Answers are grounded in this paper and cite their sources.
This source-localised EEG-ERP study (N = 25) compared single-epoch source-level magnitude across three categories of text–image pairings: internet memes, literal congruent controls, and semantically incongruent controls, in both native Russian and foreign English. The data were analysed using linear mixed-effects models with participants and items as random intercepts. Memes differed from the literal congruent controls at every component examined. In contrast, differences from the incongruent controls were restricted to the P3b and LPC windows. Language modulation was observed in the early and intermediate windows (P100, P3a and P3b), but not in the late ones (N170, N400 and LPC). These findings suggest that the non-literal, integrative nature of text–image relationships in memes is associated with a broader and more sustained pattern of source-level activity compared to literal correspondence or simple semantic mismatch. Furthermore, they indicate that the early and intermediate stages of this processing are sensitive to the language of presentation.
No takes yet. Share an insight, caveat, or question.
Karimova et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: