Cross-linguistic differences in the semantic networks for animal fluency between English and Korean speakers with and without aphasia

This is the authors' abstract. We don't add key points for this paper.

Background Semantic fluency tasks are widely used to assess lexical – semantic retrieval in healthy aging and aphasia; however, growing evidence indicates that semantic organization is influenced by culturally grounded lexical experience. Prior cross-linguistic work suggests that Korean speakers rely on culture-specific semantic schemas – such as zodiac animals – during animal fluency, yet it remains unclear how such differences manifest in retrieval strategies or semantic network structure in aging and aphasia.Aims The current study investigated cross-linguistic differences in clustering, switching, and semantic network structure during animal fluency in English- and Korean-speaking older adults (OA) and persons with aphasia (PWA).Methods & Procedures Participants included English- and Korean-speaking OA (n = 30) and PWA (n = 67). All participants completed a one-minute animal fluency task. Clustering and switching were computed using SNAFU, and group-level semantic networks were constructed using co-occurrence – based Semantic Network Analysis. Global graph metrics – average shortest path length (ASPL), clustering coefficient (CC), and modularity (Q) – were compared across language groups.Outcomes & Results Behaviorally, English- and Korean-OA produced comparable total number of correct responses; however, English-OA generated significantly larger cluster sizes than Korean-OA. In PWA, English speakers produced significantly more correct items and significantly more switches than Korean speakers. Semantic network analyses revealed that Korean-OA networks demonstrated significantly longer ASPL, significantly lower CC, and significantly higher Q relative to English-OA. In PWA, Korean speakers also exhibited significantly longer ASPL than English speakers, indicating reduced global connectivity. Together, these results demonstrate that significant cross-linguistic differences in semantic network structure are evident in healthy aging and remain observable in aphasia.