Perceptual bases of visual concepts

描述(申请人提供)：我们的总体目标是发现大脑如何处理视觉信息，神经活动如何与视觉感知相关，以及视觉处理如何与其他大脑认知系统相互作用。我们的具体目标是阐明鸽子是如何识别和概念化视觉刺激的。通过研究鸽子-一种高度可视化的动物，它很容易学习，但没有语言或哺乳动物的新皮质，其历史可以仔细地控制和系统地变化-视觉识别和概念化的过程可能会更快和更容易地被发现。如果鸽子的视觉辨别行为类似于人类，那么概念化的过程可能是由常见的神经生物学机制调节的，这些机制不依赖于语言能力或人脑。鸽子可能成为复杂视觉加工的行为学和生物学研究的强大模型系统。我们的建议旨在观察人类和鸽子的认知和概念化的感知过程是否相似。鸽子将接受几种不同的操作条件反射程序的训练，以区分线条画和计算机渲染的自然和人工刺激。随后，鸽子将接受以下刺激物的测试：(1)降低训练刺激物的质量，(2)重新排列部分，(3)深度旋转图像。这些测试刺激在人类身上产生高度特异性的效应，这鼓励了这样一种想法，即物体识别是通过结构描述来中介的，该结构描述指定了物体各部分的神经表示以及这些部分之间的关系。如果人和鸽子类似地处理这些不同的视觉刺激，那么我们用鸽子做的实验结果应该与人相似。经验上的趋同将证明自然界的经济性，以及语言对于视觉识别和概念化的多余。经验分歧将意味着不同的神经生物学或语言机制在人和鸽子的视觉识别和概念化中起中介作用。无论是哪种情况，这一研究项目的结果都应该对视觉识别和概念化的基本机制有相当大的帮助。除了我们提议的研究的科学意义之外，它与健康的相关性也很大。发展物体识别和概念化的健全动物模型可能使我们更好地理解人类视觉系统在健康和疾病中的行为和生物学机制。有效的动物模型也可能有助于为设计评估视觉表现的非语言诊断测试铺平道路。此外，比较鸽子和人类识别物体的方式，可以为计算机和认知科学家构建能够识别现实世界中复杂刺激的人造设备提供新的见解。发现不同的生物系统如何实现同样的适应性壮举，可能会对那些试图创造不同的人工和假肢物体识别系统的人有很大帮助。公共卫生相关性：物体识别和视觉概念化动物模型的发展可能使我们能够更好地理解人类视觉系统在健康和疾病中的行为和生物学机制。有效的动物模型也可能有助于为评估视觉表现设计实用的诊断测试；尤其重要的是，言语行为不需要参与这样的表现评估，这使得非语言测试对婴儿、幼儿和临床人群特别有用。我们的研究结果也应该有助于我们理解视觉系统是如何发展的，以及在疾病或受伤后如何促进其再生。

DESCRIPTION (provided by applicant): Our general aim is to discover how the brain processes visual information, how neural activity is related to visual perception, and how visual processing interacts with other brain systems underlying cognition. Our specific aim is to elucidate how pigeons recognize and conceptualize visual stimuli. By studying the pigeon-a highly visual animal which can readily learn, but which does not have language or a mammalian neocortex and whose history can be carefully controlled and systematically varied-the processes of visual recognition and conceptualization may be more quickly and readily discovered. If the visual discrimination behavior of pigeons resembles that of humans, then the processes of conceptualization may be mediated by common neurobiological mechanisms which do not depend on linguistic competence or the human brain. The pigeon may become a powerful model system for both behavioral and biological studies of complex visual processing. Our proposal aims to see whether the perceptual processes of recognition and conceptualization are similar in humans and pigeons. Pigeons will be trained with several different operant conditioning procedures to discriminate line drawings and computer renderings of natural and artificial stimuli. The pigeons will later be tested with stimuli that: (1) degrade the training stimuli, (2) rearrange its parts, and (3) rotate the image in depth. These test stimuli produce highly specific effects in humans, which encourage the idea that object recognition is mediated by a structural description specifying a neural representation of the object's parts and the relations among those parts. If people and pigeons similarly process these various visual stimuli, then the results of our experiments with pigeons should parallel those with people. Empirical convergence would attest to the economy of nature and to the superfluity of language for visual recognition and conceptualization. Empirical divergence would imply that different neurobiological or linguistic mechanisms mediate visual recognition and conceptualization in people and pigeons. In either case, the results of this research project should shed considerable light on the basic mechanisms of visual recognition and conceptualization. Beyond the scientific significance of our proposed research, its health relevancy is considerable. Developing sound animal models of object recognition and conceptualization might better enable us to understand the behavioral and biological mechanisms of the human visual system in both health and disease. Effective animal models may also help pave the way for devising nonverbal diagnostic tests for assessing visual performance. In addition, comparing how pigeons and people recognize objects could provide new insights for computer and cognitive scientists to construct artificial devices which can recognize complex stimuli in the real world. Discovering how different biological systems accomplish the same adaptive feat might greatly help those attempting to create different artificial and prosthetic systems of object recognition. PUBLIC HEALTH RELEVANCE: The development of animal models of object recognition and visual conceptualization might better enable us to understand the behavioral and biological mechanisms of the human visual system in both health and disease. Effective animal models may also help pave the way for devising practical diagnostic tests for assessing visual performance; particularly important here is the fact that verbal behavior need not participate in such performance assessments, making nonverbal tests especially useful with infants, young children, and clinical populations. Our results should also advance our understanding of how the visual system develops and how to promote its regeneration after disease or injury.