Abstract: In social cognition, people use face-extracted information for social categorization and also judgments of personality, and this act can lead to possible consequences. The primary focus of previous models on social perception of faces is on social and physical facial features, which are important for researchers to understand these processes. Two- and three-dimensional models of social perceptions of faces currently exist, of which the former includes trustworthiness and dominance, while the latter includes approachability, youthful-attractiveness, and dominance as dimensions. This paper discusses social perceptions of faces through analyzing the processes in previously developed models of social perception of faces. This will include two- and three-dimension models using the relationship between the dimensions of attractiveness and trustworthiness. Future directions for research are also included. Future research should aim to develop a common computational account of various aspects of social perceptions of faces and characterize the functions of the brain regions involved in this perception.

Key words: social perceptions of faces, model, trustworthiness, dominance, attractiveness

摘要： 面孔作为一种复杂的特殊刺激，是人类表达自身情绪、认知他人情感的重要工具和途径。面孔识别是人类社会生活的一项重要技能，在缺乏外部线索时，它有助于我们对他人面孔的熟悉度、年龄和种族等方面有一个较明确的认识，从而更有利于人们社会交往和适应环境。大量研究表明，基于面孔的物理特征可以推断面孔所有者的内在特质，而这一推断又会影响个体的行为决策。因此，对面孔的认知不仅要关注面孔结构化特征，也要关注面孔社会化特征。针对面孔结构化特征，学者们提出了一些经典模型，如Bruce-Young模型、NBC模型等；针对面孔社会化特征，Alexander Todorov及其合作者采用主成分分析法，构建了包含信任度和支配度的面孔社会知觉“二维”模型。Clare A.M. Sutherland及其合作者提出，吸引力和信任度是两个独立的维度，由此构建了包含可接近性、年轻的—吸引力和支配度的面孔社会知觉“三维”模型。本文认为，面孔社会知觉模型从“二维”到“三维”的发展可以从吸引力和信任度两个维度的关系进行探讨。面孔在社会互动中传递着大量的信息，其中较为重要的是面孔信任度和面孔吸引力。其中，信任度评价具有决定性意义，信任一个不值得信任的人可能会给自身造成危害，而没信任一个值得信任的人往往意味着失去了合作的机会；吸引力评价则具有进化上的意义，吸引力往往与潜在配偶的繁殖能力有关，尤其体现在长期择偶关系之中。总结以往的面孔识别模型与本文介绍的面孔社会知觉模型，未来基于面孔特征的识别或特质推断可以从以下几方面进行思考：第一，选取生态化效度更高的面孔刺激材料；第二，进行跨文化检验模型的普适性；第三，从面孔识别模型向面孔社会知觉模型的发展需要进一步探索内在的认知神经机制；第四，进一步拓宽面孔社会知觉的研究领域。

关键词: 面孔社会知觉, 模型, 信任度, 支配度, 吸引力