Fig. 4: Summary model for distinct components of linguistic meaning derived from a conjunction SB-MEMA.

Model derived from present analyses of reference resolution (referential vs. non-referential sentence endings) and its difficulty (limited vs. strong narrowing), and semantic coherence (coherent vs. incoherent non-referential sentences). Included is the color scheme for Type, Token, and Referent matched to colored bars next to regional descriptions, and also for regions showing some conditional difference for Reference, Coherence, and Narrowing (matched to colored clusters on brain plots) and for combinations of these (e.g., R + C = Reference and Coherence, but not Narrowing). Conjunction SB-MEMAs are plotted depicting regions sensitive to conditional contrasts, regardless of directionality (threshold: %BGA > 5%, t > 2.57, patient coverage ≥ 2), across two main time windows of interest (500–700 ms, 700–900 ms). Functional descriptions in green (bottom entries) are inferred from existing literature and are not intended to be monolithic, but specific to the present task context of naming-to-definition (see Discussion).