Object, face, body and scene representations in the human brain

人脑中的物体、面部、身体和场景表征

• 批准号: 10703936
• 负责人: Christopher Baker
• 金额: $ 203.08万
• 依托单位: NATIONAL INSTITUTE OF MENTAL HEALTH
• 依托单位国家: 美国
• 资助国家: 美国
• 起止时间: 至
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10703936
• 关键词: American; Area; Behavioral; Blinking; Body part; Brain; Brain imaging; Brain region; Categories; Color; Complex; Computer Models; Data; Databases; Dimensions; English Language; Environment; Event; Evolution; Eye; Face; Face Processing; Functional Magnetic Resonance Imaging; Goals; Grain; Hand; Human; Image; Interruption; Knowledge; Learning; Leg; Light; Measures; Mental Health; Names; Nature; Neurosciences; Participant; Perception; Performance; Persons; Positioning Attribute; Psychology; Reporting; Research; Resources; Retina; Role; Saccades; Sampling; Semantics; Series; Shapes; Signal Transduction; Social Behavior; Stimulus; Stream; Time; Variant; Visual; Visual Cortex; Visual Perception; Visual system structure; Work; behavior measurement; cognitive process; computer science; crowdsourcing; experience; experimental study; falls; foot; insight; interest; large-scale database; mental representation; nervous system disorder; neural; neurotransmission; response; tool; visual information

The goal of this research is to understand how we see what we see: how does the brain analyze the light falling on the retina of the eye to reveal a world full of objects, people and things? Ongoing work focuses on two main areas of research: i) understanding the multidimensional representations of objects, and ii) determining the nature of object representation during occlusion (NCT00001360). i) Understanding the multidimensional representations of objects Real-world scenes are incredibly complex and heterogeneous, yet we are able to categorize them and identify objects and people within those scenes effortlessly. While prior studies have identified brain regions that appear to be specialized for processing faces, object and scenes, it remains unclear what the precise roles of these different regions are and what information they contain. One of the major challenges in understanding visual perception in the brain is the wide range of different objects and scenes that we experience. Despite this breadth, studies often focus on a small number of hand selected object or scene categories, but it becomes unclear how representative any subsequent findings are. To overcome this challenge we developed a large-scale database (THINGS) of 1,854 diverse types of object sampled systematically from concrete picturable and nameable nouns in the American English language. This database provides a rich resource of object concepts and object images and offers a tool for both systematic and large-scale naturalistic research in the fields of psychology, neuroscience, and computer science. We used this database in a large-scale behavioral experiment using online crowdsourcing, sampling 1.46 million trials in more than 5,000 participants. Using a computational model of the task, we were able to identify 49 core dimensions of our internal mental representations of objects, providing a comprehensive and fine-grained characterization of this representational space. In addition to the behavioral data we also collected extensive fMRI and MEG data on images from the THINGS database. The work described above provides a critical framework for now investigating the underlying neural representations of objects. We are establishing how these core dimensions are reflected in the neural signals over both space (fMRI) and time (MEG). Our analyses reveal highly distributed representations of the dimensions that include both high level and early visual cortex. Further, the temporal dynamics show wide variation in the evolution of responses that may reflect the differential contribution of visual and semantic information. These results provide important insights into the cognitive processes supporting our understanding of objects. ii) Determining the nature of object representation during occlusion Visual information is interrupted frequently due to occlusion, eyeblinks, and saccades. However, objects in the visual environment seem to persist through these perceptual gaps. We have been examining the nature of object representations during perceptual gaps caused by occlusion. First we used MEG to track how object representations unfold over time: before, during and after occlusion. Focusing on color, shape and position as different object features, when the moving objects were fully visible, there was some evidence for information about all features in the MEG signal. During occlusion, while some position information persisted, we could find no evidence for representation of other features. These results challenge the notion of a perception-like representation of moving objects during occlusion and open up new questions about how the visual system overcomes perceptual gaps to support the perception of a meaningful, continuous stream of information. Second, we conducted a series of behavioral experiments to further probe the nature of representations during occlusion. Participants were asked to report on the position of an object undergoing occlusion or on the time when the object would reappear from behind the occluder. While participants were accurate in estimating position during the early part of occlusion, their performance quickly deteriorated, suggesting a limited representation of occluded objects consistent with the MEG data. However, participants were very accurate in determining whether an object re-emerged from behind the occluder at the correct time or not, suggesting that integration between the pre- and post-occlusion periods might be critical for understanding occlusion events and might underlie our sense of object persistence. Elucidating how the brain enables us to recognize objects, scenes, faces and bodies provides important insights into the nature of our internal representations of the world around us. Understanding these representations is vital in trying to determine the underlying deficits in many mental health and neurological disorders.

这项研究的目的是了解我们如何看到我们所看到的：大脑如何分析落在眼睛视网膜上的光线，以揭示一个充满物体，人和事物的世界？ 正在进行的工作集中在两个主要研究领域：i）理解对象的多维表示，以及ii）确定遮挡期间对象表示的性质（NCT 00001360）。 （一）理解物体的多维表征 现实世界的场景是非常复杂和异构的，但我们能够对它们进行分类，并毫不费力地识别这些场景中的对象和人。虽然先前的研究已经确定了似乎专门用于处理面部，物体和场景的大脑区域，但仍不清楚这些不同区域的确切作用以及它们包含的信息。 理解大脑中视觉感知的主要挑战之一是我们所经历的各种不同的物体和场景。尽管如此广泛，研究往往集中在少数手选择的对象或场景类别，但它变得不清楚如何代表任何后续的研究结果。为了克服这一挑战，我们开发了一个大型数据库（THINGS），其中包含1,854种不同类型的对象，这些对象从美国英语中具体的可图示和可命名的名词中系统地采样。该数据库提供了丰富的对象概念和对象图像资源，并为心理学，神经科学和计算机科学领域的系统和大规模自然主义研究提供了工具。 我们在一个大规模的行为实验中使用了这个数据库，这个实验使用了在线众包，在5,000多名参与者中抽样了146万次试验。使用该任务的计算模型，我们能够识别出我们对物体的内部心理表征的49个核心维度，为这个表征空间提供了全面且细粒度的表征。 除了行为数据，我们还从THINGS数据库中收集了大量的fMRI和MEG图像数据。上述工作为现在研究对象的潜在神经表征提供了一个关键框架。我们正在建立这些核心维度如何反映在空间（fMRI）和时间（MEG）的神经信号中。我们的分析揭示了高度分布的代表性的尺寸，包括高层次和早期的视觉皮层。此外，时间动态显示广泛的变化，在响应的演变，可能反映了视觉和语义信息的差异贡献。 这些结果提供了重要的见解支持我们的理解对象的认知过程。 ii）确定遮挡期间对象表示的性质 视觉信息经常由于遮挡、眨眼和扫视而中断。然而，视觉环境中的物体似乎通过这些感知间隙而持续存在。我们一直在研究物体表征的性质，在知觉差距所造成的闭塞。首先，我们使用MEG来跟踪对象表示如何随着时间的推移展开：遮挡之前、期间和之后。将颜色、形状和位置作为不同的物体特征，当运动物体完全可见时，有一些证据表明MEG信号中存在关于所有特征的信息。在遮挡期间，虽然一些位置信息持续存在，但我们找不到其他特征的表征证据。这些结果挑战了在遮挡期间移动物体的感知表示的概念，并提出了关于视觉系统如何克服感知间隙以支持有意义的连续信息流的感知的新问题。 其次，我们进行了一系列的行为实验，以进一步探讨性质的表征在闭塞。参与者被要求报告一个对象的位置进行闭塞或对象将重新出现从后面的遮挡器的时间。虽然参与者在闭塞的早期部分准确估计位置，但他们的表现迅速恶化，这表明与MEG数据一致的闭塞对象的代表性有限。然而，参与者在确定物体是否在正确的时间从遮挡物后面重新出现时非常准确，这表明遮挡前后的整合对于理解遮挡事件可能是至关重要的，并且可能是我们对物体持久性的感觉的基础。 阐明大脑如何使我们能够识别物体，场景，面孔和身体，为我们对周围世界的内部表征的本质提供了重要的见解。理解这些表征对于试图确定许多精神健康和神经系统疾病的潜在缺陷至关重要。