Klippel Alexander
Department of Geography, The Pennsylvania State University.
Top Cogn Sci. 2011 Oct;3(4):722-40. doi: 10.1111/j.1756-8765.2011.01159.x.
This article discusses an approach to characterizing movement patterns (paths/trajectories) of individual agents that allows for relating aspects of cognitive conceptualization of movement patterns with formal spatial characterizations. To this end, we adopt a perspective of characterizing movement patterns on the basis of perceptual and conceptual invariants that we term movement choremes (MCs). MCs are formally grounded by behaviorally validating qualitative spatio-temporal calculi. Relating perceptual and cognitive aspects of space and formal theories of spatial information has shown promise to foster understanding of the semantics of movement patterns. Specifically, we discuss our approach in relation to existing qualitative formalisms such as the region connection calculus and the Egenhofer's intersection formalisms. We show that the MC approach is compatible with these approaches but offers additional opportunities to improve the cognitive adequacy of these formalisms. We summarize this paper using a movement taxonomy that also provides guidance for future research.
本文讨论了一种刻画个体智能体运动模式(路径/轨迹)的方法,该方法能够将运动模式的认知概念化方面与形式空间刻画联系起来。为此,我们采用一种基于感知和概念不变量来刻画运动模式的视角,我们将这些不变量称为运动编排(MCs)。MCs通过对定性时空演算进行行为验证而在形式上得到支撑。将空间的感知和认知方面与空间信息的形式理论联系起来,已显示出有助于增进对运动模式语义的理解。具体而言,我们将我们的方法与诸如区域连接演算和埃根霍费尔的交集形式体系等现有的定性形式体系相关联进行讨论。我们表明,MC方法与这些方法兼容,但提供了额外的机会来提高这些形式体系的认知充分性。我们使用一种运动分类法总结本文,该分类法也为未来的研究提供指导。