This repository contains Matlab code (.m) and data files (.mat, .csv, .nii) required to reproduce the results of the study entitled "A neural mechanism of social responsibility", by Maria Gaedeke, Tom Willems, Omar Salah Ahmed, Bernd Weber, Rene Hurlemann, and Johannes Schultz. A preprint of this study can be found on bioRxiv (Link: https://doi.org/10.1101/2020.05.25.107300). The final Version of Record (reviewed preprint) is available at eLife (https://elifesciences.org/reviewed-preprints/105391). The fMRI data of the study are available on OpenNeuro (Link: https://openneuro.org/datasets/ds005588).
BehaviouralData contains the raw behavioural data (.mat format) from the two studies reported in the manuscript. Code contains Matlab code to run the experiment (socialDecisionRespo_fMRI.m), reproduce the analyses and generate the figures (see below for details). Within Code, the folders Figures and csv contain outputs from the analyses, fMRI contains fMRI-specific Matlab scripts, and bin contains helper functions. fMRIresults contains Nifti files resulting from the analysis of the fMRI data and used in the Figures.
To replicate the results Figures using Matlab: First, download the repository and add all folders to the path. Then run:
- master_behavAnalysis.m to generate Figures 2 and 3, and Supplementary Figure 1
- master_fMRI_showResults to generate Figures 4 and 5, and Supplementary Figure 2
To replicate the fMRI results using SPM12 in Matlab based on the raw data, download the raw data from OpenNeuro (Link: https://openneuro.org/datasets/ds005588). Then, use the following scripts located in Code/fMRI:
- a_preprocessing.m for realignment / motion correction, normalization / unwarping, and smoothing
- b_FFX.m to fit individual general linear models to the BOLD data
- c_gPPI.m to run functional connectivity analyses using the gPPI toolbox (download on https://www.nitrc.org/projects/gppi)
- d_RFX_fullFact.m to run most group-level "random effects" models
- e_RFX_pairedTtest_guiltEffect.m for a specific group-level test
- f_apply_YuKoban_guilt_signature_map.m to compare our findings to previously published results (uses code from https://github.com/canlab/Neuroimaging\_Pattern\_Masks)