電子報資訊圖片呈現方式與閱聽人認知型態對閱讀成效、閱報時間與使用態度之影響

Abstract

近年網際網路的迅速擴張，帶動報紙電子化的浪潮，然而，電子報出版人 對於如何建構一項親和易用的電子出版品尚在摸索階段。本研究即從訊息 設計的角度，探討資訊圖片(information graphic)在電子報的閱報環境 中，對不同認知型態(cognitive style)讀者的影響，其結果可作為未來 電子報建立個人化(customized)訊息接收環境時的設計依據。本研究採4 5 4的二因子實驗設計，兩個自變項分別是：「資訊圖片的呈現方式」以 及「使用者的認知型態」。資訊圖片的呈現方式分為四種：「無圖」、「 簡單的資訊圖片」、「加入裝飾的資訊圖片」和「隱藏的資訊圖片」；使 用者的認知型態則分為四類：「語文導向」、「視覺導向」、「高語文－ 視覺」以及「低語文－視覺」。依變項有四：閱讀成效、閱報時間、圖片 使用次數與使用態度。參與實驗的受試者為國立交通大學大學部及研究所 的學生。實驗共分兩階段進行：第一階段以Kirby、Moore與 Schofield(1988)所修訂的「語文－視覺認知型態量表」(Verbalizer- Visualizer Questionnaire, VVQ)測驗結果將受試者區分為四種認知型態 ，受測人數共兩百六十八人。然後，研究者再從中徵求自願者一百七十六 人參與第二階段實驗，依其認知型態隨機分派到四種資訊圖片呈現組別中 ，實際上機閱讀一份模擬的電子報。受試者閱畢後接受離線的閱讀成效測 驗並回答一份使用態度問卷；電腦並透過程式記錄使用者的瀏覽歷程與時 間。本研究的主要發現為：一、在閱讀成效上：當新聞訊息重複出現在文 字與圖片中時，資訊圖片有助於該訊息的記憶。二、在閱報時間上：圖片 加入裝飾後，會增加使用者的瀏覽時間（本實驗中，各組下載圖片的時間 幾乎沒有延遲）。另外，低語文－視覺組的受試者瀏覽時間長於語文導向 的受試者。三、在使用態度上：1.各組皆對圖片表示高度歡迎與需求。2. 各組皆對資訊圖片的功能表示正面的肯定，認為資訊圖片能幫助他們掌握 、理解、記憶新聞中的重要訊息，並增加閱讀興趣。3.簡單的資訊圖片組 與加裝飾的資訊圖片組對電子報的整體印象及視覺吸引力的評價皆高於無 圖組。4.當圖片是隱藏在一個小圖像(icon)中時，使用者會期待該圖片有 較多裝飾性的插畫。四、認知型態除「瀏覽時間」一項外，在多項比較中 皆未造成統計上的顯著影響。五、圖片呈現方式與認知型態交互作用後， 未在任何一項比較中造成統計上的顯著影響。 With the popularity of the Internet and the World Wide Web (WWW), web-based newspapers become prevailing rapidly in the recent years. Web-based newspapers are different from traditional printed newspaper in designing and packaging message effectively. The purpose of this study is to investigate the effects of informational graphic, or infographic, presentation modes and reader*s cognitive styles on their recognition tests, browsing time and attitudes in a web-based newspaper environment. Infographic in thisstudy is defined as information concisely presented in a visual form.A 4 * 4 two factors experiment was executed in this study. The two independent variables are: infographics presentation modes and readers* cognitive styles. The dependent variables include subjects* information recall, browsing time and their attitude toward the web-based newspaper.Two hundred and sixty-eight subjects were participated in the experiment stage 1. They were tested on the Verbalizer-Visualizer Questionnaire (VVQ) developed by Kirby, Moore, & Schofield in 1988. Based on subjects* test scores, they are classified into four cognitive styles groups: verbal- oriented, visual-oriented, high verbal-visual and low verbal- visual. Among 268 subjects, 176 voluntary participants were randomly assigned to four versions of web-based newspapers in the experiment stage 2. The four versions of newspapers used in this study were designed by the researcher, and were different in terms of graphics presentation manner: no graphics (text- only), plain infographics, infographics with adornment and hidden infographics. Upon completion of reading all news, subjects received a recognition test and a five-point Likert scale questionnaire which assessed their attitude toward the reading experiences.The main results of this study are:1. Recognition performance: Readers* recognition scores were significantly better when message were redundant in both text and graphics those with only textual stories.2.Browsing time: Subjects who read infographics with adornment spent significantly more time in browsing the newspaper than the other three groups. The low verbal-visual group spent significantly more time in browsing newspaper than the verbal-oriented groups.3. User* Attitude:(1)All subjects expressed strong desire for graphics.(2)All subjects considered infographics to be useful tools helping them understand and remember the story, make them be more interested in reading news story.(3)Those who read plain graphic and adornment graphic thought the web- based newspaper more appealing than those who read text-alone stories.(4)When the graphics were hidden in a button, the reader expected decorative graphics when they were full-size presented.4.Although cognitive styles was thought to influence people*s information-processing performance, this study found that the cognitive style only had significant influence on browsing time. There was no significant interaction between graphic presentation modes and cognitive styles on any of the dependent variables.