The Role Of Key Visuals In Reducing Working Memory Load During L2Discourse Comprehension

Key visuals are the graphic displays that introduce students to the main ideasof a text and the interrelationship among these ideas. Plenty of research hasdemonstrated that using key visuals to provide learners with backgroundinformation can improve their comprehension for the subsequently presented text.However, the psychological mechanism through which the backgroundinformation functions has not been well researched.The purpose of this study is to investigate the influence that key visuals haveon the second language (L2) discourse comprehension from the perspective ofworking memory (WM) load. WM load refers to the amount of mental activity thatoccurs within WM. The capacity of WM is extremely limited. Owing to thislimitation, comprehension can take place only when WM is capable of holding andprocessing certain mental activities at a given time. An abundance of research hasdemonstrated that WM plays a crucial role in L1comprehension. However,relatively little research has been conducted in the L2field, most of which is on L2reading comprehension. Additionally, most of the previous studies only measuredthe storage component of WM, neglecting the processing ability. For this reason,more research is needed to investigate the role of WM in L2comprehension, inwhich the method of measuring the WM capacity should be improved.Recently, in a few psychological studies, the researchers found that anincrease in the learners’ knowledge decreases the role of WM capacity as the determinant of complex task performance. The knowledge refers to the priorknowledge and the preexisting experiences stored in learners’ long-term memory(LTM). Therefore, the study hypothesized that key visuals may activate theknowledge stored in LTM, which effectively reduces the load on WM.To provide an in-depth understanding of the psychological process underlingof the effects of key visuals, the experiment was carried out in both L2listeningand L2reading comprehension which are assumed to exert different amounts ofload on WM. Two pre-posttests were designed. The participants in this study were90students who were non-English majors in a Chinese university. They first tookthe English listening and reading comprehension pretests which contained a recallprotocol and a sentence verification task. One week later, they took the posttestswith the form similar to the pretests; but in the posttests, the participants wereprovided with the key visuals relevant to the testing passages prior to the listeningand reading comprehension tasks. After the posttests, they took the listening andreading span tests which measured their listening and reading WM capacitiesrespectively. The present study improved on Waters and Caplan’s (1996) span test,using recall span, reaction time and sentence judgment as components of the spantests, to measure both processing and storage components of WM. To furtherexamine the role of key visuals in discourse comprehension, the researcher dividedthe participants into three WM groups, i.e. the high, the mid and the low WMgroups according to their scores in the span tests.The correlation, regression analysis and mixed-model ANOVA wereconducted to analyze the data. The prediction that key visuals would improve L2discourse comprehension by reducing load on WM was supported by the followingfindings: the participants performed better in listening and reading comprehension in the posttests than in the pretests; the participants were less constrained by theirlimited WM capacity in the posttests than in the pretests; differences in thediscourse comprehension performance between the WM groups were smaller in theposttests than in the pretests. These results indicate that the backgroundinformation provided by the key visuals decreases the role of WM capacity as thedeterminant of L2discourse performance. The statistical information also showedthat using recall span, reaction time and sentence judgment as components of WMmeasure was reliable and the source of individual difference in memory capacitydid not reside in the passive storage capacity only. The study has also found thatthe participants’ listening and reading comprehension abilities were constrained bytheir capacity of WM, and that compared with reading comprehension, listeningcomprehension required more WM capacity.However, there were also incompatible findings: in the readingcomprehension posttest, the performance of the mid WM group and the high WMgroup was not improved in the sentence verification task, and there even appearedsome regression in their sentence verification skill. This was explained in the lightof Cognitive load theory and Dual-process theory.The study has both theoretical and pedagogical implications. The findingsdeepen our understanding of the psychological nature of L2listening and L2reading comprehension, and also enrich the psycholinguistic and second languageacquisition theories. Moreover, it suggests that in L2teaching, teachers can utilizesome techniques to reduce the learners’ WM load in comprehension tasks anddesign some teaching programs to help them make up for their WM limitations.