Executive, language, or both? An examination of the construct validity of verbal fluency measures

The purpose of the present study was to evaluate whether phonemic and semantic verbal fluency were more related to aspects of language processing than executive functioning (EF). An exploratory factor analysis was performed on a college-aged sample of 320 healthy participants using principle axis factoring and promax rotation on nine measures of EF. The first three factors, labeled: working memory, fluid reasoning, and shifting/updating, were extracted and used as latent executive variables. Participants were also split into low, medium, and high phonemic and semantic verbal fluency ability groups. Phonemic and semantic fluency correlated similarly across all three extracted EF factors and word knowledge. Using one-way analysis of variance (ANOVAs), there was a main effect for both phonemic and semantic verbal fluency groups and all outcome variables (i.e., the EF factors and word knowledge). Tukey HSD post hoc analyses showed that those in the low verbal fluency ability groups had significantly lower scores across all outcome measures compared to the high verbal fluency ability groups. Across all analyses, semantic fluency had stronger relations with the EF factors, signifying a large executive component involved in the task. Both phonemic and semantic fluency were similarly related to multiple dimensions of EF and word knowledge and should be considered executive language tasks.