Still an Ineffective Method With supertrials/ERPs-Comments on "decoding Brain Representations by Multimodal Learning of Neural Activity and Visual Features"

• Published in: IEEE transactions on pattern analysis and machine intelligence Vol. 45; no. 11; pp. 1 - 3
• Main Authors: Bharadwaj, Hari M, Wilbur, Ronnie B, Siskind, Jeffrey Mark
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.11.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Blanking; Brain - physiology; brain-computer interface; Convolution; EEG; Electroencephalography; Electroencephalography - methods; Evoked Potentials - physiology; human vision; Humans; Machine Learning; neuroimaging; neuroscience; object classification; Signal Processing, Computer-Assisted; Signal to noise ratio; Statistical methods; Support vector machines; Three-dimensional displays; Training
• ISSN: 0162-8828; 1939-3539; 2160-9292; 1939-3539
• DOI: 10.1109/TPAMI.2023.3292062

A recent paper claims that a newly proposed method classifies EEG data recorded from subjects viewing ImageNet stimuli better than two prior methods. However, the analysis used to support that claim is based on confounded data. We repeat the analysis on a large new dataset that is free from that confound. Training and testing on aggregated supertrials derived by summing trials demonstrates that the two prior methods achieve statistically significant above-chance accuracy while the newly proposed method does not.