Learning opportunities and outcomes of L2-L2 and L2-HL learner interaction during a collaborative writing task

Task-based pair work encompasses comprehensible input, negotiation of form and meaning, and modified output, thus affording learners with optimal opportunities for language learning. Specifically, exchanges in which interactants turn their attention to certain linguistic targets within meaningful interaction, called form-focused episodes (FFEs), are believed to promote language learning. A growing body of research on learner-learner interaction has shown that learners internalize the information discussed in FFEs, as measured by tailor-made posttests (Adams, 2006, 2007; LaPierre, 1994; Swain & Lapkin, 1998). However, there has been a mismatch to date between the meaning-based nature of the treatment tasks and the exclusively form-focused nature of the tailor-made posttests used to assess learning gains. Furthermore, research has focused almost exclusively on L2-L2 pairs, leaving aside the question of what happens when heritage language (HL) learners are involved.;This dissertation examines not only the connection between learner-learner interaction and language learning, but also whether learner linguistic background (L2, HL) plays a differential role in terms of both learning opportunities and outcomes. Learning opportunities were operationalized as the FFEs that arose in the task-based interaction of matched (L2-L2) and mixed (L2-HL) dyads, and learning gains were assessed through the incorporation of linguistic information from successfully resolved FFEs in immediate and two-week delayed post-treatment individual writing tasks. Participants were 24 L2 learners of Spanish (English native speakers) and 8 HL learners (bilingually-raised Spanish/English speakers) enrolled in the same intermediate Spanish course. Research questions were addressed by analyzing the audio-recorded interactions of 8 L2-L2 and 8 L2-HL dyads engaged in a collaborative writing task, as well as the texts produced by the learners during and after the treatment.;Results indicated that there were no differences between matched and mixed dyads in terms of the total number of FFEs, the frequency of preemptive and reactive FFEs, and the occurrence of morphosyntactic FFEs. By contrast, L2-L2 and L2-HL dyads differed with respect to frequency of orthographic and lexical FFEs and successful resolution of FFEs. Within mixed dyads, L2 and HL learners differed in terms of frequency of initiation of FFEs as well as the type of FFEs initiated. Nevertheless, both L2 and HL learners were able to correctly resolve FFEs equally as often. With regards to learning gains, L2-L2 dyads incorporated information from the FFEs in both the immediate and the delayed post-treatment individual writing tasks significantly more often than learners in L2-HL dyads. Within mixed dyads, L2 learners were more likely to use linguistic information provided by their HL peers than the other way around in the immediate post-treatment writing task. Overall, considering the greater learning gains of L2-L2 dyads and of L2 learners in L2-HL dyads, the data suggest that HL learners may not benefit from the interaction as much as their L2 counterparts, although benefits are not entirely one-sided in favor of L2 learners.