Task modality, working memory, and L2 outcomes: An individual-differences approach

Task modality has been claimed to influence opportunities for second language (L2) grammar development (see Gilabert, Manchón, & Vasylets, 2016), yet the relevance of learner differences in cognitive capacity in oral and written pedagogic tasks is still not well understood. Oral tasks are hypothesized to place higher demands on memory due to the time pressure associated with speaking and the fleeting nature of oral language (e.g., Gilabert et al., 2016); however, some accounts suggest that written tasks can be more demanding, as writing prompts ample monitoring and the activation of graphemic representations (see, e.g., Kellogg, 2007; Sauerland & Sporer, 2011). This study sought to examine how working memory capacity relates to L2 learners’ (a) grammar development and (b) perceptions of task demands in oral and written form-focused tasks. Beginner-level learners of L2 Spanish completed two computerized focused tasks in which they produced the target language and received feedback in the oral modality (Speaking, n = 27) or the written modality (Writing, n = 28). Two target structures of varying salience were considered in the study. After each task, participants rated their perceptions of required mental effort and task difficulty. L2 grammar development was gauged using pre-, post-, and delayed post-tests of production and written and aural acceptability judgment. Participants’ executive, phonological, and visuospatial working memory capacities were measured with an automated operation span task, a non-word recognition task, and a forward Corsi block-tapping task, respectively. Regression analyses revealed that working memory capacity was predictive of L2 grammar learning outcomes and task demand ratings in the Speaking group only. Specifically, phonological and visuospatial working memory were associated with production and acceptability judgment performance accuracy, whereas executive working memory was related to learners’ ratings of perceived mental effort. Differences were observed based on target structure salience. Theoretical and pedagogical implications are discussed.