Key Takeaways

• Just as with artifact rejection, the overarching goal in designing an artifact correction strategy is to maximize the likelihood that you will obtain an accurate answer to the scientific question your study is designed to answer. This typically involves reducing noise to maximize statistical power, avoiding artifactual voltages, and avoiding unwanted changes in the sensory input.
• Some of the assumptions of ICA are known to be incorrect, and it is therefore imperfect. In practice, ICA works best for large and frequent artifacts, such as blinks. I therefore recommend a conservative correction strategy in which you remove only the small set of ICs that correspond with well-understood artifacts and that are well isolated by the ICA decomposition.
• You can significantly improve the ICA decomposition process by performing it on a copy of the EEGset that has had addition steps applied to minimize activity that would otherwise interfere with the decomposition. This includes deleting break periods, deleting segments with extra-huge C.R.A.P., and filtering the data from 1-30 Hz. The ICA weights are then copied back to the original EEGset.
• You can determine which ICs are artifacts by looking at the scalp distribution, time course, and frequency content. You can use ICLabel to help you determine which ICs to correct, but you should combine this with manual inspection.
• In most cases, it is best to combine correction for ocular artifacts with rejection of epochs with large C.R.A.P. (along with rejection of epochs with blinks or eye movements that might interfere with the perception of the stimulus).
• Artifact correction and rejection are imperfect, and you should check their effectiveness with the following steps:
  • Assess whether blinks are actually a confound if not corrected (i.e., whether they differed between experimental conditions)
    • If so, assess whether the artifact correction process eliminated the confound.
      • If the confound was not fully eliminated, assess whether the remaining confound was large enough to have a meaningful impact on the ERP effects.
  • Assess whether data quality (SME) was improved by correction of EOG artifacts and whether it was further improved by also rejecting epochs with large C.R.A.P.

References

Artoni, F., Delorme, A., & Makeig, S. (2018). Applying dimension reduction to EEG data by Principal Component Analysis reduces the quality of its subsequent Independent Component decomposition. NeuroImage, 175, 176–187. https://doi.org/10.1016/j.neuroimage.2018.03.016

Berger, H. (1929). Ueber das Elektrenkephalogramm des Menschen. Archives Fur Psychiatrie Nervenkrankheiten, 87, 527–570.

Bigdely-Shamlo, N., Mullen, T., Kothe, C., Su, K.-M., & Robbins, K. A. (2015). The PREP pipeline: Standardized preprocessing for large-scale EEG analysis. Frontiers in Neuroinformatics, 9. https://doi.org/10.3389/fninf.2015.00016

Box, G. E. P. (1976). Science and Statistics. Journal of the American Statistical Association, 71(356), 791–799. https://doi.org/10.1080/01621459.1976.10480949

Delorme, A. (2023). EEG is better left alone. Scientific Reports, 13, 2372. https://doi.org/10.1038/s41598-023-27528-0

Dimigen, O. (2020). Optimizing the ICA-based removal of ocular EEG artifacts from free viewing experiments. NeuroImage, 207, 116117. https://doi.org/10.1016/j.neuroimage.2019.116117

Drisdelle, B. L., Aubin, S., & Jolicoeur, P. (2017). Dealing with ocular artifacts on lateralized ERPs in studies of visual-spatial attention and memory: ICA correction versus epoch rejection. Psychophysiology, 54, 83–99. https://doi.org/10.1111/psyp.12675

Groppe, D. M., Makeig, S., & Kutas, M. (2009). Identifying reliable independent components via split-half comparisons. Neuroimage, 45, 1199–1211. https://doi.org/10.1016/j.neuroimage.2008.12.038

Heise, M. J., Mon, S. K., & Bowman, L. C. (2022). Utility of linear mixed effects models for event-related potential research with infants and children. Developmental Cognitive Neuroscience, 54, 101070. https://doi.org/10.1016/j.dcn.2022.101070

Kappenman, E. S., Farrens, J. L., Zhang, W., Stewart, A. X., & Luck, S. J. (2021). ERP CORE: An open resource for human event-related potential research. NeuroImage, 225, 117465. https://doi.org/10.1016/j.neuroimage.2020.117465

Klug, M., & Gramann, K. (2021). Identifying key factors for improving ICA-based decomposition of EEG data in mobile and stationary experiments. European Journal of Neuroscience, 54, 8406–8420. https://doi.org/10.1111/ejn.14992

Luck, S. J. (2014). An Introduction to the Event-Related Potential Technique, Second Edition. MIT Press.

Luck, S. J., Stewart, A. X., Simmons, A. M., & Rhemtulla, M. (2021). Standardized measurement error: A universal metric of data quality for averaged event-related potentials. Psychophysiology, 58, e13793. https://doi.org/10.1111/psyp.13793

Näätänen, R., & Kreegipuu, K. (2012). The mismatch negativity (MMN). In S. J. Luck & E. S. Kappenman (Eds.), The Oxford Handbook of Event-Related Potential Components (pp. 143–157). Oxford University Press.

Pion-Tonachini, L., Kreutz-Delgado, K., & Makeig, S. (2019). ICLabel: An automated electroencephalographic independent component classifier, dataset, and website. NeuroImage, 198, 181–197. https://doi.org/10.1016/j.neuroimage.2019.05.026

Winkler, I., Debener, S., Müller, K.-R., & Tangermann, M. (2015). On the influence of high-pass filtering on ICA-based artifact reduction in EEG-ERP. 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 4101–4105. https://doi.org/10.1109/EMBC.2015.7319296

Zhang, G., Garrett, D. R., Simmons, A. M., Kiat, J. E., & Luck, S. J. (2024). Evaluating the effectiveness of artifact correction and rejection in event-related potential research. Psychophysiology, 61, e14511. https://doi.org/10.1111/psyp.14511