The ventral visual pathway and the construction of identity-specific representation

本提案的主要目标是更好地了解大脑如何编码人脸识别。我们将使用心理物理学的方法与经颅磁刺激（TMS）和功能性磁共振成像（fMRI）来测试人类大脑的枕颞和前颞皮质区如何使我们能够获得并保持对已知的，熟悉的人的稳定表征，从而支持人脸识别。该提案的中心假设是，不是有一个单一的人脸识别区域，而是一个分层的，但分布式和循环的网络负责人脸识别。为了验证这一假设，我们研究了枕面部区域（OFA）和前颞叶（ATL）在识别过程中的作用，因为这两个区域位于所提出的层次结构的两端。我们将主要问题分为两部分。首先，使用TMS，我们将测试OFA/ATL在从陌生到熟悉的面孔过渡过程中的作用。我们将比较OFA和ATL刺激的效果，因为参与者以两种根本不同的方式获得身份特异性信息。参与者将暴露于大量的自然可变的面孔，以促进基于统计的平均面孔的创建，或者他们将看到面孔以及沿着额外的与人相关的语义信息（姓名、职位）。我们将评估OFA和ATL在这些任务中的因果作用。我们希望，在这两项任务中，OFA/ATL的刺激如何干扰学习的方式将为我们提供有关面孔加工网络结构的信息。其次，我们将研究腹侧视觉系统中熟悉面孔的获得表征。在心理物理学实验中，我们将测试它随时间的稳定性。在fMRI引导的状态依赖TMS实验中，我们将测试其在OFA和ATL中的图像独立性。最后，使用功能磁共振成像和一个新版本的多变量模式分析（表征相似性分析），我们将测试的差异和相似性的细粒度表示不熟悉与熟悉的面孔。总的来说，我们的目标是揭示负责面部识别的神经机制。

The major objective of the present proposal is to better understand how the brain encodes faces for recognition. We will use psychophysical methods together with transcranial magnetic stimulation (TMS) and functional magnetic resonance imaging (fMRI) to test how the occipito-temporal and anterior-temporal cortical areas of the human brain enable us to acquire and maintain a stable representation about known, familiar persons, thereby supporting face recognition. The central hypothesis of the proposal is that instead of having a single face recognition area a hierarchical, but distributed and recurrent network is responsible for face recognition. For testing this hypothesis we study the role of the occipital face area (OFA) and the anterior temporal lobe (ATL) in recognition processes, as these two areas are located towards opposite ends of the proposed hierarchy. We split our major question into two parts. First, using TMS, we will test the role of OFA/ATL in the process as faces transition from unfamiliar to familiar. We will compare the effect of OFA and ATL stimulation as participants acquire identity-specific information in two fundamentally different ways. Participants will either be exposed to a large amount of naturally variable faces to facilitate the creation of a statistically based average faces or they will see faces along additional person-related semantic information (names, job-titles). We will evaluate the causal roles of OFA and ATL in these tasks. We expect that the manner how the stimulation of OFA/ATL interferes with learning during the two tasks will give us information regarding the structure of the face-processing network. Second, we will study the acquired familiar face representation within the ventral visual system. In psychophysical experiments we will test its stability over time. In fMRI-guided, state-dependent TMS experiments we will test its image independence within the OFA and ATL. Finally, using fMRI and a novel version of multivariate pattern analysis (representational similarity analysis) we will test the differences and similarities of the fine-grained representation of unfamiliar versus familiar faces. Overall, our aim is to reveal the neural mechanisms that are responsible for face recognition.