How robustly do multivariate EEG patterns track individual-subject lexico-semantic processing of visual stimuli?

Petit, Selene; Brown, Alyse; Jessen, Emil Trenckner; Woolgar, Alexandra

doi:10.1080/23273798.2023.2177315

Cited by 2 publications

(1 citation statement)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Isik et al (2014), groups of 10 trials were averaged before using a 5-fold cross-validation, and in Duncan et al (2023), groups of 3 trials were averaged before using a 10-fold cross-validation. Goddard et al (2022) averaged over 16 trials, before using an 8-fold cross-validation, and Petit et al (2023) used the median of 5 trials and a leave-one-pseudotrial-out cross-validation with 22 folds. Given the variation across experiments, it can be difficult to make decoding analysis decisions regarding a new set of data.…”

Section: Introductionmentioning

confidence: 99%

Optimising analysis choices for multivariate decoding: creating pseudotrials using trial averaging and resampling

Scrivener,

Grootswagers,

Woolgar

2023

Preprint

Self Cite

View full text Add to dashboard Cite

Multivariate pattern analysis (MVPA) is a popular technique that can distinguish between condition-specific patterns of activation. Applied to neuroimaging data, MVPA decoding for inference uses above chance decoding to identify statistically reliable condition-specific information in neuroimaging data which may be missed by univariate methods. However, several analysis choices influence decoding success, and the combined effects of these choices have not been fully evaluated. We systematically assessed the influence of trial averaging and resampling on decoding accuracy and subsequent statistical outcome on simulated data. Although the optimal parameters varied with the classifier and cross-validation approach used, we found that modest trial averaging using roughly 5-10% of the total number of trials per condition improved accuracy and associated t-statistics. In addition, a resampling value of 2 could improve t-statistics and classification performance, but was not always necessary. We provide code to allow researchers to optimise analyses for the parameters of their data.

show abstract