Contents lists available at ScienceDirect Journal of Second Language Writing journal homepage: www.elsevier.com/locate/jslw The associations between working memory and the efects of four diferent types of written corrective feedback Shaofeng Li a,⁎ , Saeed Roshan b a Florida State University, Tallahassee, FL 32306, US b Auckland University of Technology, Auckland 1010, New Zealand ARTICLEINFO Keywords: Written corrective feedback Working memory Second language Acquisition ABSTRACT Working memory has been posited to play an important role in afecting the processes of writing (Kellogg, 1996). However, to date there has been limited research on its role in second language writingandnoresearchonwhetheritisassociatedwiththeefectsof written corrective feedback in second language learning. This study examines the associations between two types of working memory—complex working memory and phonological short-term memory—and the efective- ness of four types of written corrective feedback: direct corrective feedback, direct corrective feedback plus revision, metalinguistic explanation, and metalinguistic explanation plus revision. Seventy-nine adult learners of English as a foreign language performed three writing tasks and took two working memory tests: a reading span test (complex working memory) and a non-word span test (phonological short-term memory). The results showed that (1) complex working memory was a positive predictor of the efects of metalinguistic explanation and the efects of direct corrective feedback plus revision, and (2) phonological short-term memory was a negative predictor of the efects of direct corrective feedback plus revision. The results suggest that the role of working memory varies as a function of feedback type and that complex working memory and phonological short-term memory may have opposite associations with the efectiveness of written feedback. 1. Introduction Written corrective feedback (CF) refers to responses to second language (L2) learners’ linguistic errors in their written production. The contribution of CF to L2 development has been the focus of a growing number of studies in L2 research over the last 20 years. To date, the bulk of the research has investigated whether learners’ L2 writing performance improves after receiving written CF, and whether the efectiveness of written CF varies as a function of CF type and error type (e.g., Bitchener, 2008; Rummel & Bitchener, 2015; Van Beuningen, De Jong, & Kuiken, 2012). However, there has been limited investigation into the role individual diference variablesplayinlearners’processingofwrittenCFandthelearningthatresults.Thus,therehasbeenacallformoreresearchintothe moderating efects that individual diference factors have on how learners respond to and use the written CF they receive (Ellis, 2008).SeveralstudieshaveexaminedwhethertheefectsofwrittenCFarerelatedtolearners’individualdiferencesinlearnerbeliefs (Rummel, 2014), analytic ability (Shintani & Ellis, 2015),andanxiety(Sheen, 2011).However,nostudyhasexaminedtherelationof working memory to the efectiveness of written CF, which is somewhat surprising in light of the importance of working memory in second language learning (Wen, Mota, & McNeill, 2015)andinthevariousprocessesandcomponentsofwriting(Kellogg, Whiteford, https://doi.org/10.1016/j.jslw.2019.03.003 Received 26 September 2018; Received in revised form 3 March 2019; Accepted 3 March 2019 ⁎ Corresponding author. E-mail addresses: sli9@fsu.edu (S. Li), sroshan@aut.ac.nz (S. Roshan). Journal of Second Language Writing 45 (2019) 1–15 Available online 13 April 2019 1060-3743/ © 2019 Elsevier Inc. All rights reserved. T