Suppr超能文献

基于性别的人脸识别原型形成。

Gender-based prototype formation in face recognition.

机构信息

Institut Universitaire de France, Centre Européen des Sciences du Gouˆt et de l’Aliment, University of Burgundy, Dijon, France.

出版信息

J Exp Psychol Learn Mem Cogn. 2011 Jul;37(4):888-98. doi: 10.1037/a0022963.

Abstract

The role of gender categories in prototype formation during face recognition was investigated in 2 experiments. The participants were asked to learn individual faces and then to recognize them. During recognition, individual faces were mixed with faces, which were blended faces of same or different genders. The results of the 2 experiments showed that blended faces made with learned individual faces were recognized, even though they had never been seen before. In Experiment 1, this effect was stronger when faces belonged to the same gender category (same-sex blended faces), but it also emerged across gender categories (cross-sex blended faces). Experiment 2 further showed that this prototype effect was not affected by the presentation order for same-sex blended faces: The effect was equally strong when the faces were presented one after the other during learning or alternated with faces of the opposite gender. By contrast, the prototype effect across gender categories was highly sensitive to the temporal proximity of the faces blended into the blended faces and almost disappeared when other faces were intermixed. These results indicate that distinct neural populations code for female and male faces. However, the formation of a facial representation can also be mediated by both neural populations. The implications for face-space properties and face-encoding processes are discussed.

摘要

在两项实验中,研究了性别类别在面孔识别过程中的原型形成中的作用。参与者被要求学习个人面孔,然后识别它们。在识别过程中,将个人面孔与混合面孔混合,这些混合面孔是相同或不同性别的混合面孔。这两项实验的结果表明,即使从未见过,也可以识别由所学个人面孔混合而成的混合面孔。在实验 1 中,当面孔属于同一性别类别(同性混合面孔)时,这种效果更强,但也出现在跨性别类别中(跨性别混合面孔)。实验 2进一步表明,这种原型效应不受同性混合面孔呈现顺序的影响:当面孔在学习过程中一个接一个呈现或与异性面孔交替呈现时,效果同样强烈。相比之下,跨性别类别中的原型效应对面孔混合到混合面孔中的时间接近非常敏感,当其他面孔混合时,几乎消失。这些结果表明,不同的神经群体编码女性和男性面孔。然而,面部表示的形成也可以由两个神经群体介导。讨论了对面部空间特性和面部编码过程的影响。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验