This directory contains the analysis scripts accompanying the manuscript: Finzi, Wagner, Itthipuripat & Aron, 2017, Stopping cognition: unexpected events recruit prefrontal inhibitory control and interrupt working memory (in prep)
Analysis and scripts adapted from: 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.
For more details on data and the paradigm please refer to the manuscript (Finzi, Wagner, Itthipuripat & Aron, in prep) when available or contact me at: rebecca.d.finzi@gmail.com. Additional information and the behavioral/EEG data is available on OSF.
This analysis pipeline requires eeglab (version 13_6_5b) which can be downloaded from the EEGLAB website.
- preproc.m
- preprocessing of stop data
- preprocessing of WM data
- merging of the two datasets
- add_error_codes.m
- add response error behavioral data to EEG data
- run_ica.m
- run ica on the merged datasets
- post_ICA.m
- apply weights to EEG data
- compute dipoles
- brain_comp_select.m
- select the ICs that appear to reflect brain activity
- create_study.m
- select only the brain components (determined in #5)
- epoch datasets
- create eeglab STUDY
- precompute component measures for clustering
- clustering.m
- cluster components
- save right frontal beta cluster
- computeERSP.m
- compute event related spectral perturbations for right frontal beta clustering
- AV_ERSP.m
- average cluster ERSP images for each condition, compute significance and plot differences
- computeERSP_st.m
- compute single trial ERSPs for each subject in right frontal beta cluster
- regression.m (calls regressERSP_Z.m)
- run regression for each subject testing effect of power in time frequency space on behavioral response error
- compute significance on group level for standard and novel conditions using standardized beta coefficients
- plot significant results for each condition
- for novel condition, pool across subjects and plot power (at peak) response error for visualization purposes