身體活動介入對執行功能影響之後設分析

Abstract

近年來身體活動領域之研究焦點已從早期的基礎認知功能，轉向對執行功能之成效探討。然而，目前雖已累積大量的實驗研究，但結果卻分歧而難以形成共識，造成研究者解釋的困難，也使實務運用者在選用介入方案時無所適從。因此，本研究之目的為採取後設分析法，完整蒐集與統整過去相關研究，以探討身體活動介入對執行功能影響之成效。同時，亦探討不同身體活動介入特徵、參與者特徵、實驗設計、測量向度，對執行功能成效的影響。 經由國內外電腦資料庫(PsycArticles、ERIC、SPORTDiscus、MEDLINE、ASP、臺灣期刊論文索引系統)搜尋、並複核過去後設分析研究與回顧性文獻所提及之研究報告，共計94篇文獻符合本研究之納入標準。接著，依據Hedges與Olkin(1985)所發展之技術進行各篇文獻之效果值整併與調節變項分析，並以Rosenthal(1991)之Fail-safe N方法，檢驗本研究結果受出版偏誤影響之情形。 結果顯示，身體活動介入對執行功能具有顯著低度正向效果(d+=0.205)，且受出版偏誤影響不大。另外，調節變項分析結果則顯示，不同身體活動介入特徵、參與者特徵、實驗設計、與測量向度均能有效增進執行功能。其中，以「參與者年齡」和「測量向度」為影響介入效果之重要調節變項。就「參與者年齡」而言， 6~26歲參與者，隨著年齡增加，介入效果降低。就「測量向度」而言，身體活動介入對計畫/排序的影響，顯著高於工作記憶。其餘調節變項雖未達統計顯著性，但仍呈現可能趨勢。最後，本研究根據研究結果與討論，對未來研究者與實務運用者提出相關建議。In recent years, the research area of physical activity has shifted in focus from basic cognitive function to higher order cognitive function—executive function. Although there is a substantial body of literature related to the effects of physical activity on executive function, the research findings are mixed. It’s not only difficult to explain by researcher, but also hard to choose an appropriate intervention program by practitioner. Thus, the purpose of this meta-analysis was to provide an updated comprehensive analysis of the extant literature on physical activity and executive function and explore the effects of moderators which include physical activity characteristics, participant characteristics, experimental designs, and types of executive functions. Searches of electronic databases (PsycArticles、ERIC、SPORTDiscus、MEDLINE、ASP、index to Taiwan Periodical Literature System) and examinations of reference lists from relevant review articles resulted in 94 studies meeting inclusion criteria. An overall effect size (ES) and average ESs relative to moderator variables were then calculated according to Hedges and Olkin’s method (Hedges& Olkin, 1985). In addition, the possible impacts of publication bias were analyzed by a Fail-safe N analysis (Rosenthal, 1991). The overall ES was significantly positive and small (d+=0.205, k=201). A Fail-safe N analysis was applied, concluding that the meta-analysis didn’t suffer much from publication bias. Positive effects were also found in all physical activity characteristics, participant characteristics, experimental designs, and types of executive functions. Examination of potential moderators indicated that participant’s age and types of executive functions were significant moderators. For the participant’s age moderator, there is a positive association between participant’s age and ESs when the participants were 6 to 26 years old. For the types of executive functions moderator, the effect of physical activity on planning/sequencing is significantly larger than the effect on working memory. Although other moderator variables didn’t reach significance, they still reveal potential trends. Finally, some suggestions were provided for future practitioners and researchers.