博物館網站與使用者認知風格之研究：以國立臺灣師範大學數位校史館為例

Abstract

認知心理學與眼動行為的交互作用研究風起雲湧於1970年代中期，而網站資訊結構亦於1976年由Richard Saul Wurman所提出，本研究立基於此三大面向，以國立臺灣師範大學數位校史館網站為研究個案，進行博物館網站與使用者認知風格之大規模研究，有效樣本為127人。 研究工具採用1985年Childers、Houston與Heckler依據雙碼理論發展的認知風格量表(SOP scale)以及國立臺灣師範大學電機工程學系何宏發教授2012年自行研發之眼動儀(EyeNTNU-180)。研究方法則先以問卷調查法，探索校史館網站使用者之文字/圖像認知風格，以進行分群，後以實驗法進行網站測試，自變項包含：受試者文字/圖像認知風格、網站圖文配置與任務類型，網站圖文位置分為左圖右文與左文右圖兩種版本，任務類型則包含已知項目搜尋與探索性搜尋兩種任務；依變項則包含任務達成效率（答題秒數）與效能（正確性）、以及瀏覽歷程和眼動數據之開始時間點、第一次凝視持續時間、總凝視時間與凝視次數等七項。 具體而言，本研究發現共有十項：(1)圖像型認知風格者亦會聚焦於文字資訊，(2)文字型認知風格者對於文字資訊處理較快易聚焦，(3)開始時間點易關注於情境導覽列文字區，(4)左方資訊較易吸引使用者聚焦，(5)已知項目搜尋對受試者而言較簡單明確，(6)探索性搜尋任務瀏覽歷程視覺化呼應Pirolli與Card於1999年提出的資訊覓食理論，(7)開始時間點關注於其他區呼應Marchionini於1997年提出的個體資訊架構概念元素理論，(8)第一次凝視持續時間持續聚焦於文字資訊，(9)總凝視時間和第一次凝視持續時間相呼應，(10)對有興趣與資訊需求的區域，受試者有較多的凝視次數。茲彙整三項研究結果如下： 一、 校史館網站使用者的圖文認知風格與網站圖文配置交互影響使用者的任務達成效能。 二、 校史館網站使用者的圖文認知風格與網站圖文配置交互影響使用者的任務達成效率。 三、 校史館網站使用者的圖文認知風格與網站圖文配置交互影響其眼動行為。 針對研究結果對於個案網站提出五項建議：(1)視覺化資訊結構，幫助使用者辨識網頁資訊；(2)明確網站資訊，重組組織系統；(3)提高語意明確性、標示系統再進化；(4) 新增輔助導覽系統，建立方向感；(5)增加站內搜尋功能，強化搜尋系統；期能提升網站好用性，進而協助使用者創造更高的使用效率與效能。

The reciprocal effect study of cognitive psycology and eye movement arose from the mid-1970s, and Richard Saul Wurman at the American Institute of Architecure conference of 1976 coined the term information architecture. Bulid on above theorys, the aim of this research was focus on the reciprocal effect of museum website display and user’s cognitive style, and how to affect user’s eye movement, and task accomplishment ratio. To address this objective, a multimethod research design was employed that involved questionnaire survey and eye tracking, and the research tools was applied the Childers, Houston, & Heckler (1985) Style of Processing (SOP) Scale and the Hong-Fa Ho, Chien Yi Liu (2012) eye-tracker (EyeNTNU-180). Moreover the user’s cognitive style, website display, and search tasks was set as independent variable, and task accuracy, finish time, task process, first fixation latency, first fixation duration, total contact time, and number of fixation was set as dependent variable. Specifically, the research findings and results came from 127 participants using two kinds of the National Normal University Archives Websites prototypes. Our main findings have shown ten phenomenons as follows: (1) Imagers also focused on the verbal information. (2) Verbalisers processed verbal information more efficiently. (3) Based on first fixation latency, at the beginning of the tasks users tended to focus on the verbal information in the contectual navigation. (4) Based on total contact times and number of fixation, the left side information was more attractive to users. (5)To users, known-item search task was much easier. (6) Exploratory search task was echoed Pirolli, P., & Card, S.(1990) Information foraging Theory. (7) Based on first fixation latency, most users tended to focus on the others sections on the website echoing Marchionini (1997) personal information infrastructure components theory. (8) Based on first fixation duration, users tended to continue fixate on verbal information. (9) Total contact time was echoed first fixation duration. (10) Based on number of fixations, becasued of interest or information needs, the users’ number of fixations was increased. In conclusion, the research results indicated that users’ cognitive style and university archives websites display will reciprocal affect task accomplishment effectiveness and efficiency, also influence users’ eye movement. Consequently, we propose five suggestions for improvement the National Taiwan Normal University Archives website as follow: (1) Visualize the information architecture to help users recognize. (2) Reorganize the organization system to clear and recategorize. (3) Rename the labeling system to fit in with users’ nature language. (4) Add the supplemental navigation system to build up users’ sense of direction. (5) Set up the search system to help users’ search information. Hope above suggestion can help improve university archives website usability, furthermore make progress in users’ search efficiency and effectiveness.