基于特征推理的图标搜索特性实验研究
发布时间：2019-03-16 16:05
【摘要】：采用视觉搜索实验范式,通过行为和眼动追踪数据揭示了基于形状相似、功能关联和语义匹配特征3类推理条件的图标搜索特性,通过方差分析法比较了不同特征推理条件对各测量指标的影响.实验结果表明:形状相似性引导下的目标搜索时间以及对于各搜索项目的平均注视时间均最短,所消耗的认知资源最少,无需深度信息加工即可进行目标识别;基于功能关联性的图标搜索过程中注意资源分配最广;而语义词引导下的目标搜索正确率最高.基于形状相似性和功能关联性特征推理条件下的视觉搜索行为可简化为搜索项目与启动图标之间的模式匹配以及功能层级隶属关系判断过程;以语义词为启动特征的视觉搜索首先需经过语义词特征的图形化过程,再对搜索项进行匹配判断.3类搜索情境中均存在自上而下的信息加工特征.研究结果为人机界面中图标设计提供了理论指导.
[Abstract]:The visual search experiment paradigm is used to reveal the icon search characteristics based on three kinds of reasoning conditions: shape similarity, functional association and semantic matching, through behavior and eye movement tracking data. The influence of different feature reasoning conditions on each measurement index was compared by variance analysis. The experimental results show that the target search time under the guidance of shape similarity and the average fixation time for each search item are the shortest and consume the least cognitive resources, so the target recognition can be carried out without the need for in-depth information processing. In the process of icon search based on functional relevance, the attention resource allocation is the broadest, while the target search accuracy rate is the highest under the guidance of words. The visual search behavior based on shape similarity and functional relevance features can be simplified as pattern matching between search items and startup icon and the judging process of functional hierarchy membership relationship. The visual search with semantic word as priming feature first needs to go through the graphical process of semantic feature and then make matching judgment to the search term. There are top-down information processing features in all three search situations. The results of this study provide theoretical guidance for the design of icons in man-machine interface.
【作者单位】： 东南大学机械工程学院;
【基金】：国家自然科学基金资助项目(71271053,71471037)
【分类号】：TP391.41
，
本文编号：2441701
[Abstract]:The visual search experiment paradigm is used to reveal the icon search characteristics based on three kinds of reasoning conditions: shape similarity, functional association and semantic matching, through behavior and eye movement tracking data. The influence of different feature reasoning conditions on each measurement index was compared by variance analysis. The experimental results show that the target search time under the guidance of shape similarity and the average fixation time for each search item are the shortest and consume the least cognitive resources, so the target recognition can be carried out without the need for in-depth information processing. In the process of icon search based on functional relevance, the attention resource allocation is the broadest, while the target search accuracy rate is the highest under the guidance of words. The visual search behavior based on shape similarity and functional relevance features can be simplified as pattern matching between search items and startup icon and the judging process of functional hierarchy membership relationship. The visual search with semantic word as priming feature first needs to go through the graphical process of semantic feature and then make matching judgment to the search term. There are top-down information processing features in all three search situations. The results of this study provide theoretical guidance for the design of icons in man-machine interface.
【作者单位】： 东南大学机械工程学院;
【基金】：国家自然科学基金资助项目(71271053,71471037)
【分类号】：TP391.41
，
本文编号：2441701
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2441701.html
