ABSTRACT
Historically, accuracy has served as a key source of evidence in the elaboration of Second Language Acquisition (SLA) theories since the beginning of the field in the 1960s. Accuracy was a main construct within the Contrastive Analysis Hypothesis which existed in a strong and a weak version. The first core issue assesses the place of accuracy within a number of current SLA theories, mainly Usage-based Theory, Complex Dynamic Systems Theory, and Interactionist Theory. The second core issue presents how accuracy has been studied in learner corpus research, while the third discusses some of the challenges involved in studying accuracy in spoken learner corpus data. Detecting errors in spoken learner corpus data is further complicated by the presence of disfluencies which can be difficult to distinguish from errors themselves. The rich metadata provided in learner corpora make them an ideal testbed to study the combined impact of multiple independent variables on accuracy profiles.
