Recent theoretical advances viewing language aptitude as multifaceted have led to accumulated evidence on certain constructs beyond the specifications of traditional aptitude tests. In this chapter, we consider the potential role of implicit statistical learning (ISL) mechanisms in second language acquisition (SLA). First, terminological and conceptual issues are examined with the goal of clarifying the processes and knowledge that ISL tasks are claimed to involve. Second, we consider justifications for research into the potential link between ISL and SLA by reviewing the argument for a default implicit processing mode in adult L2 learning. Third, we provide a synthesis of studies assessing the link between ISL and L2 outcomes based on the following research questions: (1) Which studies have examined the association between ISL tasks and L2 outcomes and what are their characteristics? (2) Which measures have been used to investigate the construct of ISL? And (3) What are the accumulated findings of the studies? The overall results of a Bayesian meta-analysis confirmed a positive, albeit weak, link between ISL and SLA (r =.13, [.05,.22]). Moderator analyses and a critical review of ISL tasks suggest avenues for future research.

The current study investigated to what extent two aptitude components, one for explicit and the other for implicit learning, could predict the acquisition of English grammatical structures by late L2 English learners in a naturalistic acquisition context. Sixty-five L2 English learners of Chinese Mandarin, as well as a group of English native speakers, performed a visual-world eye-tracking task. This task involved real-time processing of two grammatical properties of the English nominal phrase that differ in terms of L2 psycholinguistic difficulty: (1) definiteness (a difficult structure) and (2) mass–count (an easy structure). Predictors were implicit learning aptitude, measured by the serial-reaction time (SRT) task, and explicit learning aptitude, measured by subtests of the Modern Language Aptitude Test (MLAT Part 4) and the LLAMA_F test. The overall findings suggest that explicit and implicit learning mechanisms are recruited differentially for learning different grammatical properties.

This chapter explores how tasks connect with psycholinguistic perspectives. Two preliminary sections will explore the nature of learning (and the contrast between explicit and implicit approaches) and the general options available in task-based instruction (tasks themselves, and then task conditions). Then the two areas, learning and task-based instruction, will be related to one another, to explore whether a task-based approach is more consistent with explicit or implicit learning. The general conclusion is that a task-based approach is extremely flexible, and consistent with either. The chapter then moves on to consider second language task-based performance, and its measurement, describing and contrasting the Limited Attention Capacity Hypothesis and the Cognition Hypothesis/SSARC model.

This chapter focuses on three dimensions of tasks we consider essential to task-based language learning. First is their meaningful dimension, which brings learners into contact with patterns and functional uses of a wide range of language features. It also provides them the context for interpreting grammatical features and unlocking relationships between them, and for investing new language with their own shared socio-cultural meanings. Second, the accessibility of new language is promoted by a combination of implicit task-based processes and explicit classroom processes. Thirdly, we adopt Spada and Lightbown’s concept of integration to bring together the meaningful and accessibility principles, with tasks providing the context for both.