Data for "Mapping brain resilience: Higher educational attainment reduces the regional impact of stroke lesions on stroke severity"

Description

Data for Umarova, Gallucci, Hürlimann & Sperber: "Mapping brain resilience: Higher educational attainment reduces the regional impact of stroke lesions on stroke severity" Data on a study investigating the interaction between lesion location and cognitive reserve (represented by its proxy measure years of education) on stroke severity and outcome. Online materials include: - Main_analysis_interaction_CR: the code and results of the main anaylsis - Sensitivity_analyses: the code and results of sensitivity analyses, controlling for additional covariates (sex & age/ lesion volume) and evaluating the validity of results via a small permutation test In each results folder, following files are present: - BF_CR_interact_map_xxx: map of the Bayes factors for the interaction effect of cognitive reserve and voxel-wise lesion status - betas_CR_interact_map_xxx: (only provided for main results) the voxel-wise beta weights of the interaction derived from a standard GLM defined analogue to the Bayesian statistical model - h1_map_BF_CR_interact_map_xxx: binned BF map highlighting voxels with moderate/strong/very strong/extreme evidence for the alternative hypothesis - h0_map_BF_CR_interact_map_xxx: binned BF map highlighting voxels with moderate/strong/very strong/extreme evidence for the null hypothesis - noev_map_BF_CR_interact_map_xxx: binned BF map highlighting voxels without conclusive evidence for any hypothesis (Bayes Factors >0.333 and <3) Results are stored as NIFTI files (.nii) that can be viewed by various open brain viewer tools, such as MRIcroGL (https://www.nitrc.org/projects/mricrogl/) All code was run in R 4.2.1 with the Bayes Factor Package BayesFactor 0.9.12-4.4.

Institutions

• University Hospital of Bern

  Bern, Bern

Categories

Licence