Degrees of Instructional Explicitness, Depth of Processing, Learning styles and L2 Development: A study on the Spanish Imperfect subjunctive

Abstract

The aim of the present study was three-fold: First, it intended to investigate the effects that different types of instruction varying in explicitness (e.g., Rosa & O'Neill, 1999; Rosa & Leow, 2004), had on 88 intermediate-level Spanish learners' development of the Imperfect subjunctive. Second, it aimed to gain some insight into participants´ minds by employing think-aloud protocols to account for the internal processes that occurred while they carried out a dual task (i.e., reading for meaning and form), and to examine how these processes correlated with post-task performance. Finally, its third goal was to probe deeper into participants' Learning styles (Dornyei, 2005), and, more specifically, into their preference for a deductive or inductive approach to learning grammar rules in the L2. According to their preference, a match or mismatch situation was created with learners matching the treatment received (deductive approach = More explicit condition) or mismatching it (deductive approach = Less explicit conditions) to account for the matched/mismatched design's impact on post-task performance (Valtz, Tare, Jackson, & Doughty, 2013). Participants were semi- randomly assigned to one of six experimental conditions differing in explicitness and in whether they thought aloud or not to control for potential reactivity (Bowles, 2010) (+ Explicit +/- Think aloud (TA), - Explicit +/- TA, Baseline +/- TA). The study consisted of 3 sessions and it followed a pre/post/delayed test design with 2 weeks between each of the sessions. Results showed that the Less explicit condition appeared to perform as well as the More explicit condition immediately after treatment and better 2 weeks after treatment for both interpretation and production tests when compared to the Baseline condition performance. With regards to Depth of Processing, the More explicit condition seemed to elicit more instances of deep processing than the Less explicit or Baseline conditions, and deeper processing significantly correlated with higher comprehension and with a more accurate subsequent production of the target form. With respect to Learning styles, the mismatch condition did positively correlate with better performance at Interpretation immediately after treatment, unlike the matching condition, which failed to yield any positive correlation with post-task performance.