Task complexity and interaction: (combined) effects on task-based performance in Dutch as a second language

This paper presents the overarching conclusions of three consecutive investigations into task-based L2 performance. It aims at giving a better understanding of how changes in the number of elements referred to in a task affect L2 production, and how this relates to cognitive task complexity. Furthermore, it evaluates differences between monologic and dialogic tasks, and searches for combined effects of the factors `± elements' and `± monologic'. Analyses examined the oral task performances of 152 participants by using global measures of linguistic complexity, accuracy, and fluency as well as a task specific measure. Results revealed hardly any effects of the manipulation of the number of elements. Dialogic tasks, however, consistently guided L2 performers towards greater accuracy, lexical complexity, and fluency. The discussion compares these findings to native speaker baseline data, reviews the results in light of Robinson's (2005) Cognition Hypothesis, and highlights the cognitive impact of the factor `± monologic'.