Background/Objectives: Previous research has investigated the impact of cognitive as well as emotional predictors, including WMC and language anxiety on language production. These studies often adopt the Complexity, Accuracy and Fluency (CAF) framework to assess language production. However, outcomes on language production related to meaning have played an underrepresented role in this model, resulting in previous findings that were inconsistent. Methods: The present study addresses this limitation by integrating the meaning-oriented Communication Content Function (CCF) model, based on Systemic Functional Linguistics, into the CAF framework, resulting in a revised model for language production evaluation under experimental conditions, which we call the Formal and Functional Evaluation (FFE) model. This model allows for a more comprehensive investigation of the impact of WMC and language anxiety on language production. Using data from L3 acquisition of German by L1 Cantonese and L2 English speakers, we examine how WMC, language anxiety, and language proficiency interact in shaping production outcomes. Results: Multiple regression analyses show that the main effects of WMC and language anxiety are not significant in most language production outcomes, whereas the main effect of language proficiency is significant in almost all outcomes. Also, a significant interaction between WMC × language proficiency is observed, although language anxiety × language proficiency as well as language anxiety × WMC are not. Conclusions: These results suggest that cognitive and affective factors influence language production primarily through interaction effects moderated by language proficiency.
Ran et al. (Mon,) studied this question.