Task-modality effectsa study of task complexity effects in speech and writing

Resumen

One of the major areas of interest in task-based language learning and teaching (TBLT) has been the psycholinguistically-oriented strand, which explores the way in which task design may interact with second language (L2) learners` cognitive response in creating distinct opportunities for L2 learning and use. The main appeal of TBLT ideas is that a language task is presented as a manageable and, at the same time, powerful tool of instruction and research. What is necessary, however, is a clear understanding of the manner in which different task characteristics might affect learners` performance and development. There are two major theoretical frameworks which provide an account of the purported effects of task features on L2 learning and use. These theoretical models have been developed by Peter Robinson (the Cognition Hypothesis. Robinson, 2011a) and Peter Skehan (the Trade-off Hypothesis, Skehan, 1998, 2009). The main construct of interest in both models is the construct of task complexity. Skehan and Robinson define this construct in a slightly different ways and they also make somewhat different predictions about the effects that task complexity might have on L2 products and processes. Robinson`s and Skehan`s predictions have been put to the empirical test and this research has produced mixed results. In addition, as the models have been created to account primarily for oral production, it is still an empirical question whether or not the predictions made by the two models can also be applied to written performance. In other words, it is still unknown whether or not task complexity produces the same effects in speech and in writing. Moreover, mode (oral versus written) by itself may constitute an important task feature that can condition language use and learning opportunities. At a more general level, it is also the case that current understanding of the idiosyncrasy of the language-learning potential of oral and written tasks is limited. Given this state of affairs, the objective of the current dissertation is twofold. Our first aim (operationalized as our first research question) is to explore whether or not L2 linguistic performance is different when the same task is performed in speech and in writing. Our second aim (corresponding to our second research question) is to investigate any potential effects of increasing task complexity in oral production as compared to written production. In terms of task complexity manipulations, we employed the Cognition Hypothesis as a theoretical framework and we specifically investigated the impact of the resource-directing variable of reasoning demands on L2 performance, which was operationalized as complexity, accuracy and time on task. To answer our research questions, we conducted a study with 78 participants who were Catalan/Spanish learners of English as a foreign language. Half of the participants performed the simple and complex versions of an argumentative, instruction-giving task orally, the other half did it in writing. In the analysis, we compared speakers` and writers` performance in terms of linguistic (lexical and structural) complexity, propositional complexity (operationalized as idea units), accuracy, and time on task. Our results revealed marked differences between oral and written production. Thus, we found that speakers produced more idea units, while writers achieved higher scores for subordination, mean length of analysis-of-speech units, lexical diversity, extended idea units, and time on task. As for the effects of task complexity, the participants’ written production showed more variation between the complex and the simple versions of the task. Changes in the written production also showed a better fit to the theoretical predictions advanced in the Cognition Hypothesis.