Dissertation/Thesis Abstract

Concurrent verbalization, task complexity, and working memory: Effects on L2 learning in a computerized task
by Medina, Almitra Dadin, Ph.D., Georgetown University, 2008, 341; 3315458
Abstract (Summary)

The field of SLA has witnessed recently an increased effort in operationalizing and measuring attention and awareness through the use of concurrent verbal reports (think-aloud protocols); however, the argument that the provision of these reports poses an additional task or alters cognitive processes—the issue of reactivity—has motivated very recent investigations to examine the effect of thinking aloud during task completion. The role of verbalization in L2 learning is still an open empirical question, however, especially in terms of variables that may moderate the findings.

The current study investigated reactivity of concurrent verbalizations and sought to uncover whether think-aloud findings are influenced by task complexity and working memory (this latter as measured by a sentence span and computation span test, analyzed individually) and examined whether these variables have implications for L2 learning.

Participants were intermediate Spanish learners randomly assigned to one of four groups, depending on verbalization condition (think aloud vs. silent) and task complexity (more vs. less complex).

Results show that thinking aloud had no effect, neither positive nor negative, on L2 learning, as measured by a recognition and written production test, and that this lack of reactivity on accuracy was not altered by the complexity of the experimental task nor was related to participants' working memory (WM). Thinking aloud was suggested to have changed cognitive processes to some degree, however, as reactivity surfaced for latency, such that concurrently verbalizing thoughts delayed task completion time. This change, nonetheless, was not enough to have a substantial impact on learning.

Further results show that task complexity was found to have a significant effect on the written production of the target structure (the imperfect subjunctive in Spanish), as those who engaged in the more complex task were better able produce the imperfect subjunctive in writing than those in the less complex task. These results remained significant even when WM was accounted for. And finally, WM as measured by the computation span was found to significantly inference recognition, but not production; that is, higher WM was related to better accuracy in recognizing the target structure, supporting the 'higher is better' hypothesis in just this one case.

Indexing (document details)
Advisor: Leow, Ronald P.
School: Georgetown University
Department: Spanish & Portuguese
School Location: United States -- District of Columbia
Source: DAI-A 69/07, Dissertation Abstracts International
Subjects: Linguistics
Keywords: Computerized, Concurrent verbalization, Second language, Task complexity, Working memory
Publication Number: 3315458
ISBN: 978-0-549-66279-2
Copyright © 2021 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy