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

这项研究的目的是了解我们如何看到我们所看到的：大脑如何处理落在眼睛视网膜上的光，以产生我们对充满物体，人，事物和地方的世界的感知？ 在过去的一年中，我们继续研究大脑中自下而上（感觉驱动）和自上而下（内部驱动）加工之间的相互作用，重点是工作记忆和行为目标（任务）的影响，使用复杂的视觉刺激，如视觉对象和场景。 1)心理意象 在没有感觉输入的情况下，心理意象完全依赖于自上而下的信号。在此之前，我们进行了详细的比较视觉图像和知觉的个别复杂的物体使用功能磁共振成像。我们发现，（1）我们可以解码参与者在多个大脑区域看到或想象的特定物体的身份，（2）图像和感知信息在整个视觉处理流中的分布不同。这项先前的工作集中在参与者刚刚看到的物体的图像上，只需要短期记忆。我们目前正在扩展这项工作，以研究长期记忆中物体的图像，训练受试者在一天内学习特定的物体，然后在稍后的一天想象它们，同时我们测量大脑的反应。 2)工作记忆 工作记忆是指在短时间内将信息主动地保存在脑海中的过程。与心理意象一样，工作记忆是在没有感觉输入的情况下产生的。我们研究了哪些大脑区域参与了工作记忆过程中的信息维护，以及大脑区域的参与如何根据观察者的行为目标而变化（Lee，Kravitz，Baker，2013，Nature Neuroscience）。参与者在相同的视觉输入（对象的图片）上执行两项任务，其中一项任务要求参与者保持视觉属性和其他非视觉属性（对象的名称）。我们发现，在维持视觉特性，对象身份可以解码从大脑活动的视觉，但不是前额叶皮层，而相反的非视觉特性。因此，在工作记忆中保持信息的能力是大脑的一种普遍而灵活的特性，各个区域的作用取决于目标。 3)行为目标/任务 不同的任务需要从视觉刺激中提取不同类型的信息，这取决于观察者的行为目标。我们一直在研究复杂视觉刺激的表征如何根据参与者正在执行的任务而变化。 首先，我们发现我们可以从大脑多个区域的活动中解码参与者对给定视觉对象执行的任务。此外，我们发现强调视觉刺激的物理特性（例如颜色）的任务和强调概念特性（例如现实世界的大小）的任务之间有很大的区别。 其次，我们现在将这项工作扩展到视觉场景。在此之前，我们发现大脑中被认为对场景识别至关重要的区域中的场景表征主要反映场景的空间属性（例如，它们是开放的还是封闭的），而不是语义属性（即场景类别）。在我们目前的工作中，我们正在调查这些表征如何根据任务而变化，要求参与者专注于所呈现场景的特定方面。 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。

The goal of this research is to understand how we see what we see: how does the brain process the light falling on the retina of the eye to produce our perception of a world full of objects, people, things and places? During the past year we have continued to investigate the interaction between bottom up (sensory driven) and top-down (internally driven) processing in the brain, focusing on working memory and effect of behavioral goals (task), using complex visual stimuli, such as visual objects and scenes. 1) Mental Imagery Mental imagery, which occurs in in the absence of sensory input, relies entirely on top-down signals. Previously, we conducted a detailed comparison of visual imagery and perception for individual complex objects using fMRI. We found that (1) we can decode the identity of the specific object participants view or imagine in multiple brain regions, and (2) imagery and perceptual information are distributed differently throughout the visual processing stream. This prior work focused on imagery of objects that participants had just seen, requiring short-term memory only. We are currently extending this work to investigate imagery of objects from long-term memory, training subjects to learn particular objects on one day and then imagine them on a later day while we measure brain responses. 2) Working memory Working memory refers to the process of actively holding information in mind over short time intervals. As with mental imagery, working memory occurs in the absence of sensory input. We investigated which brain regions are involved in maintaining information during working memory and how the involvement of brain regions varies according to the behavioral goals of the observer (Lee, Kravitz, Baker, 2013, Nature Neuroscience). Participants performed two tasks on identical visual input (pictures of objects) with one task requiring participants to maintain visual properties and the other non-visual properties (name of the object). We found that during the maintenance of visual properties, object identity could be decoded from brain activity in visual, but not prefrontal, cortex, whereas the opposite held for non-visual properties. Thus, the ability to maintain information during working memory is a general and flexible property of the brain, with the role of individual regions being goal-dependent. 3) Behavioral goals/task Different tasks require different types of information to be extracted from visual stimuli, depending on the behavioral goals of the observer. We have been investigating how the representations of complex visual stimuli vary according to the task a participant is performing. First, we found that we could decode the task a participants was performing on a given visual object from activity in multiple regions throughout the brain. Further, we found a strong distinction between tasks that emphasized physical properties of the visual stimuli (e.g. color) and tasks that emphasized conceptual properties (e.g. real-world size). Second, we are now extending this work to visual scenes. Previously, we found that scene representations in a region of the brain thought to be critical for scene recognition primarily reflect the spatial properties of scenes (e.g. whether they are open or closed) and not the semantic properties (i.e. scene category). In our current work, we are investigating how these representations change according to task, by asking participants to focus on particular aspects of the scenes presented. 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.