臺灣健康成年人於修訂版家庭圖片測驗中主體與情境脈絡相關記憶之常模研究

Abstract

主體記憶與情境脈絡記憶在日常生活中皆扮演重要角色，然而過去對於情境脈絡記憶的研究較少，且多半使用缺乏生態效度的刺激材料。家庭圖片測驗以生活化的材料測量對於角色、地點及活動的記憶，為既有之能夠同時涵蓋主體與情境脈絡記憶兩面向的測驗工具。然而現行之家庭圖片測驗採用角色基準分數、僅計單一總分、且無提供年齡以外之人口學變項研究資料，無法直接評估對主體與情境脈絡記憶之表現，亦不利臨床使用。有鑒於此，本研究欲改善現行計分系統，並發展個別分數之常模，主要目標為：一、探討角色、地點及活動分數之可分離性；二、探討發展人口學變項校正常模之必要性；三、評估各分數之信度及其對應主體與情境脈絡記憶功能之效度；四、本常模資料之適用性。本研究收集臺灣232名20至88歲、教育程度0至18年之健康成年人於修訂版家庭圖片測驗之常模資料，其中86名受試者同時接受效標測驗。半年後隨機選取30名受試者進行再測以考驗信度。此外比較臨床病人在本常模之表現以檢驗效度。結果顯示：一、角色、地點及活動分數之正確率有顯著差異，隨年齡增加而降低的程度亦不同，故應分離探討之；二、年齡為主要預測測驗表現之人口學變項，但性別及教育程度也有顯著影響，故有發展人口學變項校正常模之必要；三、本測驗具良好之信度；效度部分則顯示角色及地點分數符合預期，分別對應主體與情境脈絡記憶之構念，然活動分數之特性需要後續研究進一步探討；四、本常模資料具有良好之代表性、新近性及適切性。本研究修改家庭圖片測驗，建立適用於評估主體與情境脈絡記憶功能的常模，並提供校正分數百分等級對照表，供臨床工作者於早期偵測、鑑別診斷、預後評估及認知復健方向應用。未來應持續增加常模樣本數，並收集不同臨床病人組之測驗表現，以累積實徵的效度研究資料。

Background: Both content and context memory play important roles in daily life while few past studies focused on the latter one and their task materials were lack of ecological validity. The Family Pictures subtest is an available tool for measuring three major real-life information (character, location, and activities) which correspond to the concept of content and context memory. However, the limitations of the current Family Pictures restrict the utility for evaluating the content and context memory performances. Objective: The study was to modify the Family Pictures and provide norms for content and context memory measures. The specific objectives were to explore: (1) The dissociability of the character, location, and activities memory aspects. (2) The necessity of demographic-corrected norms for the three measures of content- and context-related memory. (3) The issue of psychometric properties. (4) The issue of norm appropriacy. Methods: The present study collected 232 participants and used stratified sampling by current age (ranging from 20 to 88 years old), gender, education (ranging from 0 to more than 16 years), and area of residence. Eighty-six participants meanwhile received the criterion-related neuropsychological tests. Six months later, 30 participants of the norm groups received re-test year for reliability. Clinical patient group was also collected for verifying the validity. Results: (1) These sub-scores, respectively measuring the character, location, and activities memory aspects, of the modified Family Pictures are dissociable. (2) Not only age but also gender and education have influence on each sub-score, and therefore demographic-adjusted norms are needed. (3) The three sub-scores have good reliability. The results of validity indicated that the character and the location sub-scores separately correspond to the concepts of content- and context-related memory. More studies are needed to clarify the properties of the activities sub-score. (4) The normative data have good representativeness, recency, and relevance. Conclusion: The present study modified the Family Pictures subtest, established an appropriate normative data for evaluating the content and context memory function, and provided the referential table for percentile ranks. It is useful for early detection, differential diagnosis, prognosis evaluation, and cognitive rehabilitation. The further work on a large scale aims at normative collection and validity study in patients with brain diseases.