Suppr超能文献

研究人脸识别行为中个体差异的稳定性。

Investigating the stability of individual differences in face recognition behavior.

作者信息

Arrington Myles N, Scherf K Suzanne

机构信息

Department of Psychology, Pennsylvania State University, 113 Moore Building, University Park, PA, 16802, USA.

Social Science Research Institute, Pennsylvania State University, 113 Moore Building, University Park, PA, 16802, USA.

出版信息

Sci Rep. 2025 Mar 19;15(1):9425. doi: 10.1038/s41598-025-90317-4.

Abstract

Individual differences in face recognition abilities are characterized as heritable and resilient to change. However, this work is largely based on inter-individual differences, tends to include participants with extreme behavior (e.g., prosopagnosia, super-recognizers), and does not accommodate patterns of bias in intra-individual recognition behavior. Here, we investigated the continuity and stability of intra-individual differences in face recognition behavior among emerging adults using two tasks of unfamiliar face recognition that differ in the gender of the faces to be recognized. Although the estimate of stability is high (0.71) across the sample, there are instabilities in the behavior of many individual participants. For example, approximately 16.7% of the sample exhibited a discrepancy between tasks that was larger than 1 SD. Also, stability was more characteristic of extreme behavior. This is a bit surprising given the potential for close generalization of performance across these two tasks (identical structure and similar stimuli). Inter-individual differences in participant characteristics (i.e., gender, age, social skills) do not explain this variability. These findings are difficult to accommodate into current models of individual differences in face recognition behavior.

摘要

人脸识别能力的个体差异具有遗传性且不易改变。然而,这项工作主要基于个体间差异,倾向于纳入具有极端行为的参与者(例如,面孔失认症患者、超级识别者),并且没有考虑个体内部识别行为中的偏差模式。在此,我们使用两项不熟悉人脸识别任务,其中要识别的面孔性别不同,研究了新兴成年人在人脸识别行为中个体内部差异的连续性和稳定性。尽管整个样本的稳定性估计值很高(0.71),但许多个体参与者的行为存在不稳定性。例如,约16.7%的样本在两项任务之间表现出大于1个标准差的差异。此外,极端行为更具稳定性特征。鉴于这两项任务(结构相同且刺激相似)的表现有密切泛化的可能性,这有点令人惊讶。参与者特征(即性别、年龄、社交技能)的个体间差异并不能解释这种变异性。这些发现难以纳入当前人脸识别行为个体差异模型。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/f561/11923172/a28a75471be8/41598_2025_90317_Fig1_HTML.jpg

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验