README.md

Code for the Cognitive Flexibility in Aging Paper

Grahek, I., Leng, X., Fengler, A., & Shenhav, A. (2025) Slower transitions between control states lead to reductions in cognitive flexibility over the lifespan. bioRxiv. https://www.biorxiv.org/content/10.1101/2025.08.27.672689v1.abstract

Contents

---

1. analyses Directory

Contains scripts and models for response time, accuracy analyses, and Drift Diffusion Model (DDM) analyses.

1.1. brms Folder

Purpose: Scripts related to Bayesian regression models using brms.

• Cluster/:

  • bash/: Job submission scripts for cluster computing.
  • models/: Scripts for fitting specific models.
  • output/: Model output files (currently empty; limited by GitHub file size).

• CAC_Aging_Brms.Rmd:
  R Markdown file for analyzing brms models and generating figures.

---

1.2. hssm Folder

Purpose: Scripts to fit Drift Diffusion Models using the HSSM package.

• analysis/:

  • analyze_models.ipynb: Jupyter notebook for analyzing HSSM models and saving posteriors.
  • hssm_analysis_helpers.py: Helper functions used in the Jupyter notebook.
  • DDM_analyses_and_plots.Rmd: R Markdown file for DDM model analysis and figure creation.
  • .csv files: Posterior samples generated from the notebook, used in Markdown analyses.
  • .rds files: Regression models derived from Markdown analyses.

• data/:

  • .csv files: Preprocessed data used to fit models.

• models/:

  • DDM models fitted to the data.

• output/:

  • Model files (currently empty due to size limitations).

---

2. data Directory

Contains raw and preprocessed behavioral data.

• ProcessedData/:

  • .csv file with preprocessed data.

• RawData/:

  • Contains raw data files (not uploaded due demographic information).

---

3. preprocessing Directory

Purpose: Scripts for preprocessing raw data into format suitable for analysis.

• Preprocessing.R: Main script for data preprocessing.
• PreprocessingSummary.csv: Summary of the excluded participants and total participant numbers after preprocessing.

---