http://dx.doi.org/10.6129/CJP.2005.4704.03
Chinese Journal of Psychology, 47(4), 2005， 339-351
唐大崙（中國文化大學大眾傳播系）；李天任（中國文化大學資訊傳播研究所）；蔡政旻（中國文化大學資訊傳播研究所）
摘要
過去一世紀，心理學家對於人類喜好程度 的測量多建立在主觀的問卷調查、或者是配對 比較程序上。偶而也有研究者假設喜好程度會 表現在視線軌跡中，但是從來沒有人認真對這 個假設進行實徵性探索。本研究首次以客觀的 視線軌跡分析，配合受測者主觀填答的方式， 分析受測者對於喜好之色彩與視線軌跡所透露 出的訊息是否有關。本研究測量的依變項包含 凝視點個數、總凝視時間、返回凝視次數、凝 視順序與喜好程度。於穩定控制亮度之實驗環 境中，採用八種主要色樣本，分別塗到色票、 杯子、T 恤、椅子、機車、磁片及背包等七種 物品上，同一種物品的八種顏色同時呈現，並 做到各顏色所在空間位置的對抗平衡。受測者 依序觀看這些刺激材料的同時，以眼球追蹤儀 器記錄眼球運動訊息，事後再要求受測者依喜 好程度排序。
經由多變量變異數分析結果發現，越受喜 愛的物品色彩，其被凝視之總時間、凝視次數 與視線返回次數越高，而且大約在第二次到第 七次凝視的時候，即已在視線軌跡中表現出喜 好的影響。商品類別與色彩類別在凝視時間、 凝視次數與視線返回次數這三個依變項上，也 有顯著主效果。這結果不只顯示視線軌跡可以 反映內在喜好程度，建立了另一個客觀測量喜 好程度的典範，也透露出物品的顏色與細節都 是吸引視線的重要因素，而且文末也討論這類 實驗對喜好研究的意義。
關鍵詞：色彩喜好、眼球追蹤、凝視時間、凝 視次數
An Exploratory Study on Relationship Between Preference and Scanpath ––– Evidence from Color Preference Sorting Task
Da-Lun Tang（Department of Mass Communication,Chinese Culture University）；Tien-Rein Lee（Department of Information Communication, Chinese Culture University）；Chung-Min Tsai（Department of Information Communication, Chinese Culture University）
Abstract
Researchers have used subjective rating methods such as survey and pair-comparison procedures to study preference of human beings in the past century. Less empirical studies were done using the eye-tracking method. This paper provides first empirical data using eye-tracking experimental method to explore the relation between subjects' color preference and their scanpath. Visual stimuli used in this study were seven different kinds of objects (color chip, cup, T-shirt, chair, motorcycle, floppy disk, and backpack) with their 8 primary colors (red, orange, yellow, yellow-green, green, blue-green, blue, and purple). Identical objects with eight different colors were presented on the computer screen each time. Subjects were given five seconds to browse each object with eight different colors per trial. An eye tracking machine was used to track their eye movements. After subjects finish browsing all objects, they were asked to rank the color preference for each object. Results from MANOVA showed that the fixation time, fixation counts and return counts were significantly longer and higher for more preferred colors. In addition, subjects had longer fixation time and higher fixation counts for color chips than other kinds of objects. In general, we concluded that eyetracking paradigm provides another method to measure color preference objectively. Moreover, colors and other details of objects may be another important factors to attract subjects’ eye movements.
Keywords: color preference, eye tracking, fixation time, fixation counts
