面孔吸引力对微表情识别的影响
文献类型：学位论文
| 作者 | 林琼斯 |
| 答辩日期 | 2023-06 |
| 文献子类 | 继续教育硕士 |
| 授予单位 | 中国科学院大学 |
| 授予地点 | 中国科学院心理研究所 |
| 其他责任者 | 王甦菁 |
| 关键词 | 面孔吸引力 微表情识别 眼动 情绪识别 注意力 |
| 学位名称 | 理学硕士 |
| 学位专业 | 应用心理学 |
| 其他题名 | The Effect of Facial Attractiveness on Micro-Expression Recognition |
| 中文摘要 | Facial expressions are important non-verbal cues that humans use to convey information and express emotions. As a special facial expression, micro-expression (ME) is an extremely quick and uncontrollable facial movement that lasts for a short time and reveals thoughts and feelings that an individual attempts to cover up. ME recognition is widely used in the fields of national security, judicial interrogation, clinical and education fields as an effective clue for detecting deceptions, as MEs occurred too quickly and are very difficult to detect, scholars have long endeavored to explore and improve individuals' ability to recognize MEs. Facial attractiveness is a relatively stable characteristic of faces, which affects the individual's behavior and neural activity to a certain extent, and plays an important role in individual attitudes, judgments, and major decisions. Though much more difficult to detect and recognize, ME recognition is similar to macro-expression recognition in that it is influenced by facial features. Previous studies suggested that facial attractiveness could influence facial expression recognition processing. However, it remains unclear whether facial attractiveness could also influence ME recognition. The investigation of this issue could improve the theoretical research of the ME recognition mechanism, deepen the understanding of the role of facial attractiveness in the process of expression recognition, and provide a theoretical basis for researchers to design more systematic and efficient ME recognition training tools. In this research, two studies were conducted to explore the processing of ME recognition. Study 1 used ME recognition task and facial attractiveness rating task to investigate the accuracy and reaction time of three different MEs (positive, neutral, and negative) at two attractiveness levels (attractive, unattractive) in a static condition or dynamically. The results showed that there was a significant interaction between ME and facial attractiveness, MEs were recognized much faster on attractive faces than on unattractive ones. Furthermore, attractive happy faces were recognized faster in both the static and the dynamic conditions, highlighting the happiness superiority effect. Study 2 was divided into two experiments. Experiment 2 used eye-tracking technology to further investigate the effect of facial attractiveness on ME recognition and its eye movement pattern. Through the eye movement analysis of four eye movement indicators (Total Fixation Duration, First Fixation Duration, Percentage of Fixation Time, and Fixation Count) in three areas of interest (eye/brow, nose, and mouth), the results showed that facial attractiveness changed the eye movement pattern of individuals in the process of ME recognition. We observed that the eye/brows always attracted the highest proportion of fixations and fixation time, followed by the nose, and the least in the mouth area. Compared with unattractive faces, participants gazed at attractive faces for longer and more times in the eye/brow area. For happy expressions, participants gazed at attractive faces in the eye/brow area for longer and more times; for disgust expressions, participants gazed at unattractive faces in the nose area for longer and gazed at attractive faces in the eye/brow area more frequently. The study also discovered that during the ME recognition process, participants looked first at the eye/brow and nose areas, then at the mouth area. The first fixation time in the eye/brow area of unattractive faces was significantly longer than that of attractive faces, indicating that participants would prioritize the eye/brow area of attractive faces. In experiment 3, the pre-training model based on the DAN algorithm was used for testing. The results showed that the model had higher accuracy in the eye/brow area and had higher sensitivity to happy expressions, indicating that the attention resource allocation of human eye recognition and machine recognition was consistent to a certain extent. These findings indicate that when attractive faces are associated with positive MEs, facial attractiveness has a facilitating effect on ME recognition; the attention bias caused by facial attractiveness is affected by expression valence. For attractive faces, participants pay more attention to the eye/brow and nose areas of happiness, and there is no significant difference for unattractive faces. Therefore, our results provide evidence for the influence of facial attractiveness on ME recognition and explore its behavioral characteristics and attentional mechanisms. The results are consistent with the theories of the evaluative congruence account and the Happy-Face-Advantage, and provide more evidence to support the theories. Through the cross-study of computer vision technology and cognitive psychology, this study expands the theoretical model of influencing factors of ME recognition, provides new evidence for the research of ME recognition based on the attention mechanism, and provides new clues for the development of ME database and ME recognition training in the field of ME detection and recognition. |
| 英文摘要 | 面部表情是人类传递信息、表达情感的重要非言语线索，微表情作为面部表情的特殊形式，是指人们极力压抑或掩饰真实情绪时无法自控的面部动作。微表情是识别谎言重要且独特的线索之一，普遍应用于国家安全、司法、医疗、教育等领域。由于微表情发生速度极快且难以检测，长期以来，研究者致力于探索影响微表情识别的关键因素和提高个体的微表情识别能力。面孔吸引力是面孔相对定的特质，在一定程度上影响个体的行为表现和神经活动，对个体的态度、判断和重要决策等起关键作用。目前研究者对面孔吸引力在情绪识别过程中作用的研究主要集中在宏表情领域，面孔吸引力作为影响表情识别的重要面部特征，是否会对微表情识别产生影响?针对这一问题的探索，可以完善微表情识别机制的理论研究，深化面孔吸引力对表情识别影响的理解，为研究人员设计更加系统高效的微表情识别训练工具奠定理论基础。 研究一采用微表情识别任务和面孔吸引力评级任务，通过两个实验范式探讨被试在不同面孔吸引力水平下对微表情识别的认知加工。实验la和实验1b分别考察被试在BART和METT范式下，对高吸引力面孔和低吸引力面孔的三种不同微表情效价(积极、中吐和消极)判断的正确率和反应时。结果发现，高吸引力面孔条件下的高兴表情能更快更准确地被识别，快乐优势效应明显，微表情类型和面孔吸引力的交互作用显著，表明面孔吸引力影响了被试对微表情情绪类别的判断。研究二分为两个实验，实验二利用眼动追踪技术进一步考察面孔吸引力对微表情识别的作用及其眼动模式，通过对被试在眉眼区、鼻区和嘴区的总注视时间、首次注视时间、注视时间百分比以及注视次数进行分析，研究结果表明面孔吸引力改变了个体在微表情识别过程中的注视轨迹。被试在眉眼区的注视时间和注视次数最多，鼻区次之，嘴区最少。高吸引力面孔比低吸引力面孔分配到更多的注意资源，尤其是在眉眼区。从微表情类型上看，被试更关注高兴表情一高吸引力面孔的眉眼区，而对厌恶表情一低吸引力面孔鼻区的注视时间更长。在微表情识别过程中，被试先注视眉眼区和鼻区，最后是嘴区，说明眉眼区是微表情识别的重要线索区域。实验三利用基于DAN算法的预训练模型进行测试，结果发现模型在眉眼区的识别正确率更高，并且对高兴表情有较高的敏感度，说明人眼识别与机器识别的注意力资源分配在一定程度上具有一致性。上述结果表明，当高吸引力面孔与积极微表情相关联时，面孔吸引力对微表情识别具有促进作用;个体对具备高吸引力水平的面孔存在注意偏向，这种注意偏向受表情效价的影响，在高吸引力面孔条件下，被试更关注高兴表情面孔的眉眼区和鼻区，低吸引力面孔则不存在显著差异。 本研究证实了面孔吸引力对微表情识别的影响及其注意机制，研究结果与评价一致性理论和快乐优势效应理论相一致，并为该理论提供了更多的证据支持。本研究通过计算机视觉技术与认知心理学的交叉研究，拓展了微表情识别影响因素研究的理论模型，为基于注意力机制的微表情识别研究提供了新证据，为微表情检测与识别研究领域在微表情数据库开发及微表情识别训练等方面提供了新线索。 |
| 语种 | 中文 |
| 源URL | [http://ir.psych.ac.cn/handle/311026/45134]  |
| 专题 | 心理研究所_认知与发展心理学研究室 |
| 推荐引用方式 GB/T 7714 | 林琼斯. 面孔吸引力对微表情识别的影响[D]. 中国科学院心理研究所. 中国科学院大学. 2023. |
入库方式： OAI收割
来源：心理研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。