What’s in the semantic fluency task? An exploratory study of the dynamics of semantic fluency during bilingual development and the role of clustering and switching
Résumé
The semantic fluency task (SFT) is frequently used with disordered populations to assess both lexical retrieval and executive functions. In bilinguals, the task seems sensitive to dominance effects (e.g., Lam & Yoon, 2023), and might be a good candidate as an objective measure of dominance patterns. However, systematic investigation of the links between the latter and results on the SFT for each language is still scarce (see Paap et al., 2019 for review). Moreover, hardly anything is known about the dynamics of SF skills over bilingual language development. A recent study (Lethinen et al., 2024) suggests that it might be interesting to take into account not only overall scores, but also measures of clustering and switching patterns (Troyer et al., 1997) that seem to evolve during bilingual language development. While both have been shown to be necessary for efficient processing of the SFT, their evolution in bilingual language development is not clear yet. Clustering refers to the capacity to exploit subcategories of the semantic category elicited (e.g., for animals: pets, farm animals, wild animals...) while switching refers to the ability to switch to a new subcategory when the preceding is no longer productive. The use of these strategies is likely to depend on language proficiency. Moreover, categorization, and hence clustering in the SFT, is likely to vary with respect to cultural settings.In the present study, we will explore acquisitional data of the SFT collected with 25 Syrian learners of French (age 18-68 years) with variable proficiency who spent from 1 month -10 years in France. The participants were tested twice, at T1 and approximately 12 months later (T2). This allows us to document the evolution of semantic fluency both longitudinally and cross-sectionally taking into account L2 level at data-collection, age at arrival in France and length of immigration. All tasks were performed first in French (L2) and then in Syrian Arabic (L1). Two semantic categories known to be particularly productive were used: animals and fruit. These data will be compared to the semantic fluency data of 25 matched Lebanese bilinguals (a population culturally and linguistically close to the Syrian speakers) in both Lebanese Arabic and French. This comparison will allow us a) to document the evolution of clustering and switching strategies in relation to bilingual proficiency, and b) explore cultural specificities in the content of the two semantic categories used.
