Lion – tiger – stripes: Neural correlates of indirect semantic priming across processing modalities

“Lions” do not have “stripes”. However, via the word “tiger” both words are closely connected within the semantic network. With the present study we pursued two goals: to detect neural correlates of (1) directly and indirectly related word pairs by means of priming, and (2) to assess the effect of presentation modality.Stimuli were presented with a short SOA of 350 ms as subjects performed a lexical decision task during fMRI measurement. Four experimental conditions were compared: directly related (picture–frame), indirectly related (anvil–nail), unrelated (steamboat–needle) and nonword trials (chalk–edan), presented in a uni- (word–word) and cross-modal (auditory–word) version.Behavioral data revealed a modality-independent priming effect only for direct semantic priming. On a neural level, directly linked words led to left-lateralized activations in fronto-temporo-parietal areas. Indirect priming led to right-hemispheric fronto-parietal signal changes. Common areas of activation for uni- and cross-modal priming were found within the left middle temporal gyrus and right precuneus for direct priming and within the right insula for indirect priming. The comparison of the semantic distances (direct > indirect) showed one region activated modality-independent: the precuneus.Direct priming is associated with activation clusters corresponding to a large left-lateralized network. Indirect priming recruits right-hemispheric regions, reflecting widespread semantic fields and attentional components. The modality-independent comparison of direct and indirect priming revealed common areas of activation supporting an amodal rather than multiple semantic systems. The activation related to semantic distances underpins the special role of the precuneus. This region is involved in semantic priming and association processing whereas episodic memory contents might be addressed.