The Effects Of Working Memory On Second Language Reading Comprehension

Recent years have witnessed a rapid growth of interest in the effects of working memory(WM)on second language acquisition(SLA).As an important input mode of SLA,numerous studies have suggested that reading comprehension is closely associated with WM.However,many previous studies generally regard reading comprehension as a global construct without involving its multidimensional construct such as literal comprehension,inferential comprehension,and evaluative comprehension.Additionally,reading span task(RST)and operation span task(OST),as two common measures of WM in SLA studies,do not evaluate the memory load effectively when measuring WM.It is also unclear whether different memory load will affect second language(L2)reading.Therefore,motivated by previous work,the present study tries to explore the relationship between WM under different memory load and L2 reading comprehension,particularly its specific dimensions of literal comprehension,inferential comprehension and evaluative comprehension.The study aims to answer the following three research questions:(1)Are there any differences among literal comprehension,inferential comprehension,and evaluative comprehension in L2 reading between high-and low-proficiency English learners?(2)Are there any differences in WM under different memory load between high-and low-proficiency English learners?(3)How does WM under different memory load affect literal comprehension,inferential comprehension,and evaluativecomprehension in L2 reading?Fifty college students were recruited in the present study.According to the College English Test Band 4(CET-4)results,they were further divided into two groups including high-proficiency group(CET-4≥475)and low-proficiency group(CET-4<425).Then,all participants were asked to finish a reading comprehension test which was chosen from the multiple-choice reading comprehension(MCRC)measures in curriculum based measurement(CBM)system.A total of 40 questions were written for the test.Of the 40 questions,14 questions assessed literal comprehension,14 questions assessed inferential comprehension,and 12 questions assessed evaluative comprehension.In addition,WM was measured by a modified version of OST in which an n-back paradigm was inserted to control the memory load(n depending on the load level,that is,1,2,or 3),and all participants’ performance was assessed in terms of accuracy and reaction time(RT).Besides,for comparison purposes,each participant’s WM was also measured by using unmodified OST and unmodified n-back task respectively.Finally,statistical analyses were conducted by SPSS 21.0 to evaluate group differences.The study got the following three results:(1)Compared with low-proficiency group,high-proficiency group performed better on all three levels of reading comprehension(literal comprehension,inferential comprehension,and evaluative comprehension)in L2 reading,and much better on inferential comprehension.(2)The WM capacity of high-proficiency group was greater than that of low-proficiency group,and the RT of high-proficiency group was much shorter than that oflow-proficiency group under higher memory load condition(2-back load condition).(3)Under higher memory load condition,the RT of high-proficiency group was significantly negatively correlated with literal comprehension scores and inferential comprehension scores,while under lower memory load condition(1-back load condition),the RT of low-proficiency group showed a significant negative correlation with literal comprehension scores.The findings indicated that different proficiency English learners differed on three levels of reading comprehension,inferential comprehension in particular.Furthermore,the correlation between WM under different memory load and three levels of L2 reading comprehension was different,significantly different on inferential comprehension,suggesting that inferential comprehension probably plays an important role in L2 reading comprehension.Several limitations should be noted in this preliminary study.First,the modified WM measurement paradigm should be further evaluated in a larger sample of participants for the purpose of confirming its reliability and validity.Besides,with regard to the reading comprehension test,further research should address more question types such as short answer questions.To conclude,the study will deepen our understanding of the three levels of reading comprehension and may provide new insights into the relationship between WM and L2 reading comprehension.