Determining the independence of face dimensions through an extended multidimensional signal detection theory

人们可以很容易地感知到一张脸的几个不同的属性，如身份、情感表达、年龄和运动。一个重要的问题是，大脑是否独立地处理这些属性，因此，例如，无论她是高兴、愤怒、动动眼睛、摇头，还是我们多年没有见过她，都很容易识别出她。虽然有影响力的人脸识别理论提出，某些人脸属性确实是独立于其他属性进行处理的，但回顾文献表明，它充满了相互矛盾的结果，可能是因为缺乏关键的实验控制。目前的项目试图确定面孔的身份、表情和运动是否被独立处理，使用行为和大脑独立性的测量方法。更好地了解健康人是如何感知面孔特性的，反过来可以更好地理解某些障碍(如抑郁症、焦虑症、精神分裂症和自闭症)面孔感知障碍背后的原因。为了解决之前研究中相互矛盾的发现，研究人员建议在一系列实验中使用一种新的方法，使用行为和大脑测量面孔感知的方法。首先，使用计算机图形学实现对呈现给参与者的人脸属性的严格控制：创建人脸和表情的三维模型，并从这些模型获得图像和视频。由于计算机图形学的进步，这些图像和视频看起来很自然，但它们的形状和运动可以高精度地控制。其次，运用知觉的数学模型对数据进行分析。这些模型严格定义了感知处理的特征，如独立性，以及测量这些特征的方法，同时将它们与参与者在行为任务中使用的决策策略分开。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

People can easily perceive several separate properties of a face, such as its identity, emotional expression, age, and movement. An important question is whether the brain processes each of those properties independently, so that, for example, it is easy to identify a friend regardless of whether she is happy, angry, moving her eyes, shaking her head, or even if we have not seen her in years. Although influential theories of face recognition propose that some face properties are indeed processed independently from others, a review of the literature shows that it is full of contradictory results, probably due to lack of critical experimental controls. The present project seeks to determine whether face identity, expression, and motion are processed independently, using both behavioral and brain measures of independence. A better understanding of how face properties are perceived in healthy individuals could in turn lead to a better understanding of the reason behind impairments in face perception in some disorders, such as depression, anxiety, schizophrenia, and autism.To resolve the contradictory findings from previous research, the investigator proposes to use a novel methodology in a series of experiments that use both behavioral and brain measures of face perception. First, tight control over the properties of faces presented to participants is achieved using computer graphics: three-dimensional models of faces and expressions are created, and images and videos are obtained from those models. Because of advances in computer graphics, such images and videos look quite naturalistic, but their shape and movements can be controlled with high precision. Second, mathematical models of perception are used to analyze the data. These models provide rigorous definitions of features of perceptual processing such as independence, and ways to measure those features while separating them from decision-making strategies used by participants in behavioral tasks.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.