Motivation for participation:

I have been investigating the dynamic representation of geographic space for a number of years. My initial goal was to find a representational approach for GIS that also conforms to human conceptualization of the world and how people organize spatial knowledge. This has since expanded to include a broader perspective of modeling "reality" as a means of understanding how people learn and gain understanding about their environment.

My perspective, therefore, is to apply a multidisciplinary approach to these issues, drawing upon concepts from a number of disciplines in addition to geography; from philosophy, perceptual psychology and linguistics, to database management systems, "traditional" GIS and mathematics.

Position statement: Geographic knowledge as a multi-representational and dynamic system

Our World View as a Cognitive Map

The most frequently used metaphor for describing the form of our geographic knowledge is the Cognitive Map. This term within the context of human cognition usually refers to the cognitive representation of geographical-scale space, and includes the immediate space of one's own neighborhood as well as very large and complex spatial entities such as towns, cities and whole environments. Because of their size, such entities cannot normally be seen in their entirety. It is the total collection of these mental maps, each representing a specific domain (my home town, the way to work, known countries of the world, etc.), that together comprise our World View. The idea of a cognitive map is a convenient and obvious metaphor for the representation of geographic space. The map as a graphic product is, after all, one of the most familiar means of storing and communicating knowledge about geographic space. As such, the metaphor of the cognitive map has been often used without further elaboration or explanation. There are implications of this metaphor that can be very misleading with regard to conceptual structure, and care must be taken to avoid more than the very broadest of interpretations.

The mental map metaphor conjures-up the notion of some unified representation as a graphic artifact inside each individual's head that contains the sum-total of that individual's geographic knowledge - that the cognitive representation of geographic knowledge is isomorphic with the graphical map and that we retrieve information by reading this "master map" with "the mind's eye." Certainly, our knowledge is highly interrelated, like the graphic map. Nevertheless, most known aspects of our cognitive representation of geographic space do not fit the mental map metaphor.

Examining the characteristics of a graphical map display relative to cognitive structure can nevertheless be illuminating with regard to the characteristics of the mental representation of geographic space. It seems intuitive that we do not literally store geographic knowledge as maps; using cartographic symbology, with dashed lines for dirt roads, etc. Most of our geographic knowledge, is not actually stored in any sort of graphic form, cartographic or otherwise. Even though we often talk about "mental imagery", mental images are also not exact copies of reality, analogous to a photograph. They are perhaps best described as image representations that have been possibly, but not necessarily, derived from some visual stimulus. Most adults are familiar with imagery conjured-up from reading a novel or a poem, or by listening to music. According to Piagetian development theory, there are image-like representations early-on in learning, but that they also play an increasingly subordinate role as the active component of memory as thought processes mature and become capable of abstract, logico-mathematical operations and knowledge becomes better organized.

A number of researchers have asserted that knowledge is encoded as a combination of image and non-image forms including pictorial, schematic, mathematical, textual, and auditory (Anderson 1978; Ioerger 1994; Hayward and Tarr 1995). But - doesn't this get us back to the metaphor problem again? Do we really have knowledge, particularly most of our "abstract" or "higher-level" knowledge encoded into words, mathematical equations or even (graphic?!) schematic diagrams? Just on the basis of intuition, perhaps with a little introspection, this doesn't ring true. Downs has asserted that knowledge within the mind has no form at all - that it is pure relation (Downs 1985). In his view, knowledge takes on an explicit form only for the purposes of communication; to convey information and/or facts to another. This is by definition external (as opposed to internal) knowledge representation. External representation can take many forms, including words, images, diagrams and maps. We must therefore maintain a careful and conscious distinction between external and internal forms. In (externally) describing the form of internal knowledge, using the normal, external forms we use to communicate seems almost automatic, whether or not they are really appropriate.

It is undeniable that we can indeed conjure-up visual imagery in our minds, which may be the result of a remembered visual experience or a mental creation of; a real-world scene, a drawing or a map. We have similar "imagery" with respect to specific words and mathematical equations - even music. All of these can be recalled as sensory sensations, whether they are completely imagined or based upon actual sensory experience. From the Piagetian perspective, sensory sensation is the beginning of learning. We also recall that the first type of knowledge within this framework is variously called figurative, declarative or (on a geographic scale) landmark knowledge. This consists of what might be termed experiential or observational knowledge - i.e., stored sensory sensation and is not abstract or derived knowledge. On this basis it would seem reasonable to say that such knowledge could be cognitively encoded as stored sensory sensations - which indeed could include a visual recollection of words on a printed page. Can we say that all other types of knowledge, i.e., the more "abstract" or "higher-level" knowledge is pure relation, as Downs has suggested? We have already discussed learning as a process of grouping and abstracting. Wouldn't this at least include the modification of remembered sensory sensations over time to fit accumulating knowledge and evolving points-of-view (in addition to simple forgetting)? Another argument potentially supporting the idea that at least some "abstract" or "higher-level" knowledge is encoded into visual, auditory or other external forms is again the argument that such external forms can and do channel how we think.

Tulving (1972) made a distinction between episodic and semantic. He described episodic memory as the kind of memory that receives and stores information about specific events and groups of events (i.e., episodes), and the space-time relationships among those events. This is the type of memory that deals with remembered experience. Semantic memory is the organized knowledge a person possesses about concepts and their interrelationships. Unlike episodic memory, semantic memory does not refer to unique episodes or events, but rather to universal principles. Information in episodic memory is recorded directly from perception and is susceptible to forgetting. Semantic memory, although it can be recorded directly (by, say, reading a textbook), is often derived through a combination of perception and thought. Through the learning process, certain events and episodes become associated with concepts in semantic memory as examples. It therefore seems reasonable to view our cognitive representation of geographic space as both semantic and episodic. Reading or talking about a neighborhood in our home town, or about some other familiar city may prompt visual memories of a restaurant we visited there, etc. Thus, events and episodes are remembered as sensory sensations, and are also a component of "higher-level" knowledge.

Another problematic aspect of the cognitive map metaphor is that it easily leads to the implicit assumption that the cognitive structure of our spatial knowledge is dynamic only in the sense of its original construction and subsequent modification as we learn. Nevertheless, there is a significant amount of empirical evidence showing that we have multiple cognitive representations for the same geographic environment (Kuipers 1978; Garling, Book et al. 1984; Bryant, Tversky et al. 1992; Franklin 1992.) Not only can we have representations in point-oriented, linear or survey form for a given domain because of our level of knowledge, our representation for knowledge belonging to the same domain changes depending upon the task at hand. Thus, people have a route-oriented representation for the path they drive to and from work every day and may generate a route-oriented representation when driving around a familiar city or when asked to give directions. As a specific experimental example, Taylor & Tversky have shown that people who read about an extended environment from either a route-perspective description or a survey-perspective description could answer questions utilizing apparently flexible perspectives. The perspective used in the text for the original learning task did not seem to affect the ability to answer such questions (Taylor and Tversky 1992). This gives further support to the notion that our brains structure representations on-the-fly.

In contrast to the properties of a map as a graphical artifact, then, the structure of our knowledge representation is not static and monolithic, but is dynamic and multifaceted (Montello 1992). Our cognitive spatial representation can be discontinuous or linear (route-oriented), segmented, and incomplete. Accommodation of these properties has lead Barbara Tversky (1993) to describe our World View as a cognitive collage rather than a cognitive map. This, however, derives from the map metaphor and retains some of the same problems, such as the lack of reflecting the dynamic nature of our cognitive spatial representation. Perhaps a better way to describe it is as another multi-stable system. As such, the various representational forms of the environment are not stored in any static fashion. Representations are dynamically generated, depending upon the initializing circumstances and can be in any of a number of forms and modes (visual, verbal, etc.).

How is spatial knowledge encoded?

The classical approach assumes a complex intern al representation in the mind that is constructed through a series of specific perceived stimuli, and that these stimuli generate specific internal responses. Research dealing specifically with geographic-scale space has worked from the perspective that the macro-scale physical environment is extremely complex and essentially beyond the control of the individual. This research, such as that of Lynch and of Golledge and his colleagues, has shown that there is a complex of behavioral responses generated from correspondingly complex external stimuli, which are themselves interrelated. Moreover, the results of this research offers a view of our geographic knowledge as a highly interrelated external/internal system. Using landmarks encountered within the external landscape as navigational cues is the clearest example of this interrelationship.

Portugali (1996) has recently extended this view and explicitly acknowledged a complex interrelationship between our internal representation and the external environment. Furthermore, he asserts that elements in the external environment act as an interrelated knowledge representation, external to ourselves, that functions in concert with our internal knowledge representation.

The rationale is as follows: We gain information about our external environment from different kinds of perceptual experience; by navigating through and interacting directly with geographic space as well as by reading maps, through language, photographs and other communication media. With all of these different types of experience, we encounter elements within the external world that act at symbols. These symbols, whether a landmark within the real landscape, a word or phrase, a line on a map, or a building in a photograph, trigger our internal knowledge representation and generate appropriate responses. In other words, elements that we encounter within our environment act as knowledge stores external to ourselves.

Each external symbol has meaning that is acquired through the sum of the individual perceiver's previous experience. That meaning is imparted by both the specific cultural context of that individual and by the specific meaning intended by the generator of that symbol. Of course, there are many elements within the natural environment not "generated" by anyone, but that nevertheless are imparted with very powerful meaning by cultures (e.g., the sun, moon and stars). Manmade elements within the environment, including elements such as buildings, are often specifically designed to act as symbols as at least part of their function. The sheer size of downtown office buildings, the pillars of a bank facade and church spires pointing skyward are designed to evoke an impression of power, stability or holiness, respectively.

These external symbols are themselves interrelated, and specific groupings of symbols may constitute self-contained external models of geographic space. Maps and landscape photographs are certainly clear examples of this. Elements of differing form (e.g., maps and text) can also be interrelated. These various external models of geographic space correspond to external memory.

From the perspective just described, the sum total of any individual's knowledge is contained in a multiplicity of internal and external representations that function as a single, interactive whole. The representation as a whole can therefore be characterized as a synergistic, self-organizing and highly dynamic network.

References

Anderson, J. R. (1978). "Arguments Concerning Representations for Mental Imagery." Psychological Review 85(4): 249-276.

Bryant, D. J., B. Tversky, et al. (1992). "Internal and External Spatial Frameworks for Representing Described Scenes." Jornal of Memory and Language 31: 74-98.

Downs, R. (1985). The Representation of Space: Its Development in Children and in Cartography. The Development of Spatial Cognition. R. Cohen. Hillsdale, NJ, Lawrence Erlbaum Associates: 323-344.

Franklin, N. (1992). "Spatial Representation for Described Environments." Geoforum 23(2): 165-174.

Garling, T., A. Book, et al. (1984). "Cognitive Mapping of Large-Scale Environments." Environment and Planning 16(1): 3-34.

Hayward, W. G. and M. J. Tarr (1995). "Spatial Language and Spatial Representation." Cognition 55: 39-84.

Ioerger, T. R. (1994). "The Manipulation of Images to Handle Indeterminacy in Spatial Reasoning." Cognitive Science 18: 551-593.

Kuipers, B. (1978). "Modeling Spatial Knowledge." Cognitive Science 2: 129-153.

Montello, D. R. (1992). The Geometry of Environmental Knowledge. International Conference GIS - From Space to Territory: Theories and Methods of Spatio-Temporal Reasoning. A. U. Frank, I. Campari and U. Formentini. Pisa, Italy, Springer-Verlag.

Portugali, J., Ed. (1996). The Construction of Cognitive Maps. The GeoJournal Library. Dordrecht, Kluwer Academic Publishers.

Taylor, H. and B. Tversky (1992). "Descriptions and Depictions of Environments." Memory & Cognition 20(5): 483-496.

Tulving, E. (1972). Episodic and Semantic Memory. Organization of Memory. E. Tulving and W. Donaldson. New York, Academic Press: 382-403.

Tversky, B. (1993). Cognitive Maps, Cognitive Collages, and Spatial Mental Models. Spatial Information Theory: A Theoretical Basis for GIS. A. Frank and I. Campari. New York, Springer-Verlag. 716: 14-24.

Short Vita

Current Position

Professor of Geography and Associate of the Earth Systems Science Center, The Pennsylvania State University

Professional Experience

After receiving a Ph.D. in Geography from the State University of New York at Buffalo in 1977 specializing in geographic information systems, she worked at the U.S. Geological Survey National Center in Reston, Virginia for three years. She then took a position at the University of California at Santa Barbara before joining The Pennsylvania State University in 1986.

Her research interests are in the areas of geographic knowledge representation theory, spatio-temporal data models, spatial cognition, spatial analysis methodologies and GIS design. Current research efforts include the representation of time and temporal dynamics in spatial databases, human cognition of space-time, and extension of formal analysis techniques in a GIS context.

She is currently a member of the Editorial Board of the International Journal of Geographical Information Systems, a member of the American Association of Geographers (AAG) National Committee on Digital Data Formats and Standards, and is the Penn State representative to the University Consortium for Geographic Information Sciences (UCGIS) and member of that organization's Research Committee . She is a past member of the Board of Directors of the American Cartographic Association and the Board of Directors of the University Consortium for Geographic Information Sciences.

Recently, she was also co-organizer (with Dr. Barry Smith, Philosophy, SUNY Buffalo) of an interdisciplinary workshop on The Ontology of Fields, sponsored by the National Consortium for Geographic Information and Analysis, to be held in Bar Harbor, Maine, June, 1998, organizer of the Panel on Extensions to Geographic Representation at UCGIS Annual Assembly, Park City, Utah, June, 1998, Director of the Auto Carto 12 Conference, held in conjunction with American Congress on Surveying and Mapping/ American Society for Photogrammetry and Remote Sensing, Charlotte, North Carolina, 1995. She was also a keynote speaker for the Symposium on Advanced Geographic Data Modeling, sponsored by the Netherlands Geodetic Commission, held in Delft, Holland, 1994.

Selected Publications

Peuquet, Donna J. "Time in GIS and Geographical Databases" in Geographical Information Systems: Principles and Applications (Maguire, Goodchild, Rhind, and Longley, Eds.), 2nd edition (London: Longman) in press.

Qian, Liujian, Monica Wachowicz, Donna Peuquet and Alan MacEachren, "Delineating Operations for Visualization and Analysis of Space-Time Data in GIS," pp. 872-878 in Proceedings, GIS/LIS '97, Cincinnat (1997).

Peuquet, Donna J. and Liujin Qian. "An Integrated Database Design for Temporal GIS," pp. 2-1 to 2-11 in Proceedings, Seventh International Symposium on Spatial Data Handling, Delft, Netherlands(1996) (Also separately published by Taylor & Francis, 1997).

Peuquet, Donna J. and Niu Duan. "An Event-Based Spatio-Temporal Data Model for Geographic Information Systems," International Journal of Geographical Information Systems, 9(1):7-24 (1995).

Peuquet, Donna J. "A Conceptual Framework for the Representation of Temporal Dynamics in Geographic Information Systems," Annals of the Association of Americal Geographers, 84(3):441-461 (1994).