Using multivariate pattern analysis to increase effect sizes for event-related potential analyses
Published in Psychophysiology, 2024
This study demonstrates that multivariate pattern analysis (MVPA) can enhance effect sizes in ERP research, offering greater statistical power than traditional univariate methods. By applying MVPA techniques, such as support vector machine decoding and cross-validated Mahalanobis distance, to the ERP CORE dataset, the authors found that MVPA approaches yielded effect sizes that were as large or larger than those produced by conventional univariate analyses across seven widely studied ERP components. These findings suggest that MVPA can be a valuable tool for researchers aiming to detect subtle differences in ERP signals, thereby improving the sensitivity and robustness of electrophysiological studies.
Recommended citation: Carrasco, C. D., Bahle, B., Simmons, A. M., & Luck, S. J. (2024). "Using multivariate pattern analysis to increase effect sizes for event-related potential analyses." Psychophysiology, 61(7), e14570. https://doi.org/10.1111/psyp.14570
Download Paper