[UNRAVEL]

Description

• Authors: MASKED
• Date: 2025-09-10
• E-mail: MASKED

This is a research compendium for the project UNRAVEL. The compendium is organised based on BIDS standard, but not fully compatible due to custom workflows.

The results are based on the original dataset sourced from OSF repository shared under CC-BY license and reported in the paper by Kalbe and Schwabe (2022). Additional data were obtained upon request.

Directory structure

.
├── CHANGES                    # Log of changes and updates
├── Makefile                   # Master workflow to reproduce the repository
├── README.md                  # Overview and documentation
├── code                       # Code with individual workflows for each analysis component
├── data-raw                   # Minimally preprocessed source data
├── data-source                # Source data directly downloaded from OSF (not included)
├── derivatives-beh            # Preprocessed behavioral data (input for the analyses)
├── derivatives-scr            # Preprocessed skin conductance response data obtained upon request (input for the analysis)
├── description.json           # Project metadata
├── docs-mdl                   # Documentation: bayesian graphical model
├── participants.json
├── participants.tsv
├── plots-gp                   # Plots of Bayesian model extended with gaussian process
├── plots-kohonen              # Plots of the kohonen self-organizing map
├── plots-mdl                  # Plots of baseline Bayesian model of memory recognition
├── plots-mixture              # Plots of Bayesian model extended with latent mixture
├── plots-reproducibility      # Plots of data quality and reproducibility of original results
├── results-gp                 # Results of Bayesian model extended with Gaussian process
├── results-kohonen            # Results of kohonen self-organizing maps
├── results-mdl                # Results of baseline Bayesian model of memory recognition
├── results-mdl-kfold          # Results of 5-fold cross validation
├── results-mixture            # Results of Bayesian model extended with latent mixture
├── results-mixture-covariates # External covariates
├── results-reproducibility    # Results of reproducibility analysis
└── results-rl                 # Results of reinforcment learning model

Prerequisites

To successfully reproduce the analysis in this repository, ensure that you have the following tools installed:

0. Unix operating system.
1. GNU Make: This is used to automate the build process of the project.
2. Docker/Podman: Docker / Podman is used to create a containerized environment where the analysis runs in an isolated environment with all necessary dependencies. This approach also isolates the analysis from the host system.

The code was executed with GNU Make 3.81 and podman version 5.0.2 on macOS Ventura 13.6.7 (22G720) with an Apple M2 Pro chip. If you do not have podman installed, you can modify the Makefile to use the appropriate container engine (e.g., docker). Simply modify the variable ENGINE in the Makefile.

Podman Machine Info

Parameter	Value
OsArch	linux/arm64: 5.0.3
Kernel	6.8.8-300.fc40.aarch64
CPUs	8
Memory	32 GB
Disk Size	100 GB

Getting Started

Build the image

A Dockerfile is provided in the dockerfiles directory. To build the image, navigate to the dockerfiles directory and run:

cd code/dockerfiles
make build-docker-r

This will create an image called rstan:2.32.5 that contains all the dependencies needed to run the analysis.

To test the container, run:

make test-docker-r

Run the code with GNU Make

Once you have the Docker image built, you can use GNU Make to run the analysis inside the Docker container. Make will automate the process and ensure the proper environment is used.

Rebuild everything (not-recommended)

To reproduce all analyses:

make all

By default, small MCMC sampling is performed. However, this will likely overkill your system.

Rebuild specific outcome folders (recommended)

Open master Makefile and check the target analysis you want to rebuild. For example:

# to download the source data from OSF repository
make download

# to fit bayesian model (requires data and other pre-requisits)
make fit-mdls

License

This work is licensed under CC BY 4.0

Acknowledgment

Special thanks to Dr. Felix Kalbe for sharing the original dataset.

References

Kalbe, F., & Schwabe, L. (2022). On the search for a selective and retroactive strengthening of memory: Is there evidence for category-specific behavioral tagging? In Journal of Experimental Psychology: General (Vol. 151, Issue 1, pp. 263–284). American Psychological Association (APA). https://doi.org/10.1037/xge0001075