臺灣健康成年人之中文版自由和線索選擇性回憶測驗常模研究

Abstract

情節式記憶缺損是最常見之阿茲海默症（Alzheimer’s disease, AD）的典型表現，但在正常的老化過程中，也常會因注意力降低、處理訊息的效率退化等其他原因導致記憶力表現的下降。自由和線索選擇性回憶測驗（Free and Cued Selective Reminding Test, FCSRT）透過線索回憶的方式控制編碼訊息，並使用相同的語義線索促進訊息提取的過程，以此特性之優點對記憶缺損表現作出更好的區辨。目前此測驗尚缺乏具代表性之臺灣常模資料，故本研究旨在建立臺灣成人族群之中文版自由和線索選擇性回憶測驗（Taiwan version of the FCSRT, T-FCSRT）常模，並探討人口學變項對T-FCSRT測驗表現的影響，同時檢驗其心理計量特性與臨床應用的潛力。本研究共納入372位45歲至86歲健康成人，以臺灣人口組成分層抽樣不同性別、年齡及教育程度族群。結果顯示T-FCSRT各項核心分數（即IFR、ITR、CE、DFR、DTR）皆受年齡及教育程度的顯著影響，相較之下，性別則僅對自由回憶分數（IFR、DFR）產生顯著影響。此外，本研究顯示T-FCSRT具有良好的再測信度、效標關聯效度及建構效度，而為了建立具代表性臺灣人口常模，研究使用多元迴歸模型校正方法，排除人口學變項對測驗分數的影響，以校正分數建立百分等級對照表，提供T-FCSRT作為未來研究與臨床上評估情節式記憶缺損的神經心理測驗工具之ㄧ。

Episodic memory impairment is a typical clinical feature of Alzheimer’s disease (AD), but similar declines in memory performance can also occur during the normal aging process due to factors such as reduced attention and diminished information processing efficiency. The Free and Cued Selective Reminding Test (FCSRT) enhances the differentiation of memory impairment by using controlled encoding through semantic category cues and facilitating retrieval with the same cues, thereby improving the test’s sensitivity in distinguishing between different types of memory deficits. Despite its clinical utility, representative normative data for the Taiwanese population remains unavailable. Therefore, this study aimed to establish normative data for the Taiwan version of the FCSRT (T-FCSRT) in healthy adults, examine the effects of demographic variables on test performance, and evaluate its psychometric properties and clinical applicability. A total of 372 healthy adults aged 45 to 86 years were recruited using stratified sampling to reflect the demographic distribution of the Taiwanese population across sex, age, and education levels. The results revealed that all T-FCSRT core scores including IFR, ITR, CE, DFR, and DTR were significantly affected by age and education level. In contrast, sex had a significant effect only on free-recall measures (IFR and DFR). In addition, the T-FCSRT showed adequate psychometric properties, including test-retest reliability, criterion-related validity, and construct validity. To develop population-representative norms, a multiple regression-based approach was used to adjust scores for demographic variables. Percentile rank tables were then created based on the corrected scores, providing a culturally appropriate and psychometric framework for interpreting T-FCSRT performance. These findings support the T-FCSRT as a valuable neuropsychological tool for assessing episodic memory deficits in both clinical and research contexts.