Contribution of low-level image statistics to EEG decoding of semantic content in multivariate and univariate models with feature optimization

Eric Lützow Holm, Diego Fernández Slezak, Enzo Tagliazucchi

Abstract: Highlights•Optimized univariate models outperform multivariate models in EEG visual decoding.•Model performance is confounded by low-level image features.•Model optimization can increase the sensitivity of model towards this confound.•Low-level image feature confounds also appear when decoding of concept categories.