Thomas Barkowsky

Spatial Cognition: Reasoning, Action, Interaction

The Transregional Collaborative Research Center SFB/TR 8 Spatial Cognition pursues interdisciplinary research on a broad range of topics related to the representation and processing mechanisms for intelligent spatial behavior in technical and in natural systems. This contribution gives an overview of the field of research worked on in the SFB/TR 8 Spatial Cognition and presents three representative examples that illustrate the activities in the three research areas Reasoning, Action, and Interaction.

Schematizing Maps: Simplification of Geographic Shape by Discrete Curve Evolution

Shape simplification in map-like representations is used for two reasons: either to abstract from irrelevant detail to reduce a map users cognitive load, or to simplify information when a map of a smaller scale is derived from a detailed reference map. We present a method for abstracting simplified cartographic representations from more accurate spatial data. First, the employed method of discrete curve evolution developed for simplifying perceptual shape characteristics is explained. Specific problems of applying the method to cartographic data are elaborated. An algorithm is presented, which on the one hand simplifies spatial data up to a degree of abstraction intended by the user; and which on the other hand does not violate local spatial ordering between (elements of) cartographic entities, since local arrangement of entities is assumed to be an important spatial knowledge characteristic. The operation of the implemented method is demonstrated using two different examples of cartographic data.

Cognitive Requirements on Making and Interpreting Maps

We present an approach to modeling human interpretation of (real) geographic maps. While in Geographic Information Systems (GIS) the limitations for describing geographic knowledge mainly stem from the limitations of this knowledge itself, paper maps pose additional constraints on the representation of spatial configurations. We examine maps as representation media with respect to cartographic restrictions involved in the map making process. Some cognitive factors of cartographic generalization are indicated. We present our aspect map approach allowing for describing maps formally as pictorial representations. The approach postulates the use of meta-knowledge to enable adequate map interpretation. Phenomena of cartographic interpretation and misinterpretation are illustrated employing two kinds of hierarchic structures of spatial aspects of maps. The notions we present can be employed in augmenting the ‘cognitive adequacy’ of automated map making and map reading.

The Cognitive Reality of Schematic Maps

In graphics and language, schematisation is an important method to emphasize certain aspects and to deemphasize others. Different disciplines use schematisation for different reasons. In cartography, graphic schematisation is one aspect of map generalisation. In contrast, cognitive science addresses schematisation as a method to intentionally emphasize certain aspects of knowledge beyond technical necessity; therefore, the notion of schematic map is proposed to denote maps that employ schematisation for cognitive representational reasons. This chapter discusses different views of schematisation from cartography, linguistics, and artificial intelligence. Connections to qualitative reasoning in artificial intelligence are drawn. We address human spatial cognition and present examples of task-oriented representations. Finally, multimodality for conveying spatial knowledge and its application in schematic maps are discussed.

Spatial Representation with Aspect Maps

This paper describes the aspect map approach to model the processing of geographic maps. Geographic maps are described as spatial representation media which play an important role in many processes of human spatial cognition. We focus on the aspectuality of representation and therefore deal with aspect maps: spatial organization structures in which one or more aspects of geographic entities are represented. The aspect map architecture is presented, an AI model of processing geographic maps. Two processes contained in this model are investigated in more detail. The first is the transformation of one aspect map into another aspect map which only retains selected entities and aspects (extraction). The second process is the combination of two aspect maps, in order to obtain a third aspect map. The results of an empirical study show that the formal approach can describe and distinguish the ways in which people solve this task.

Schematic Maps for Robot Navigation

An approach to high-level interaction with autonomous robots by means of schematic maps is outlined. Schematic maps are knowledge representation structures to encode qualitative spatial information about a physical environment. A scenario is presented in which robots rely on highlevel knowledge from perception and instruction to perform navigation tasks in a physical environment. The general problem of formally representing a physical environment for acting in it is discussed. A hybrid approach to knowledge and perception driven navigation is proposed. Different requirements for local and global spatial information are noted. Different types of spatial representations for spatial knowledge are contrasted. The advantages of high-level / low-resolution knowledge are pointed out. Creation and use of schematic maps are discussed. A navigation example is presented.

Schematic Maps as Wayfinding Aids

Schematic maps are effective tools for representing information about the physical environment; they depict specific information in an abstract way. This study concentrates on spatial aspects of the physical environment such as branching points and connecting roads, which play a paramount role in the schematization of wayfinding maps. Representative classes of branchingpoints are identified and organized in a taxonomy. The use of prototypical branching points and connecting road types is empirically evaluated in the schematization of maps. The role played by the different functions according to which the map is classified is assessed, and main strategies applied during the schematization process are identified. Implications for navigational tasks are presented.

Mental Processing of Geographic Knowledge

The contribution presents a computational modeling approach to geographic knowledge processing in mind. Geographic knowledge is assumed to be stored in a piecemeal manner. Spatial knowledge fragments form a hierarchical structure of lean knowledge. An actual mental image representation is constructed when needed to perform a specific task. In this construction process missing information is complemented to create a determinate mental image. - First, the artificial intelligence perspective taken is elaborated. After a short review of conceptions on mental processing of spatial knowledge from psychology and artificial intelligence we outline the model MIRAGE. The internal structure and the operating of the model is elaborated using an exemplary scenario. Problems in constructing mental images from given pieces of knowledge are demonstrated and discussed. The paper concludes with a discussion of the approach with respect to its modeling objective. We point to further research questions and to potential applications.

Casimir: An Architecture for Mental Spatial Knowledge Processing

Mental spatial knowledge processing often uses spatio-analogical or quasipictorial representation structures such as spatial mental models or mental images. The cognitive architecture Casimir is designed to provide a framework for computationally modeling human spatial knowledge processing relying on these kinds of representation formats. In this article, we present an overview of Casimir and its components. We briefly describe the long-term memory component and the interaction with external diagrammatic representations. Particular emphasis is placed on Casimirs working memory and control mechanisms. Regarding working memory, we describe the conceptual foundations and the processing mechanisms employed in mental spatial reasoning. With respect to control, we explain how it is realized as a distributed, emergent facility within Casimir.

Spatial principles in control of focus in reasoning with mental representations, images, and diagrams

The effective control of attentional focus is an essential requirement in mental reasoning based on mental models and men tal images, as well as in the interaction with external diagrams. In this paper, we argue for spatial or ganization principles common to various mental subsystems that entail a non-centralistic con trol of focus. We give a brief overview of mental spatial rea soning and present a review of psy chological findings related to cognitive con trol. We review existing mod eling approaches that realize control of focus in imagery, scene recognition, and men tal animation. Based on these founda tions, we identify basic spatial or ganizing principles that are shared by the diverse subsystems col laborating in mental spatial reasoning. We discuss the implica tions of these principles in the frame work of a computational modeling ap proach and give an outline of the con ception of control of focus in our com putational architecture Casimir.