Experience is known to be a key element involved in the modulation of face-processing abilities as manifested by the inversion effect, other-race, and other-age effects. Yet, it is unclear how exposure refines internal perceptual representations of faces to give rise to such behavioral effects. To address this issue, we investigated short- and long-term experienced stimulus history on face processing. Participants performed same-different judgments in a serial discrimination task where two consecutive faces were drawn from a distribution of morphed faces. The use of stimulus statistics was measured by testing the gravitation of representations toward the experienced mean (regression-to-the-mean), and the dynamic of the biases was tested by investigating trial-by-trial performance. Own-race and own-age faces were tested alongside other-race and other-age faces employing a within-subject design. Results demonstrated greater regression biases in other-race and other-age faces than in own-race and own-age faces. Perceptual narrowing, measured by the ability to form and use the representation of the overall mean of the nonnative faces, varied with proficiency levels, with only those with low proficiency in face recognition showing the use of overall stimulus history for other-race faces. In contrast, the use of stimulus history for other-age faces was similarly affected by statistics in the low- and high-proficiency groups. The results demonstrate that narrowing is associated with specialization levels occurring more robustly for other-race faces, for which exposure is limited during sensitive periods in development. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

中文翻译：

---

面部辨别中的经验依赖性偏差揭示了知觉专业化和狭窄之间的关联。


众所周知，经验是涉及面部处理能力调节的关键因素，如倒置效应、他者种族和其他年龄效应所表现。然而，目前尚不清楚曝光如何完善人脸的内部感知表征以产生这种行为效果。为了解决这个问题，我们调查了面部处理的短期和长期体验刺激历史。参与者在一系列判别任务中执行相同不同的判断，其中从变形面孔的分布中抽取两个连续的面孔。刺激统计量的使用是通过测试表征对经验平均值（回归到平均值）的引力来衡量的，偏差的动态是通过调查逐个试验的表现来测试的。本种族和同龄面孔与采用主体内设计的其他种族和其他年龄面孔一起进行了测试。结果显示，其他种族和其他年龄面孔的回归偏倚大于本种族和本年龄面孔。知觉狭窄，通过形成和使用非母语面孔总体平均值的能力来衡量，随熟练程度的变化而变化，只有那些人脸识别熟练度低的人才显示出对其他种族面孔的整体刺激历史的使用。相比之下，其他年龄面孔的刺激历史使用同样受到低熟练度组和高熟练度组统计数据的影响。结果表明，缩小与其他种族面孔的专业化水平更强烈相关，在开发的敏感时期，其暴露受到限制。（PsycInfo 数据库记录 （c） 2024 APA，保留所有权利）。