Session 5A - Colloquium

Understanding the ways in which instruction can contribute to L2 development constitutes a major line of SLA research to date (DeKeyser 2017; Leow 2015). This work indicates that explicit information (EI) can increase sensitivity to L2 items of low saliency and low reliability (Cintrón-Valentín & Ellis 2016) and systematic, deliberate practice can drive the development of different types of L2 knowledge (e.g., proceduralized, automatized, DeKeyser 1997). In addition, the effectiveness of EI and practice can vary according to the nature of the learning problem (Henry et al. 2009; McManus & Marsden 2017, 2019), indicating that not all types of EI are the same and not all opportunities for practice develop automatic L2 knowledge. Furthermore, the types of L2 knowledge developed from different instructional components (e.g., practice with and without EI) is likely not the same (Marsden & Chen 2011). Very little research, however, has examined the types of L2 knowledge developed following particular instructional components and the extent to which this knowledge changes over the course of the practice (DeKeyser 1997; McManus & Marsden 2019). One study investigating the ways in which of EI contributed to performance over the course of the practice is Fernández (2008). EI about subject-object pronouns and word order in Spanish was predicted to increase the accuracy and speed of English speakers’ L2 processing of object-verb-subject and subject-object-verb sentences, but no benefits for EI were found. One explanation for this is that the EI did not adequately address the nature of the learning problem: L1-L2 differences for subject-object information. The Unified Competition Model (UCM, MacWhinney 2012) predicts learnability problems when languages express the same meaning(s) in different ways. Fernández’s target features exemplify this learning problem: subject-object information is indexed using word order in English (e.g., Mary greets her friend), but with subject-verb agreement, animacy, direct object placement, and case marking a in Spanish. Furthermore, even though practice with and without EI indicated no differences, it is possible that these different treatments led to the development of different types of L2 knowledge (e.g., practice without EI potentially recruits inductive learning processes, Marsden & Chen 2011). The present study addressed this crosslinguistic learning problem by partially replicating Fernández (2008). We added a new treatment that included EI about the L1 in order to understand (i) how variations in the type of EI influenced performance and (ii) the extent to which type of EI led to the development of different types of L2 knowledge. Fernández’s design, procedures, and materials were used. Intermediate-level English-speaking learners of L2 Spanish (n= 111) were randomly assigned to one of the following treatments: (a) practice interpreting OVS and SVO sentences in Spanish (as in original); (b) EI about L2 with the same L2 practice (as in original); (c) EI about L2 and L1 and the same practice. The new L1 EI was designed following the UCM to focus attention on L1 cues expressing subject-object information. Results replicated the original study’s findings for accuracy and speed: L2 EI provided no additional benefits in terms of accuracy and speed of processing. However, learners receiving EI about L1 showed faster and more accurate processing. Coefficient of variation analyses (Segalowitz & Segalowitz 1993) were carried out to understand the nature of learners’ L2 knowledge and its development following the different instructional treatments, which indicated knowledge restructuring in the new group suggesting gradual automatization of L2 knowledge over time (Solovyeva & DeKeyser 2018). These results indicate that raising learners’ awareness of L1-L2 differences benefitted performance and the type of L2 knowledge developed. These findings are discussed in light of previous research. Implications for classroom foreign language learning are discussed.