Chapter 3. Measuring task complexity

The current study explored whether language proficiency mediates the perception of task difficulty and whether intended task complexity differences are reflected in the language production (i.e., fluency, accuracy, and complexity) of learners of different proficiency levels in English. 37 participants (20 advanced and 17 pre-intermediate learners of English) took part in the experiment, all of them undergraduate students at a Spanish university, aged between 18 and 25. They were provided with two tasks of hypothetically differing cognitive complexity levels manipulated along ± few elements and ± spatial reasoning dimensions. Two kinds of instruments were used to measure participants’ perceptions of task difficulty: self-reported difficulty ratings and time estimation of task completion. Complexity, accuracy, and fluency measures were used to analyze participants’ speech production on the two tasks. As far as the perception of complexity is concerned, no significant differences between the high and low proficiency groups were found. Regarding performance measures, in the high proficiency group the complex task triggered greater lexical and structural complexity and accuracy, to the detriment of fluency. In the low proficiency group, on the complex task, fluency was boosted while the other areas remained intact irrespective of cognitive task complexity. These results are discussed in terms of Robinson’s Cognition Hypothesis and Skehan’s Trade-off Hypothesis.