부산대학교 도서관

Research has shown that high levels of emotion are associated with lowered academic performance. However, the mechanisms involved in this relationship are as yet unclear. One potentially important process is the disruption of working memory. Building on previous research such as Wine (1971) and Sarason (Sarason, 1984), Eysenck and Calvo (1992) formulated a processing efficiency theory (PET) to account for the disruptive effects of anxiety on cognitive task performance. Briefly, PET suggests that anxiety can have the effect of reducing cognitive storage and processing resources in the working memory system. One effect is a reduction in task effectiveness, or the accuracy of performance. H.C. Ellis and colleagues have proposed a similar model, the resource allocation model, with special reference to depressed mood (e.g. Ellis & Moore, 1999). Under both theoretical frameworks, the negative effects of emotion are assumed to be most clearly seen on complex cognitive tasks. In addition, the negative effects of anxiety on performance are thought to be most pronounced under stressful conditions. This thesis explored the relationship between emotion, working memory and academic performance in an attempt to further understand the processes involved in educational underachievement in anxiety and depression. The first empirical investigation reported in Chapter Three was an initial step in testing the simple relationship between three emotions (trait anxiety, depression and test anxiety) and academic performance measures. Consistent with previous research moderate negative relationships were found between the two constructs. Chapter Four showed that working memory partially mediated the negative trait anxiety-academic performance relationship. Chapter Five replicated the mediation hypothesis also incorporating depression and test anxiety. A moderated mediation hypothesis was also tested. That is, that the negative links between emotion, working memory and academic performance were shown to be most pronounced when stress reactivity was high. Chapter Six replicated these moderated mediation results and tested an emotion x working memory interaction hypothesis that suggested that those with high emotion and low working memory were the poorest academic performers. Chapter Seven evaluated the moderated mediation and interaction hypotheses in a longitudinal study which allowed for some preliminary conclusions to be made concerning causality. It was shown that emotion was negatively associated with academic performance via verbal working memory over time when stress reactivity was high. Over time, high emotion and low working memory in combination were associated with lower academic performance. In Chapter Eight, the implications of the findings in this thesis for theory and practice were discussed.