Using steady-state visual evoked potentials (SSVEP) to measure brain responses to faces and objects

本研究利用稳态视觉诱发电位（SSVEP）研究人脸与物体感知的神经关联。研究表明，对人类视觉系统的重复刺激会在脑电图（EEG）中产生振荡，这种振荡模仿了重复刺激的频率——SSVEP （Regan 2009）。拟议的实验将采用一种适应范式，在这种范式中，具有相同身份（例如，“Joe”）的适应性面孔以周期性的呈现频率（4hz）在屏幕上反复闪现。根据先前的一项研究（Rossion & Boremanse, 2011），连续呈现相同的面部身份会产生神经信号的减少（即重复抑制效应）。在测试阶段，不同身份的面孔（如“弗雷德”、“鲍勃”、“比尔”等）以用于适应的刺激速率显示。不同面孔的呈现会引起脑电信号的突然增加，表明SSVEP对个体面孔身份的变化非常敏感。与传统的事件相关电位（ERPs）相比，SSVEP技术的优势在于该方法产生高信噪比，需要的实验试验大大减少，并且是高级视觉感知辨别的稳健测量方法。**然而，目前尚不清楚SSVEP是否对知觉经验和学习的影响有反应。在当前的提案中，通过测试现实世界和实验室训练的专家来解决这些问题。在第一个系列的实验中，将研究对鸟类有丰富经验的现实世界的鸟类专家和对其他种族面孔有丰富经验的单一种族人士。我们假设鸟类专家和同类面孔专家对专业领域的刺激（即鸟类专家对鸟类，同类面孔专家对同类面孔）比新手表现出更大的SSVEP。此外，SSVEP差异将与专业领域对物体和面孔的识别差异相关。其次，我们将在实验室培训研究中调查知觉专业知识的获得。我们预测，决定感知经验和调节SSVEP信号强度的是感知经验的种类，而不是感知经验的数量。在最后一组实验中，我们将检验SSVEP的大小是否区分了面部识别的个体差异以及神经正常人群和已知面部识别缺陷人群（即自闭症患者）之间的群体间差异。总之，该研究计划填补了我们对指导高级视觉过程的机制的理解的关键空白，将观察者的感知经验和能力与大脑活动的差异联系起来。

The proposed research investigates the neural correlates of human face and object perception using steady-state visual-evoked potentials (SSVEP). It has been shown that repetitive stimulation of the human visual system produces oscillations in the electroencephalogram (EEG) that mimics the frequency of the repeating stimulus - the SSVEP (Regan 2009). The proposed experiments will employ an adaptation paradigm in which an adapting face of the same identity (e.g., "Joe") is repeatedly flashed on the screen at a periodic rate of presentation (4 Hz). Based on a previous study (Rossion & Boremanse, 2011), the continuous presentation of the same face identity produces a reduction in the neural signal (i.e., repetition suppression effect). In the test phase, faces of different identities (e.g., "Fred", "Bob", "Bill", etc.) are shown at the rate of stimulation used for adaptation. The presentation of different faces cause an abrupt increase in the EEG signal indicating that the SSVEP is sensitive to the changing identities of individual faces. The advantages of the SSVEP technique over traditional event-related potentials (ERPs) are that the method yields a high signal-to-noise ratio, requires substantially fewer experimental trials and is a robust measure of perceptual discrimination in high-level vision.**However, it is not known if the SSVEP is responsive to the effects of perceptual experience and learning. These questions are addressed in the current proposal by testing real world and laboratory-trained experts. In the first series of experiments, real world bird experts who have extensive experience to birds and mono-racial people who have extensive experience to own-, but not other-race faces will be studied. We hypothesize that bird experts and own-race face experts will exhibit a greater SSVEP to stimuli in the domain of expertise (i.e., birds for bird experts, own-race faces for own-race experts) than novices. Moreover, SSVEP differences will be correlated with differences in recognition of objects and faces from the domain of expertise. Second, we will investigate the acquisition of perceptual expertise in a laboratory training studies. We predict that it is the kind, not the amount of perceptual experience that determines perceptual expertise and modulates the strength of the SSVEP signal. In the final set of experiments, we will examine whether the magnitude of the SSVEP distinguishes individual differences in face recognition and inter-group differences between neurotypical people and a population with known face recognition deficits (i.e., people on the autism spectrum). In summary, the proposed program of research fills a critical gap in our understanding of mechanisms that guide the processes of high-level vision, linking the observer's perceptual experience and abilities with differences in brain activity.