The Processing of Semantics and Syntax in Naturalistic Scene Viewing

自然场景观看中的语义和句法处理

• 批准号: 159914243
• 负责人: Professorin Dr. Melissa Le-Hoa Vo
• 金额: --
• 依托单位: Visual Attention Laboratory
• 依托单位国家: 德国
• 项目类别: Research Fellowships
• 财政年份: 2009
• 资助国家: 德国
• 起止时间: 2008-12-31 至 2011-12-31
• 项目状态: 已结题
• 来源: https://gepris.dfg.de/gepris/projekt/159914243?language=en
• 关键词: Processing; Semantics; Syntax; Naturalistic; Scene

When we look around, the world does not appear to us as a random aggregation of isolated objects. Rather, the arrangement of objects in naturalistic scenes conforms to certain rules. The aim of this proposal is to gain further insight into how we use prior experience and expectations regarding object-scene relationships to efficiently guide our attention when viewing naturalistic scenes. We want to provide the grounds for a theoretical framework, according to which objects in scenes — similar to words within sentences — are processed in both a semantic (i.e., content related) and a syntactic (i.e., structure related) manner. By selectively violating expectations regarding a scene's semantic or syntactic composition, we will be able to investigate the effects of object-scene inconsistencies on attention allocation as well as their temporal dynamics by measuring behavioral performance, eye movements, and event-related potentials. A greater knowledge of what mechanisms are involved in the effective control of attention in real-world scenes will contribute to several core cognitive science disciplines such as cognitive psychology, visual neuroscience, or machine vision.

当我们环顾四周时，世界在我们看来并不是孤立对象的随机聚集。相反，自然主义场景中物体的布置符合某些规则。该提案的目的是进一步了解我们如何利用对象场所关系的先前经验和期望，以在观看自然主义场景时有效地指导我们的注意力。我们希望为理论框架提供理由，根据哪些语义（即与内容相关）和语法（即与结构相关）方式进行处理的场景中的对象（类似于句子中的单词）。通过有选择地违反对场景的语义或句法组成的期望，我们将能够通过测量行为性能，眼动和事件相关的潜力来研究对象场景不一致对注意力分配以及它们的临时动态的影响。在实际场景中有效控制注意力的有效控制中涉及哪些机制将有助于几个核心认知科学学科，例如认知心理学，视觉神经科学或机器视觉。