社会与工程心理学研究室知识库

    本研究在图像侦查的背景下，关注图像侦查中的人机交互工具的自然性及其评价，提出符合用户心理模型的交互工具是自然性高的交互工具，并且建立以行为绩效、心理负荷、情感体验为指标的多维的、客观的交互自然性评价体系。图像侦查技术是通过侦查视频图像锁定犯罪嫌疑人的一种侦查技术手段，通过技术判定，调用可能包含嫌疑人的视频监控，逐个查阅监控视频，从中筛选出特征明显的线索，以此确定嫌疑人的行动踪迹。然而，受限于监控影像设备、视频质量、图像识别技术的局限性，现有图侦技术仍然依赖于大量的人力投入。警员往往要长时间的连续观看海量的视频图像，导致心理负荷增加，很容易处于消极的情绪体验之中，且工作效率低下。更重要的是，整个查阅视频的过程中，警员只能通过纸笔记录案件要素和线索，视频内容与案件要素缺乏关联，不利于要素的梳理和警员之间相同特征的人、车、物等时空要素的串联分析。基于对一线图像侦查人员的访谈，本研究将绘制警员对于图像侦查工具的心理模型，并开发与图像侦查心理模型相符的视频交互技术，支持草图标注，线索超链接，自动轨迹绘制等功能，以简化警员的工作流程，提高工作的效率，并将案件相关信息进行电子化保存。本研究通过两个实验，综合脑电和行为绩效指标，从工作绩效、心理负荷、情绪体验等多个角度出发，对该工具的交互自然性进行评价。其中工作效率的指标是任务完成时间，心理负荷的指标是脑电信号中的额叶theta波和顶叶alpha波，情绪体验的指标是额叶alpha波不对称性。实验一任务是图像侦查任务，两组被试分别使用符合心理模型的、交互自然性高的工具和原始的工作方式完成任务，结果显示两组在情绪体验方面有差异。实验二在实验一的基础上，将图像侦查任务拆分成三个子任务，结果显示使用交互自然性高的工具的被试任务绩效更好、心理负荷更低、情绪体验更好。研究结果支持符合心理模型的工具交互自然性更高的结论，说明使用交互自然性高的工具可以提高效率、降低心理负荷、提高情绪体验。

    Under  the  background  of  image  detection,  this  research  focused  on  the naturalness and its evaluation of the human-computer interactive tools. We proposed that interactive tool which fits users' mental model is regarded as more natural and that naturalness can be evaluated by performance, mental workload and emotional experience. Video image detection is a kind of investigation technique to detect criminal suspects by detecting video images. By watching the surveillance videos that may  contain  suspects,  selecting  the  relevant  clues  from  videos,  video  image investigation analysts can identify suspects' trajectory. However, because of the limitations of surveillance equipment, video quality and image recognition technology, existing video image investigation work still relies on a large number of human labor. Analysts often watch massive surveillance videos for a long time, resulting high mental workload, negative emotional experience and low efficiency. More importantly, in the whole process of video search, the police can only record clues through paper and pen. Poor relevance of video material and text clue hinders video image investigation analysis. Based on the interview with video image investigation analysts, their mental model of the video image detection tool was drew. Based on the mental model, a video image investigation tool was realized with functions of sketch annotation, hyperlink and trajectory drawing. Two experiments were conducted to evaluate naturalness of the newly designed tool by comparing users' performance and EEG data. Performance data was time to completion in tasks, and EEG data was frontal theta and parietal alpha which represent mental workload condition, and frontal  alpha  asymmetry  which  represent  emotional  experience.  In  the  first experiment, one group of participants used newly designed tool, the other group of participants used original pencil and paper method. Result showed difference in emotional experience between two groups. In the second task, the video image instigation task was split into three tasks. Result also showed that participants used newly designed tool spend less time on tasks, and their mental workload was lower, their emotional experience was better. This result support the conclusion that the interactive tool which fits users' mental model is more natural. Using more natural interactive tool can improve working efficiency, reduce mental workload and improve emotional experience.