Functional anatomy and neural mechanisms of face recognition in macaque monkeys using fMRI, electrophysiology, and microstimulation.

本研究的目的是探讨警觉猕猴的脸部侦测与辨识的神经机制。两个发现激发了当前的提议。首先，使用功能磁共振成像在警觉的猕猴，我们以前确定了3个大脑区域，显示增加血流量的脸相比，非面部对象。这些区域可能构成了面部处理的层次结构，因为它们沿着颞叶中的尾-喙轴排列。其次，我们的fMRI引导的单单位记录显示，在这些区域中最大的区域中，惊人的97%的神经元具有面部选择性。这两个发现使得前所未有地接触到几个同质的面部细胞群体成为可能，并打开了一个新实验的金矿，以了解大脑检测和识别面部的详细步骤。我们建议系统地描述视觉变换发生沿着这面补丁层次结构。我们的三个具体目标是：1）表征神经元在三个fMRI识别的面部斑块中的视觉调谐特性，2）确定在面部处理层次结构中如何以及在何处生成位置，大小，方向，照明和视角的不变性，以及3）将解剖输入和输出映射到每个面部斑块。

The project goal is to explore the neural mechanisms of face detection and recognition in alert macaques. Two findings motivate the current proposal. First, using fMRI in alert macaques, we previously identified 3 brain regions that show increased blood flow to faces compared to non-face objects. These regions likely constitute a hierarchy for face processing, since they are arranged along a caudal-rostral axis in the temporal lobe. Second, our fMRI-guided single-unit recordings revealed that in the largest of these regions, an astonishing 97% of the neurons are face selective. These two discoveries make possible unprecedented access to several homogeneous populations of face cells, and open a gold mine of new experiments to understand the detailed sequence of steps by which the brain detects and recognizes faces. We propose to systematically characterize the visual transformations occurring along this face patch hierarchy. Our three specific aims are to: 1) characterize the visual tuning properties of neurons in each of the three fMRI-identified face patches, 2) determine how and where invariance to position, size, orientation, illumination, and view angle are generated within the face-processing hierarchy, and 3) map the anatomical inputs and outputs to each of the face patches.