A meta-analysis of the relationship between working memory and second language reading comprehension: Does task type matter?

Abstract

Despite the increasing attention paid to the role of working memory in reading, findings and measurement of working memory have been inconsistent. The current meta-analysis aims to provide a quantitative description of the overall relationship between second language (L2) reading comprehension and working memory measured through reading span task and identify methodological features that moderate this relationship. Following a comprehensive search, 25 primary studies (23 peer-reviewed studies and 2 dissertations) were included comprising 37 unique samples (N = 2,682), all of which were coded for substantive and methodological features. The results showed that (a) there is a moderate relationship between L2 reading comprehension and working memory (r = .30), (b) reading span task features such as the scoring procedure, task language, and final word recall order moderate this relationship, and (c) the degree to which working memory’s involvement in L2 reading comprehension may vary depending on the type of reading tasks at hand. Implications are discussed in terms of conceptualization and measurement of working memory. Future directions are also offered in relation to measurement practices to encourage consistency and to improve our understanding of the link between working memory and L2 reading comprehension.

Working Memory Capacity and L2 Reading

“…Between L1-and L2-based WM tasks, various correlations have been reported (e.g., r = .72 and .84, Osaka & Osaka, 1992; r = .39, Harrington & Sawyer, 1992). For the WM-L2 reading relationship, Linck et al (2014) showed no moderating effect across L1 and L2 (r = .228 and .229), whereas Shin (2020) reported that the relationship was stronger in L2 WM tasks (r = .35) than in L1 WM tasks (r = .17). Therefore, it remains unclear whether and to what extent stimulus language moderates the WM-L2 reading relationship.…”

“…They investigated whether these relationships were moderated by variables such as participant characteristics and WM measurement features, concluding that WM capacity has a robust positive relationship with L2 proficiency. Shin (2020) focused on reading span tasks and meta-analyzed correlations between discourse-level L2 reading and WM. A small relationship was found between L2 reading comprehension and WM ( r = .30 [.24, .35], 25 studies).…”

“…These tasks can be finer-grained measures discriminating among participants of similar proficiency or can be developed ad-hoc with few items and low reliability. Shin (2020) reported that the WM–L2 reading relationship was slightly stronger when standardized reading tasks were used ( r = .28) than nonstandardized reading tasks ( r = .24). Whether the results are replicable merits examination.…”

“…Including (a) and (b) allows for considering trade-offs across recalling and processing to better capture WM span. Shin (2020) showed that WM was more strongly related to L2 reading comprehension when storage and processing task scores were used ( r = .33) than when these task scores and reaction time ( r = .30) or only storage scores ( r = .20) were used.…”

The Role of Working Memory in Attentional Allocation and Grammatical Development under Textually-enhanced, Unenhanced and No Captioning Conditions

“…Indeed, a growing amount of empirical research suggests that there is a positive relationship between working memory capacity and L2 outcomes. Results of several narrative reviews (Jackson, 2020;Jeon & Yamashita, 2014;Juffs & Harrington, 2011;Wen, 2016;Williams, 2012) and meta-analyses (Linck et al, 2014;Shin, 2020) indicate that, overall, working memory has robust, positive links with L2 processing and learning outcomes.…”