Visual and semantic determinants of mental and neural representations of objects (C11#)

感觉信号的视觉处理和我们对视觉世界的语义知识紧密交织在一起。因此，我们关于视觉-知觉维度和语义维度对我们对物体的心理和神经表征的相对贡献以及我们形成类别的能力的知识仍然是零碎的。为了应对这一挑战，我们将进行一个详细的比较的心理和神经表征的视觉处理的对象图像和表征的对象词。通过使用大规模的人类相似性判断的在线众包，底层表征维度的计算建模，以及与这些维度的编码模型配对的密集采样功能性MRI，该项目有望揭示大脑和行为中视觉感知和语义信息之间相互作用的精确性质。

Visual processing of sensory signals and our semantic knowledge about our visual world are heavily intertwined. As a consequence, our knowledge about the relative contribution of visual-perceptual and semantic dimensions to our mental and neural representations of objects and thus our ability to form categories has remained piecemeal. To address this challenge, we will carry out a detailed comparison of the mental and neural representations underlying the visual processing of object images and the representation of object words. By using large-scale online crowdsourcing of human similarity judgments, computational modeling of the underlying representational dimensions, and densely-sampled functional MRI paired with encoding models of these dimensions, this project promises to reveal the precise nature of the interplay between visual-perceptual and semantic information in brain and behavior.