The Effects Of Task Complexity On EFL Learners' Writing Performance

Since the 1980s, the relationship between task complexity and foreign language learners'linguistic performance has become the focus in many studies, and much has been discussed about the effects of task complexity in second language acquisition (SLA) research, particularly regarding the effects of task design and implementation variables on language production in terms of fluency, accuracy, and complexity. Major theories about task complexity include the Cognition Hypothesis and the Limited Attentional Capacity Model. In the Cognition Hypothesis, task complexity is manipulated along two dimensions:resource-directing dimension and resource-dispersing dimension. These two models have made different predictions about the effects of task complexity on language performance. The Limited Attentional Capacity Model does not make the resource-directing/ resource-dispersing distinction, claiming that complex task performance, along any dimensions, degrades accuracy, fluency and complexity simultaneously. Whereas studies on the effect of task complexity on oral language production have been in the forefront of investigations from the past twenty years, considerably less research has been conducted on how different task types and different levels of complexity of tasks influence written output produced by foreign language learners. In designing task-based language teaching (TBLT) lessons, it is an urgent issue for language teachers and curriculum developers to select tasks among numerous options and control them delicately so that they can elicit optimal educational efficacy, and in particular, to elicit the targeted features of writing competence.This dissertation attempted to add empirical findings to the existing small body of research on the effects of task complexity on the EFL learners'writing performance. Specifically, task complexity was manipulated along both resource-directing and resource-dispersing dimensions. Learners'L2 proficiency level was also incorporated to examine the interactive effects with task complexity on their writing performance.Eight classes of 305 non-English majors at a college in Nanjing participated in the study. Four classes were freshmen with a lower English proficiency level and the other four were sophomores with a higher English proficiency level. There were four task conditions:1) 10 minutes'pre-task planning time+3 requirements to be met in the choice of a hotel for holiday, no planning time+3 requirements to be met in the choice of a hotel for holiday,10 minutes'pre-task planning time+6 requirements to be met in the choice of a hotel for holiday, and no planning time+6 requirements to be met in the choice of a hotel for holiday. Each class was required to perform a task under one condition. Their written products were analyzed in terms of fluency, accuracy and complexity. Immediately after the participants finished their tasks, four participants from each group were selected randomly for an interview conducted in Chinese.The analysis of the quantitative and qualitative data yielded the following findings:1. Increasing task complexity along resource-directing dimension (increasing the number of requirements to be met in decision-making) had positive impacts on fluency, accuracy and lexical complexity of the learners'written language performance, but had a negative effect on syntactic complexity. The data obtained from the interview also gave some evidence to the findings summarized above. Learners paid more attention to content and language than to discourse structure in writing. A more complex task made more cognitive and/or conceptual demands on attention and memory resources, directing them to aspects of the linguistic code. So increased task complexity resulted in greater fluency, accuracy, and lexical complexity.2. With regard to resource-dispersing dimension of task complexity, pre-task planning resulted in significantly greater fluency than the no-planning task condition. Pre-task planning had little effect on accuracy, syntactic complexity and lexical complexity. The data obtained from the interviews revealed that the learners employed more cognitive strategies than metacognitive strategies when they planned before completing the task. It was also found that high proficiency learners used more strategies than the low proficiency learners and it could be assumed that high proficiency learners made better use of pre-task planning time than the low proficiency ones.3. L2 proficiency level and pre-task planning time had interactive effects on the fluency, accuracy and lexical complexity. In other words, high proficiency learners benefited more from pre-task planning time and they performed better in written fluency, accuracy and lexical complexity. Instead, low proficiency learners performed better in these three aspects when they had no time to plan.L2 proficiency level and the number of requirements to be met in decision-making did not have interactive effects on the fluency, accuracy and complexity. For both high and low proficiency learners, increasing the number of requirements to be met in decision-making could improve the fluency, accuracy and lexical complexity of writing.The findings of this study add further evidence to the Cognition Hypothesis and the Limited Attentional Capacity Model, thus enriching the research on task complexity. The study also lends some insights to SLA researchers, teachers, and syllabus designers, suggesting that it is highly important to understand the characteristics of a task, their influence on L2 performance, and the learners'L2 proficiency level so that suitable types of task can be designed or chosen for a target population to elicit the optimal target language production.