Facial recognition technology and human raters can predict political orientation from images of expressionless faces even when controlling for demographics and self-presentation.

IF 12.3 1区 心理学 Q1 PSYCHOLOGY, MULTIDISCIPLINARY
American Psychologist Pub Date : 2024-10-01 Epub Date: 2024-03-21 DOI:10.1037/amp0001295
Michal Kosinski, Poruz Khambatta, Yilun Wang
{"title":"Facial recognition technology and human raters can predict political orientation from images of expressionless faces even when controlling for demographics and self-presentation.","authors":"Michal Kosinski, Poruz Khambatta, Yilun Wang","doi":"10.1037/amp0001295","DOIUrl":null,"url":null,"abstract":"<p><p>Carefully standardized facial images of 591 participants were taken in the laboratory while controlling for self-presentation, facial expression, head orientation, and image properties. They were presented to human raters and a facial recognition algorithm: both humans (r = .21) and the algorithm (<i>r</i> = .22) could predict participants' scores on a political orientation scale (Cronbach's α = .94) decorrelated with age, gender, and ethnicity. These effects are on par with how well job interviews predict job success, or alcohol drives aggressiveness. The algorithm's predictive accuracy was even higher (<i>r</i> = .31) when it leveraged information on participants' age, gender, and ethnicity. Moreover, the associations between facial appearance and political orientation seem to generalize beyond our sample: The predictive model derived from standardized images (while controlling for age, gender, and ethnicity) could predict political orientation (<i>r</i> ≈ .13) from naturalistic images of 3,401 politicians from the United States, the United Kingdom, and Canada. The analysis of facial features associated with political orientation revealed that conservatives tended to have larger lower faces. The predictability of political orientation from standardized images has critical implications for privacy, the regulation of facial recognition technology, and understanding the origins and consequences of political orientation. (PsycInfo Database Record (c) 2024 APA, all rights reserved).</p>","PeriodicalId":48468,"journal":{"name":"American Psychologist","volume":" ","pages":"942-955"},"PeriodicalIF":12.3000,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"American Psychologist","FirstCategoryId":"102","ListUrlMain":"https://doi.org/10.1037/amp0001295","RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/3/21 0:00:00","PubModel":"Epub","JCR":"Q1","JCRName":"PSYCHOLOGY, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0

Abstract

Carefully standardized facial images of 591 participants were taken in the laboratory while controlling for self-presentation, facial expression, head orientation, and image properties. They were presented to human raters and a facial recognition algorithm: both humans (r = .21) and the algorithm (r = .22) could predict participants' scores on a political orientation scale (Cronbach's α = .94) decorrelated with age, gender, and ethnicity. These effects are on par with how well job interviews predict job success, or alcohol drives aggressiveness. The algorithm's predictive accuracy was even higher (r = .31) when it leveraged information on participants' age, gender, and ethnicity. Moreover, the associations between facial appearance and political orientation seem to generalize beyond our sample: The predictive model derived from standardized images (while controlling for age, gender, and ethnicity) could predict political orientation (r ≈ .13) from naturalistic images of 3,401 politicians from the United States, the United Kingdom, and Canada. The analysis of facial features associated with political orientation revealed that conservatives tended to have larger lower faces. The predictability of political orientation from standardized images has critical implications for privacy, the regulation of facial recognition technology, and understanding the origins and consequences of political orientation. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

即使在控制人口统计学和自我介绍的情况下,人脸识别技术和人类评分员也能从无表情的人脸图像中预测政治倾向。
在实验室中对 591 名参与者的面部图像进行了仔细的标准化处理,同时对自我介绍、面部表情、头部朝向和图像属性进行了控制。结果显示,人类(r = .21)和算法(r = .22)都能预测参与者在政治倾向量表上的得分(Cronbach's α = .94),但与年龄、性别和种族无关。这些效果与求职面试对工作成功的预测效果或酒精对攻击性的驱动效果相当。当算法利用参与者的年龄、性别和种族信息时,其预测准确度甚至更高(r = .31)。此外,面部外观与政治倾向之间的关联似乎超出了我们的样本范围:从标准化图像中得出的预测模型(同时控制年龄、性别和种族)可以从美国、英国和加拿大 3401 名政治家的自然图像中预测政治倾向(r ≈ .13)。对与政治倾向相关的面部特征的分析表明,保守派的脸型往往较大较低。从标准化图像中预测政治取向对隐私、面部识别技术的监管以及了解政治取向的起源和后果具有重要意义。(PsycInfo Database Record (c) 2024 APA, 版权所有)。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
American Psychologist
American Psychologist PSYCHOLOGY, MULTIDISCIPLINARY-
CiteScore
18.50
自引率
1.20%
发文量
145
期刊介绍: Established in 1946, American Psychologist® is the flagship peer-reviewed scholarly journal of the American Psychological Association. It publishes high-impact papers of broad interest, including empirical reports, meta-analyses, and scholarly reviews, covering psychological science, practice, education, and policy. Articles often address issues of national and international significance within the field of psychology and its relationship to society. Published in an accessible style, contributions in American Psychologist are designed to be understood by both psychologists and the general public.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信