High-level visual adaptation: exploring the neural representation of faces, words and objects

适应是心理学中用于研究大脑如何表征视觉元素和物体的一种技术。这涉及到让一个人研究一个视觉对象-适应刺激-一段规定的时间，然后显示一个测试刺激。适应刺激改变了测试刺激的感知，这种效应通常归因于被适应刺激激活的神经元的疲劳或习惯化。因此，在瀑布效应中，观看瀑布的水连续向下运动将使编码向下运动的神经元习惯化，然后当受试者看到静止图像时，疲劳的向下神经元和未受影响的向上神经元之间的不平衡将导致静止图像看起来向上漂移。这项技术最近被用来研究我们如何编码面孔。即使是对一张适应脸的简短观察5秒也会改变对随后的测试脸的感知。通过分离面部的特定属性并保持其他属性不变，这已被用于研究我们如何表征面部身份，表情，性别，年龄等，并揭示视觉系统编码的面部特定属性。我们的实验室在过去7年中在这一领域做出了重大贡献。在目前的提案中，我们的目标是利用我们在面部适应方面开发的专业知识，将类似的技术应用于其他物体表征的研究，特别是我们的大脑如何表征视觉词汇。* 与其他物体不同，人脸被认为是一种特殊类型的视觉刺激，几乎所有人都有特殊的专长。这是否使它们与其他物体不同是一个有争议的问题。然而，有了兴趣和接触，就有可能发展出对其他物体的专门知识，如汽车和鸟类。一个有趣的问题是，人们为这些其他物体开发的表征是否会获得使它们与面孔相似的属性。改编是解决这个问题的一种手段，关于视觉专业知识的性质，什么是意味着成为一个“视觉专家”。单词特别有趣，因为它们也是许多（识字的）人类的专业知识，但成像研究表明，虽然它们确实使用了大脑的相似部分，但单词更多地涉及左半球，而面孔主要涉及右半球。我们将使用适应来解决这样一个问题，即什么类型的视觉信息在词语的表征中与面孔相似--这表明视觉专业知识产生的共同属性--以及什么类型的视觉信息显示出重要的差异。这种差异对于我们理解“大脑半球的专门化”，即大脑左右半球的视觉编码是如何不同的，是非常重要的。

Adaptation is a technique used in psychology to study how the brain represents visual elements and objects. This involves having a person study a visual object - the adapting stimulus - for a defined period of time, following which a test stimulus is shown. The adapting stimulus alters the perception of the test stimulus, an effect that is often attributed to fatigue or habituation of the neurons that were activated by the adapting stimulus. Thus, in the waterfall effect, watching a waterfall with continuous downward movement of water will habituate neurons that code for downward motion, and when the subject is then shown a stationary image, the imbalance between fatigued downward neurons and unaffected upward neurons will cause the stationary image to appear to be drifting upwards.***This technique has recently been used to examine how we code faces. Even a brief view of an adapting face for 5 seconds will alter the perception of a subsequent test face. By isolating specific properties in the face and keeping others constant, this has been used to study how we represent facial identity, expression, gender, age and so on, and to reveal specific properties that the visual system encodes about faces. Our laboratory has made significant contributions in this field over the last 7 years. In the current proposal, we aim to use the expertise we have developed with face adaptation to apply similar techniques to the study of other object representations and in particular, how our brains represent visual words. ***Unlike other objects, faces are held to be a particular type of visual stimulus for which nearly all humans have a special expertise. Whether this makes them different from other objects is a matter of debate. However, it is possible with interest and exposure to develop expertise for other objects, such as cars and birds. One interesting question is whether the representations that people develop for these other objects then acquire properties that make them similar to faces. Adaptation is a means of addressing this question about the nature of visual expertise, what is means to be a "visual expert".***Words are of particular interest because they are also a class for which many (literate) humans have expertise, but imaging studies show that, while they do use similar parts of the brain, words involve the left hemisphere more while faces involve the right hemisphere mainly. We will use adaptation to address the question about what types of visual information in representations of words are similar to faces - indicating common properties generated by visual expertise - and what types show important differences. Differences would be important in contributing to our understanding of "hemispheric specialization", how the visual coding of the right and left hemispheres of the brain differ.**