This project contains the scripts and datasets accompanying the manuscript: Wagner, J., Wessel, J., Gharemahni, A. & Aron, A.R. 2017, Establishing a right frontal beta signature for stopping action in scalp EEG: implications for testing inhibitory control in other task contexts (submitted) please cite if you use these scripts for your research

In the manuscript, two stop signal data-sets standing alone, and one stop signal plus unexpected events data set were analyzed. We provide access to the EEG and behavioral data for the latter (combined) data-set along with detailed instructions and scripts to generate the results shown in the paper.

You can use the provided scripts to reproduce the results in the manuscript above. I encourage you to try the same analysis also on your data using and modifying the sripts here provided. If you do so, please cite the manuscript above.

Instructions:

To run the analysis, unzip the eeglab folder and save all the scripts, the provided eeglab13_6_5b version and the data in one folder to run the analysis. Scripts were written using MATLAB2015b (The MathWorks).

The data provided is sampled at 512Hz, high-pass filtered at 2Hz and low-pass filtered at 200Hz. For more details on the data and paradigm please refer to the manuscript above. The folder also contains the rejection matrices marking bad segments in the single datasets that have been previously identified by visual inspection.

To run the analysis run the provided scripts in the order:

1. ICApreproc.m
2. createSTUDY.m
3. clustering.m
4. computeERSP.m
5. AV_ERSP.m

At step 3 you may have to adapt some parameters, since rerunning ICA might not always give you the exact same components, albeit very similar, and clustering results may also slightly differ from the results in the manuscript.

Information on the scripts

1. ICApreproc.m

• preprocessing the data (rejecting preselected bad channels and periods of data, and rereferencing to average reference)
• running ICA (be aware that this may take a while - a few days for all subjects)
• applying weights to the EEG data
• computing dipoles
• saving datasets with weights and dipoles

2. createSTUDY.m

• epoch datasets
• create STUDY
• precompute component measures for clustering
• save STUDY

3. clustering.m

• load STUDY
• build preclustring array
• cluster components
• plot cluster average scalp maps and dipoles
• find right frontal cluster --> at this step you may have to adapt some parameters, since ICA might not give you the exact same components, albeit very similar, and clustering results may also slightly differ from the results in the paper
• plot single IC scalp maps in this cluster
• find artefact components in this cluster and remove --> at this step again you may have to adapt some parameters, since ICA might not give you the exact same components, albeit very similar, and clustering results may also slightly differ from the results in the paper
• plot single IC scalp maps in this cluster
• plot single IC ERSP images and average cluster ERSP of the right frontal cluster
• save Subject and IC indices in this cluster

4. computeERSP.m

• load subject and IC index for right frontal cluster
• load data -epoch data relative to tone for unexpected events task and and relative to go signal for stop signal task
• get latency of stop signal
• compute Event related spectral perturbations, timewarp stop signal task trials to stop signal delay
• save ERSP matrixes for right frontal cluster ICs

5. AV_ERSP.m

• load ERSP matrices
• compute common baseline
• compute significance relative to baseline over subjects
• compute significance and in between conditions for Novel vs Standard and Uccessful stop vs fail
• multiple comparison correction with fdr
• mask ERSPs
• plot average ERSPs
• plot difference between Novel-Standard and Successful-Failed Stop