Focused Tasks, Mental Actions and Second Language Learning. Cognitive and Connectionist Accounts of Task Effectiveness

This paper presents a theoretical framework to estimate the effectiveness of second language tasks in which the focus is on the acquisition of new linguistic items, such as vocabulary or grammar, the so-called focused tasks (R. Ellis, 2003). What accounts for the learning impact offocused tasks? We shall argue that the task-based approach (e.g. Skehan, 1998, Robinson, 2001) does not provide an in-depth account of how cognitive processes, elicited by a task, foster the acquisition of new linguistic elements. We shall then review the typologies of cognitive processes derived from research on learning strategies (Chamot & O'Malley, 1994), from the involvement load hypothesis (Laufer & Hulstijn, 2001), from the depth of processing hypothesis (Craik & Lockhart, 1972) and from connectionism (e.g Broeder & Plunkett, 1997; N. Ellis, 2003). The combined insights of these typologies form the basis of the multi-feature hypothesis, which predicts that retention and ease of activation of new linguistic items are improved by mental actions which involve a wide variety of different features, simultaneously and frequently. A number of implications for future research shall be discussed.