdata.m* – preprocess and saves excel file of simulated and raw data together
* ./BanditTask/Preproc_Bandit_SIMnREALrawData.xls
* ./GambleTask/Preproc_Gamble_SIMnREALrawData.xls
* *ModelProcessing.Rmd* - model comparison and extracts and saves mean parameter estimates of models. Saves files (below are examples):
* ./BanditTask/params_outputS1_lapse_47subs.csv/
* ./GambleTask/params_outputS1_prospect_18Jan20.csv
***
**./BanditTask/ scripts & folders**
* 'hBayesOUTPUT' folder - all hBayesDM objects (model data) for bandit task
* 'session1/2' folder - raw subject data (anonymised) .mat files
* *preprocessBandit.m* - preprocesses raw data into model-agnostic measures into excel file:
* ./BanditTask/Preproc_Bandit.xls
* *preprocess4modelling_Bandit.m* - creates .txt files of trial-by-trial data info for modelling
* example: ./BanditTask/S1_Bandit.txt
* *Likelihood_Predict_bandit4arm_lapse.R* – calculate predictive probability. Saves files:
* ./BanditTask/S1PredictS1_bandit4_Lapse47.csv
* ./BanditTask/S2PredictS2_bandit4_Lapse47.csv
* ./BanditTask/S1PredictS2_bandit4_Lapse47.csv
* *Likelihood_PredictOTHERS_bandit4arm_lapse.R* – calculate average of future predictive probability of all but current subject. Saves files:
* ./BanditTask/S1_OTHERparamsPredict_S2_bandit4_Lapse47subs47.csv
* *Likelihood_S1priorMeanPredictS2_bandit4arm_lapse.R* - calculate S1 group prior parameters in predicting S2 data compared with subject's own S1 parameters. Saves:
* ./BanditTask/S1PRIORS_PredictS2_bandit4_Lapse47.csv
* *simulate_bandit4arm_lapse.R* – create & save artificial data based on model parameters. Saves file (example):
* ./BanditTask/S1_bandit4_Lapse_SimData_47subs47.txt
* use this file to 1) fit lapse model & then extract parameter estimates on the simulated data for parameter recovery checks and 2) check simulated versus real data. Currently need to manually adjust headers in txt file
**./GambleTask/ scripts & folders**
* same ones as in ./BanditTask/ but called 'prospect' or 'gamble' instead of 'bandit'
