面孔表情与面孔性别信息识别的时间特性比较

发布时间：2019-01-03 20:27

【摘要】：目的：面孔是我们日常生活中常见的视觉刺激，其识别具有特异性。表情和性别信息是面孔信息的重要组成部分，对两者的识别在我们的人际交往等活动方面有着巨大的影响。以往关于表情和性别的研究众多，然而关于二者直接比较的研究却很稀缺。本研究选定不同的呈现时间对面孔表情和面孔性别信息的识别进行研究，并且设定了区组和完全随机两种条件，希望通过更加真实、更加生态化的手段对这一问题加以探讨，以便能够使得研究结果具有更好的现实意义。方法：实验一和实验二均采用相同的范式、实验设计以及相同的呈现时间（12ms、50ms，250ms和500ms）对面孔的表情和面孔性别识别进行比较，不同的是实验一中每个trial中被试有意识地将注意力投入到面孔的表情信息或者性别信息中进行反应，而在实验二中，被试在并不知道接下来是要对呈现的面孔图片的表情还是性别信息进行反应的情况下实验。实验三是实验二的补充实验，将呈现时间（50ms、100ms、150ms、200ms和250ms）更加细化，以期对实验二的一些问题进行补充和修正，并试图寻找面孔表情信息和面孔性别信息识别的关键点。结果：总的来说，面孔表情信息识别优于面孔性别信息识别，主要反映在正确率指标上：（1）实验一在12ms呈现时间条件下，判断表情任务的正确率显著高于判断性别任务的正确率（t=3.647，，p0.01）；50ms呈现时间条件下也发现了相同的结果（t=2.103，p0.05）。（2）实验二在12ms呈现时间条件下，判断表情任务的正确率显著高于判断性别任务的正确率（t=4.742，p0.001）；50ms呈现时间条件下也发现了相同的结果（t=2.256，p0.05）。（3）实验三在50ms呈现时间条件下，判断表情任务的正确率显著高于判断性别任务的正确率（t=3.647，p0.05）。结论：（1）在12ms和50ms时，面孔表情信息识别比面孔性别信息识别更准确；（2）面孔表情信息加工优于面孔性别信息加工，尤其在高度自动化水平下，面孔表情信息识别比面孔性别信息识别先达到加工足以做出分类反应的程度。
[Abstract]:Objective: face is a common visual stimulation in our daily life, and its recognition is specific. Facial expression and gender information are important components of face information. There have been a lot of studies on facial expressions and gender in the past, but the direct comparison between them is scarce. This study selected different presentation time to study the recognition of facial expression and facial gender information, and set up two conditions of block and complete randomness, hoping to explore this problem through more realistic and more ecological means. In order to make the research results have better practical significance. Methods: both experiment 1 and experiment 2 used the same paradigm, the experimental design and the same presentation time (12 Ms / 50 Ms ~ (250) Ms and 500ms) were used to compare facial expression and face sex recognition. The difference was that in experiment one, each trial participant consciously focused on facial expression information or gender information, while in experiment two, The subjects did not know whether to respond to the face image or gender information. Experiment 3 is a supplementary experiment of experiment 2. The time of presentation (50ms / 100ms / 150ms / 200ms and 250ms) will be further refined in order to supplement and correct some problems in experiment two. And try to find the key points of facial expression information and face gender information recognition. Results: in general, facial expression information recognition was superior to face gender information recognition, mainly reflected in the correct rate index: (1) experiment 1. Under the condition of 12ms presentation time, The correct rate of judging expression task was significantly higher than that of gender task (t = 3.647 / p0.01). The same results were also found in 50ms presentation time (t0. 103 / p0. 05). (2). In the case of 12ms presentation, the correct rate of judging facial expression task was significantly higher than that of gender task (t0. 742, p0. 001). The same results were also found under the 50ms presentation time (t0. 256 / p0. 05). (3). In the case of 50ms presentation, the correct rate of judging facial expression task was significantly higher than that of gender task (t0. 647 / p0. 05). Conclusion: (1) in 12ms and 50ms, facial expression information recognition is more accurate than face gender information recognition; (2) facial expression information processing is superior to face gender information processing, especially in the highly automated level, facial expression information recognition can be processed to a degree sufficient for classification response than face gender information recognition.
【学位授予单位】：辽宁师范大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：B842

【参考文献】