Introducing RELAX (the Reduction of Electroencephalographic Artifacts): A fully automated pre-processing pipeline for cleaning EEG data - Part 2: Application to Event-Related Potentials

Abstract

Electroencephalography (EEG) is commonly used to examine neural activity time-locked to the presentation of a stimulus, referred to as an Event-Related Potential (ERP). However, EEG is also influenced by non-neural artifacts, which can confound ERP comparisons. Artifact cleaning can reduce artifacts, but often requires time-consuming manual decisions. Most automated cleaning methods require frequencies <1Hz to be filtered out of the data, so are not recommended for ERPs (which often contain <1Hz frequencies). In our companion article, we introduced RELAX (the Reduction of Electroencephalographic Artifacts), an automated and modular cleaning pipeline that reduces artifacts with Multiple Wiener Filtering (MWF) and/or wavelet enhanced independent component analysis (wICA) applied to artifact components detected with ICLabel (wICA_ICLabel) (Bailey et al., 2022). To evaluate the suitability of RELAX for data cleaning prior to ERP analysis, multiple versions of RELAX were compared to four commonly used EEG cleaning pipelines. Cleaning performance was compared across a range of artifact cleaning metrics and in the amount of variance in ERPs explained by different conditions in a Go-Nogo task. RELAX with MWF and wICA_ICLabel cleaned the data the most effectively and produced amongst the most dependable ERP estimates. RELAX with wICA_ICLabel only or MWF_only may detect experimental effects better for some ERP measures. Importantly, RELAX can high-pass filter data at 0.25Hz, so is applicable to analyses involving ERPs. The pipeline is easy to implement via EEGLAB in MATLAB and is freely available on GitHub. Given its performance, objectivity, and ease of use, we recommend RELAX for EEG data cleaning.

Introducing RELAX (the Reduction of Electroencephalographic Artifacts): A fully automated pre-processing pipeline for cleaning EEG data - Part 1: Algorithm and Application to Oscillations

“…This means that the RELAX MWF_wICA method makes more data available for inclusion in analyses than most pipelines, reducing the risk that participant data needs to be excluded for having too few epochs for inclusion, or that results might be biased by exclusion of epochs containing neural activity due to inferior cleaning. Our companion article additionally indicated that MWF_wICA produced data that showed amongst the highest scores for reliability of ERPs across trials and participants, suggesting cleaning with MWF_wICA provided amongst the most consistency in detecting experimental effects (Bailey et al, 2022). The inclusion of more high-quality epochs for analysis and more reliable data has been suggested by simulations to improve study power even more than increasing sample size (Clayson, Carbine, et al, 2021; Kolossa & Kopp, 2018; Luck et al, 2021).…”

“…Some default settings can be recommended. For example, the results reported in our companion paper indicated that the ICA methods infomax or fastICA with the symmetric setting were the best performers within RELAX (Bailey et al, 2022). If users can install cudaICA, this setting is likely to be optimal, providing both speed and the best performance (Raimondo et al, 2012).…”

“…Traditional extended infomax can also be used, but is slower (Lee et al, 1999). Otherwise, we have left the default setting as fastICA with the symmetric setting (Hyvarinen, 1999), which our companion paper has indicate is (just slightly) better than the deflation setting (Bailey et al, 2022). We have set this fastICA to repeat the ICA up to three times in the case of a non-convergence issue which can adversely affect ICA decompositions, then to switch to the ‘defl’ setting if non-convergence still occurs.…”

Meditators probably show increased behaviour-monitoring related neural activity

“…Our previous study used data pre-processing methods that we have since demonstrated to be less effective at cleaning artifacts compared to the method used in the current study (BLINDED FOR REVIEW). However, different EEG data cleaning approaches have been shown to produce only minor differences in study outcomes (Robbins et al, 2020), and these differences are still aligned in direction (Robbins et al, 2020, Bailey et al, 2022a, 2022b). As such, it seems unlikely that these explanations would produce such strong evidence for the null hypothesis if the results from our current study reflect the true result (we would expect more inconclusive results).…”

“…Each stimulus was presented for 250ms with an intertrial interval of 900ms (with a 50ms jitter). EEG data were pre-processed using the RELAX pipeline, which has demonstrated optimal cleaning of artifacts and preservation of ERP signals compared to other cleaning approaches (Bailey et al, 2022a(Bailey et al, , 2022b. This cleaning pipeline filtered the data from 0.25 to 80Hz with a notch filter from 47 to 53Hz, applied both Multiple Wiener Filters (MWF) and wavelet enhanced independent component analysis (wICA) to identify and remove muscle, eye movement and blink and drift artifacts from the data, and additionally the wICA reduced line noise, heartbeat and other artifacts.…”