Chapter 6. Effects of task complexity and interaction 
on L2 performance

The Cognition Hypothesis (Robinson, 2005, this volume) claims that increases in cognitive task complexity along ‘resource-directing’ variables focus L2 learners’ attention on language. This results in more accurate and syntactically and lexically more complex speech, while negatively affecting fluency. Interactive tasks have been claimed to enhance L2 learners’ attention to language form, raising the chance that they will produce more accurate speech in dialogues than monologues. As speakers ask for clarifications, and mirror each other’s speech, pair work allegedly leads to structurally less complex and lexically less diverse output while fluency increases. When task complexity and the interactional format are manipulated at the same time, combined effects may occur. As a complex task requires more clarification work, complex dialogic tasks may lead to higher levels of accuracy, but lower levels of structural and lexical complexity. This study investigated the effects of increased task complexity and interaction: 64 L2 learners of Dutch performed simple and complex tasks either individually (+monologic) or in pairs (–monologic). Task complexity was manipulated along the dimension ‘± few elements’. The students’ output was analyzed for measures of linguistic complexity, accuracy, and fluency. It was compared to the output produced by 44 native speakers of Dutch performing the same tasks. The results provide little support for the Cognition Hypothesis, although interaction had a large impact on both L2 learners and L1 speakers. The discussion takes a cognitive perspective on attentional allocation and L2 task performance.