Effects of cognitive task complexity and online planning on second language learners' argumentative writing.

Based on Kellogg's writing model, Skehan's Limited Attentional Capacity Model (LACM), and Robinson's Cognition Hypothesis, our study investigated the effects of cognitive task complexity on syntactic complexity, lexical complexity, accuracy, fluency, and functional adequacy in Chinese L2 students' argumentative writing, when students were under an online planning condition. Sixty-eight participants from a Chinese university were recruited to complete two writing tasks with task complexity varied in terms of [+ argument elements]. The findings showed that increasing task complexity led to decreased subordination in terms of clauses per T-unit and dependent clauses per clause, increased phrasal elaboration in terms of coordinate phrases per clause, and no changes in mean length of T-unit, T-units per sentence, mean length of clause, and complex nominals per clause. Neither significant differences in accuracy nor fluency were found as a function of increasing task complexity. Detrimental effects on functional adequacy in content, organization, and overall scores were identified with the increases in task complexity. The trade-offs between syntactic and lexical complexity and between syntactic complexity and functional adequacy support the basic principle of Skehan's LACM that human's information processing capacity is limited and Kellogg's claim that learners have a limited central executive capacity in writing. Implications of the results of this research are discussed.