Expert Face Recognition: The Influence of Experience

同一个人的两张照片可能看起来非常不同，而两个不同人的照片可能看起来非常相似。因此，准确的人脸识别既需要区分(区分不同的人)，也需要身份匹配(当一个人的外貌发生变化时识别他/她)。我的研究目的是了解专家面部识别在整个生命周期中发展的潜在机制。要做到这一点，我采取了两种方法。首先，我研究了人脸识别和潜在的机制在一生中是如何变化的(例如，随着儿童遇到越来越多的新面孔和在老年期间)。其次，我比较了成年人从他们有丰富(例如，自己的种族，年轻人)和最少(其他种族，老年人)经验的类别中识别身份的能力。我建议使用三种方法来做到这一点。第一个重点放在一个很大程度上被忽视的挑战上：尽管外表自然变化(发型、化妆、视角的变化)，但仍能识别出一张脸。这是护照官员面临的挑战，也是当我们的邻居戴上新眼镜时，我们能够识别她的能力。当面孔不熟悉时，要做到这一点是困难的，特别是对儿童和观看其他种族面孔的成年人来说。我将研究一张脸变得足够熟悉以进行这种识别的过程，以及这个过程如何在整个寿命内发生变化，并在不同的面孔类别中有所不同。第二个关注的焦点是脸型和脸部质地这两个面部识别线索的依赖程度是否会随着熟悉度、感知者的年龄和脸部类别的不同而不同。成年人依靠脸型(特征形状和间距)来学习新的种族面孔，但面部纹理(肤色)来识别它们，这一模式在优秀的识别者中最为明显。我将利用形状和纹理线索的显着性来检验人脸识别中与年龄相关的差异是否可以归因于在学习过程中对面部形状的敏感度降低和/或当一张脸变得熟悉时未能依赖于纹理。我将进行一项平行研究，以检验是否同样的模式可能是导致对其他种族/年龄的面孔识别不佳的原因。第三个项目关注的是视觉工作记忆的局限性如何影响人脸识别。学习一张新的面孔需要感知者存储不同的实例和/或建立平均值；这两种机制都需要观察者对面孔进行编码，这一能力反映在视觉工作记忆中。我将测量工作记忆的容量和精确度，并在不同的年龄组(儿童、年轻人、老年人)和面孔类别(例如，年轻人和老年人；自己和其他种族的面孔)之间进行比较。总而言之，这些结果将为人们提供新的见解，了解面部感知如何在整个生命周期内根据环境输入进行精细调整。

Two pictures of the same person can look very different and pictures of two different people can look very similar. Thus, accurate face recognition requires both discrimination (telling people apart) and identity matching (recognizing a person when his/her appearance changes). The objective of my research is to understand the mechanisms underlying the development of expert face recognition across the lifespan. To do so, I take two approaches. First, I examine how face recognition and the underlying mechanisms change across the lifespan (e.g., as children encounter an increasing number of new faces and during old age). Second, I compare adults’ ability to recognize identity in faces from categories with which they have abundant (e.g., own-race, young adult) vs. minimal (other-race, older adult) experience. I propose doing so using three approaches. The first focuses on a largely ignored challenge: Recognizing a face despite natural variability in appearance (changes in hairstyle, makeup, viewpoint). This is the challenge faced by passport officers and an ability that allows us to recognize our neighbour when she dons new glasses. Doing so is difficult when faces are unfamiliar, especially for children and for adults viewing other-race faces. I will investigate the process by which a face becomes sufficiently familiar to allow such recognition and how that process changes across the lifespan and differs across face categories. The second focuses on whether reliance on two cues to facial identity—face shape and face texture—varies as a function of familiarity, perceiver age, and face category. Adults rely on face shape (feature shape and spacing) to learn new own-race faces but face texture (skin tone) to recognize them, a pattern most evident in good recognizers. I will manipulate the salience of shape versus texture cues to examine whether age-related differences in face recognition can be attributed to reduced sensitivity to face shape during learning and/or a failure to rely on texture as a face becomes familiar. I will conduct a parallel study to examine whether the same pattern might underlie poor recognition of other-race/age faces. The third project focuses on how limitations in visual working memory affect face recognition. Learning a new face requires the perceiver to store separate instances and/or to build an average; both mechanisms require the observer to encode a face, an ability that is reflected in visual working memory. I will measure the capacity and precision of working memory and compare these across age groups (children, young adults, older adults) and face categories (e.g., young vs. older adult faces; own- vs. other-race faces). Together, the results will provide novel insights about how face perception becomes finely tuned to environmental input across the lifespan.