Abstract:
Two studies were conducted to investigate the role of task complexity and affective factors, and task complexity and pre-task planning conditions, in L2 writing production. Two writing tasks with varying degrees of complexity in relation to the reasoning demands and the number of elements were designed and validated. Upperintermediate English language learners were invited to participate in this study on a voluntary basis. In Study 1, one group of the learners (n = 60) performed a simple and a complex version of an argumentative writing task, and their writing production was measured in terms of syntactic complexity, accuracy, lexical complexity, and fluency (CALF), as well as the organisation, content, and overall written text quality. In Study 2, one group of the learners (n = 40) performed the simple and complex tasks under 10-minute pre-task planning and another group (n = 40) under no-pre-task planning conditions. The learners' writing production was compared using the same measures employed in Study 1. In Study 1, increasing task complexity led to a significant desirable change in one dimension of both syntactic and lexical complexity, had a significant adverse effect on accuracy and fluency, and led to the enhancement of L2 content, organisation, and writing quality. As regards the mediating role of motivation and anxiety, significant moderate negative associations were found between some dimensions of maladaptive approaches to learning and some measures of L2 writing production in the complex writing task performance. Conversely, significant moderate positive correlations were found between some aspects of adaptive approaches to learning and some measures of L2 writing production in the complex L2 writing task performance. In Study 2, a similar trend was found for the effect of increasing task complexity along the level of reasoning and the number of elements as in Study 1. The impact of 10-minute pre-task planning on L2 writing production was significantly favourable for one dimension of syntactic complexity and fluency. There were no significant effects on accuracy and lexical complexity. Nonetheless, significant positive changes for content, organisation, and writing quality were found. The findings lend partial support to the Cognition Hypothesis and the Trade-Off Hypothesis; other theoretical, methodological, and pedagogical implications are also discussed.