语用视角下基于认知理论的空间信息符号化表达机制研究

Although 3D representation of spatial information is more vivid than traditional 2D mode and makes it easier for users to understand the objective world, its huge amount of information goes against cognizing the virtual 3D world. What kind of 3D expression form is the best spatial cognition tool with the best cognitive effects is the one of focus problems that people concern about in this field. In this project, the issue is studied in depth from the perspective of pragmatics. Based on the pragmatic relevance theory, the expression mechanism of information having adequate association with cognitive objectives or results during the process of progressive spatial cognizing is researched emphatically. From the cartography and spatial reasoning methods, the reasoning algorithm of junction tree is developed and the scale corresponding relation between the occurrence probability of information with sufficient relevance and 3D symbols is explored. In addition, the overlaying rules of 3D cartographic symbols considering the semantic features and the fusing mechanism of 3D symbolic models based on the algorithm of Delaunay triangulation on different scales are established. Furthermore, the integration of symbolic models with multiple level of detailed is realized and thus the spatial context models which are the most relevant to cognition requirement are constructed. The dynamic characteristic and proceduring of spatial cognition under 3D environment is highlighted. Finally, in this research, the theoretical knowledge and methodology of spatial cognition are enriched and improved, which reflects the cognitive value of 3D symbols representation.

形象生动的三维表现形式使空间信息易于感知。但是三维表达所产生的巨大信息量不利于对空间的理解和认知。为了达到最佳的认知效果，什么样的三维表达形式是最佳的空间认知工具，这一问题值得关注。本项目从语用学的视角来研究此问题。依据语用关联理论，重点研究与认知目标或认知结果有足够关联性的信息在渐进式空间认知过程中的表达机制。从地图学和空间推理方法出发，提出了部分联结树推理算法，探讨了有足够关联性的信息出现的概率与三维符号在尺度上的对应关系。拟建立顾及语义特征的三维符号叠置规则和基于Delaunay三角网算法的不同尺度符号模型的融合机制，目的是实现多细节层次符号模型的集成从而构建与认知需求最具关联性的空间语境模型。研究突出空间认知动态性与过程化的特点，符合人类的认知心理。该研究丰富和完善了空间认知的理论知识与方法体系，体现了三维符号表达的认知价值。

本项目从语用学的视角来研究虚拟城市环境中建筑物的三维表达形式，实现了从巨大的信息量中获取认知所需信息，最终达到了最佳的寻路效果。依据语用关联理论，首先以现有的在线三维地图为实验环境，找出了使用者迷失方向或者产生寻路困难的原因。接下来重点研究了虚拟环境下影响面向寻路的渐进式认知过程的各种主观和客观因素。最后依照寻路的认知特点，研究定义了虚拟城市环境中建筑物的全局显著度和局部显著度两种变量，设计了基于距离的差异性评价方法确定各变量的权重，利用可见性因子综合每个显著度中空间对象、寻路者、环境三者之间的关系，建立显著度度量模型，最终量化了建筑物的显著度。本研究的目的是依据显著度模型量化结果，对应构建不同的细节层次的三维建筑物模型，最终产生具有等级结构差异性的虚拟城市环境，正确引导寻路者在虚拟环境下的寻路行为。研究突出空间认知动态性与过程化的特点，符合人类的认知心理。该研究丰富和完善了空间认知的理论知识与方法体系，体现了三维符号表达的认知价值。