Contents

Processed data files are located in the data folder. They are saved as .fst files, which can be read using the fst R package (see https://www.fstpackage.org/).

Analysis scripts (R / Rmd) are located in the analysis directory. They are numbered in the intended order of execution.

Main scripts:

• 01_collect_data.R reads in the raw CSV files and creates usable data frames for the rest of the analysis.
• 02_calculate_model_estimates.R reconstructs the relevant parameter values of the adaptive fact learning model (incl. rate of forgetting).
• 03_descriptive_stats_session1.Rmd is a notebook with descriptives of performance in the first experimental session.
• 04_alpha_analysis.Rmd is a notebook analysing the rate of forgetting estimates obtained in the first session.
• 05_generate_fact_predictions.R makes fact- and domain-level predictions of rate of forgetting to be used in the second session.
• 06_plot_examples.R creates some visualisations of session 1 rate of forgetting estimates and model predictions.
• 07_make_session2_predictions.R uses the data from session 1 and the first block of session 2 to reconstruct the predictions used in the second block of session 2.
• 08_visualise_predictions.R plots the prediction matrix and map shown on the poster.
• 09_session2_H1_learning_outcomes.Rmd is a notebook testing Hypothesis 1 (improved learning outcomes).
• 10_session2_H2_prediction_accuracy_1.Rmd is a notebook testing Hypothesis 2 (more accurate predictions when combining learner and fact information).
• 11_session2_H3_prediction_accuracy_2.Rmd is a notebook testing Hypothesis 3 (more accurate predictions when predicting individual facts and/or learners instead of making a domain-level prediction).

Supporting scripts:

• bayes_funs.R implements the Bayesian model functions.
• geom_flat_violin.R implements the half-violin plot that is used in some of the figures.
• slimstampen_model_funs.R implements the adaptive fact learning model functions.

Figures generated by the analysis scripts can be found in the output folder.

The stimuli used in both sessions are located in the stimuli directory.

OSF project page

https://osf.io/snfyz/