This chapter begin with a general definition of implicit and explicit CF and the classification of common CF strategies accordingly. However, emphasis is placed on the fact the individual strategies can vary in their degree of implicitness and explicitness and also in terms of whether they are input-providing or output-prompting. The significance of both cognitive-interactionist and sociocultural theories for predicting the effectiveness of implicit and explicit CF is considered next. The bulk of the chapter reviews the results of research, starting with meta-analyses of CF studies and then moving on to consider descriptive classroom-based studies and comparative experimental studies. Recasts (which can vary markedly in how implicit or explicit they are) have been found to be the most popular type of teacher feedback strategy but not in every context. Explicit CF results in higher levels of repair of errors. However, because of the numerous factors that affect the impact that CF has on acquisition (for example, whether or not noticing occurs, the grammatical target of the correction, and the intensity of the correction), it is not surprising that the results of the research are quite mixed and it is not possible to say which type of feedback is overall the more effective. The chapter concludes with some tentative generalizations and a plea for longitudinal studies of implicit/ explicit CF.

Chapter 11 examines what evaluation studies tell us about the effectiveness of TBLT. These studies are practice oriented, addressing whether TBLT ‘works’ and what might be done to make it work more effectively. The chapter begins with a discussion of TBLT as an innovation and illustrates how the factors that affect the success of educational innovations in general can predict and explain the success or failure of particular TBLT programmes. The chapter then reviews a number of actual evaluations – both macro-evaluations of complete TBLT programmes and micro-evaluations carried out by teachers implementing specific tasks in their own classrooms. Drawing on these evaluations, the chapter ends with a discussion of the major problems that teachers face in implementing TBLT and suggests some solutions.

The first chapter returns to a number of key questions (such as the place of real-life and pedagogic tasks in a task-based course) raised in Chapter 1 and addressed throughout the book. It aims to provide a balanced position on these issues. The second part addresses a number of challenges facing TBLT – theoretical challenges, such as how to develop a model of task performance and long-term acquisition; research challenges, such as the need for longitudinal studies of tasks; and pedagogic challenges, such as the use of TBLT in different instructional contexts. In this way, this chapter looks back and also forward, identifying issues that have figured in TBLT to date and issues that must be addressed in the future.

This chapter presents theory and research that examine tasks in relation to the cognitive processes involved in L2 production in what we have called the Psycholinguistic Perspective. The chapter explores and critiques two models of task-based performance - the Limited Attention Capacity Hypothesis and the Cognition Hypothesis - which have informed a large body of research. The chapter reviews studies that investigated how task design and implementation variables impact on the complexity, accuracy, lexis and fluency of the learners’ production. The chapter also considers a key issue for TBLT, namely the relationship between task performance and L2 acquisition.

Individual learner factors play an important role in how a task is performed. This requires a perspective that draws on the theory and research that addresses the psychology of the learner – what we call the psychological perspective on TBLT. Chapter 5 surveys the large body of research on the role of cognitive aptitudes (including working memory) in mediating the effects of different instructional tasks on language performance and acquisition. The chapter will also discuss the influence of affective factors such as motivation and language anxiety on task performance and outcome. A key focus of this chapter is how these psychological variables mediate the performance of a task and the learning that results.

Chapter 4 offers a sociocultural perspective on tasks. In sociocultural theory a ‘task’ is viewed an artefact for mediating learning through interaction. It views ‘task’ as always interpreted by the participants so that what the task is intended to achieve (i.e. the task-as-workplan) may well not match what the task actually achieves when it is performed (i.e. the task-as-process). Like the cognitive-interactionist perspective, the sociocultural perspective views tasks in terms of the interactions they give rise, emphasizing the importance of the collaborative nature of the interaction for ‘learning’ (defined as other-regulation) and for ‘development’ (defined as self-regulation). It reviews a range of research that has investigated tasks from a sociocultural perspective, including studies involving ‘languaging’, dynamic assessment and Concept-based Language Instruction.

This chapter presents the theory and research related to the cognitive-interactionist perspective on TBLT. It examines how different kinds of tasks create opportunities for interaction that foster the processes involved in second language (L2) and thereby highlight the importance of social interaction for TBLT. It addresses the role that the negotiation of meaning and form play in the implementation of tasks and how negotiation is achieved through interaction, especially when there is corrective feedback. This chapter also examines to what extent how interaction fosters acquisition when tasks are performed. It concludes with an evaluation of this approach to investigating tasks, pointing out both its strengths and weaknesses.

Chapter 8 considers the methodology for implementing tasks. It identifies a range of options relating to each of the three phases of a task-based lesson – the pre-task phase, the main task phase and the post-task phase. Pre-task options have three purposes; (1) to motivate students to perform the task, (2) to prepare them to perform it and (3) to encourage the use of strategies that will help them. Special attention is given to pre-task planning and the various ways in which this can be carried out. The key option in the main task phase is the within-task focus on form. Various ways of accomplishing this are considered – in particular corrective feedback. Of the post-task options, asking learners to repeat a task has attracted most attention from researchers. The chapter does not seek to be prescriptive but it does point to particular options that research has shown to be effective.

How effective is TBLT in comparison to more traditional approaches to language teaching? This chapter seeks an answer to this question by reviewing studies that have investigated the relative effectiveness of different approaches. Overall, these studies do point to the superiority of TBLT in a variety of instructional contexts – e.g. state schools in India and English-for-specific purpose courses in the USA – but there is a clear need for more studies. Comparative method studies are notoriously difficult to design and often suffer from design flaws so the chapter concludes with some guidelines for the design of future studies.

This chapter examines four different proposals for designing a task-based course. In Prabhu’s Communicational Language Teaching Project, the syllabus has low internal structure, leaving implementation issues to be decided by teachers. In Long’s syllabus, tasks have a training function. Target tasks are identified by a needs analysis and then restructured into pedagogic tasks. Robinson’s is the most ambitious proposal as he proposes a syllabus that takes into account the cognitive complexity of tasks, their propensity for promoting the kinds of interaction that facilitate acquisition, and the cognitive abilities and affective dispositions of individual learners. Finally, Ellis identifies a range of factors that influence task complexity but suggests that sequencing tasks is largely a matter of intuition that can be guided only roughly by these factors. In Ellis’s proposal there is room for a more traditional, structural module to fit alongside a task-based module in a complete course.