15
    City Games as a Framework for Studying Spatial Information and the Dynamics of Urban Design
    Hernan Casakin
    Abstract Cognitive mapping is presented as an approach to analyze the interplay between internal and external spatial information during urban design activity. City Games is proposed for analyzing spatial information in a simulated urban environment, and for gaining insight in the dynamics of city development. Empirical results showed that students purposefully created spatial structures in the city according to clearly identifiable cluster formations. Two main cognitive strategies dealing with clusterafter-cluster, and cluster-simultaneity developments were identified during the process of constructing a city. Each cognitive strategy generated different spatial organization patterns of the city. Keywords City Games, cognitive maps, spatial representations, cognitive strategies, urban design
    
    Introduction
    Cognitive maps are characterized by a powerful capability to communicate spatial information. Cognitive maps help understand how people perceive, remember, and represent in their minds spatial information about the environment (e.g., Jeffery and Burgess, 2006; Foo et al., 2005; Konar. and Chakraborty, 2005). According to Lynch (1960), the production of cognitive maps constitutes a valuable tool for gaining further insight in the mental image that people have about the cities they live in.
    
    354
    
    From Negations to Negotiations
    
    Cognitive mapping also aids in the understanding of creative acts such as the design and construction of large-scale artifacts. During this process, people (who have the innate cognitive capability to design) establish a fluid interplay between their internal representations, and those representations of the external world (Portugali, 1996a). As the design process evolves, the designer actively constructs, modifies, and updates his/her internal representations of the design artifact. Due to the complexity of the urban design process, the production of externalizations by means of traditional sketch map drawings results unsuitable to represent spatial information. As a result, the process of establishing interactions between internal and external representations is not always understood. In this study we attempt to redefine the conventional notion of cognitive mapping as proposed in the 1960s by considering an empirical tool called City Games. This device is used for the artificial construction of a city in a simulated physical environment. Questions to be addressed in this investigation are how City Games aids to gain insight in the dynamics of a city in constant change and evolution, what are the cognitive processes used in the artificial construction of a city, and how these affect spatial information represented in the playground. The following two sections deal with an introduction to the research background. Section 3 is concerned with cognitive mapping, and the spatial quality of the city. Section 4 deals with environmental images, cognitive maps, and the design of cities. City Games is proposed as a tool for representing and analyzing the design and evolution of cities. Section 5 presents an empirical study carried out with urban design students. Main results and conclusions of the investigation are offered in the last two sections.
    
    Cognitive Mapping
    Part of the mental schema that people use to organize information is referred by psychologists as the cognitive map. Cognitive capabilities allow subjects to represent in their minds spatial information about the spatial structure of the physical environment. Due to its large scale, the environment cannot be perceived as a
    
    City Games as a Framework for Studying Spatial Information...
    
    355
    
    whole, and as a result has to be represented in smaller parts (Portugali, 1996a). In a pioneer study, Tolman (1948) referred to these partial mental representations as cognitive maps. He found that some properties of cognitive maps seemed to be essentially qualitative representations of the physical environment rather than exact representations of it. Particular attention was devoted to cognitive mapping in fields like human geography (e.g.; Lowenthal, 1967; Soini, 2001), as well as in architecture and urban design (e.g.; Lynch, 1960). Lynch studied cognitive maps with deliberate attention directed to the form of cities. His investigations centered on the relationship between the spatial structure of cities and the way people internally represent them. An advantage of cognitive maps is that they can expose commonalities or differences regarding the way people understand the environment. These tools also enable to learn how spatial information is perceived, processed, and communicated by city form (e.g.; Gould, 1973). Cognitive maps also serve to assess, and improve the spatial quality of the city. For example, Milgram and Jodelet (1976) used cognitive maps to identify the most frequent cited elements from Paris, and assessed the psychological impact that they have on people. Nassar (1998) combined Milgram and Jodelet’s concept of psychological maps, and Lynch’s (1960) idea of relating psychological information to city form to study city appearance. He carried out empirical tasks in which people were asked to construct an ‘evaluative’ cognitive map, and inform about visually liked and disliked areas of the city. Cognitive maps can be seen as the representation that a single individual has about the physical environment. We call this an individual cognitive map. In this process, the physical environment communicates visual information, and the observer creates his/her own interpretation according to past experiences. However, a cognitive map can also be the result of a vast number of individual representations carried out by a number of subjects, also called composite or collective cognitive map. The idea behind collective maps is that they represent a common view of the environment. Although people are able to create a personal interpretation about
    
    c:\prsnfinal24/chapter 15
    
    356
    
    From Negations to Negotiations
    
    the spatial information contained in the physical environment, at a certain level of abstraction they share a vision about stronger commonalities of the environmental image. Collective cognitive maps are seen as a summary of an entire sample of individual cognitive maps that aid to visualize those elements and spatial relationships of the physical environment that are more significant (e.g., Young, 1999). In the current investigation we analyze the collective representation of spatial information by empirically testing individual cognitive maps, during the artificial construction of a city.
    
    Lynch and the Spatial Quality of the City
    Four decades ago Kevin Lynch (1960) reported in his book ‘The image of the city’ the fundamental role of cognitive representations in relation to cities. His purpose was to understand what makes a city to be more legible and imageable. Lynch considered the concept of ‘imageability’ as a theoretical basis for the study of cognitive maps, urban form, and spatial quality of cities. According to Lynch, imageability “is that quality in a physical object which gives it a high probability of evoking a strong image in any given observer” (Lynch, 1960, 9). The image of a city is actually the overlap of many individual images “…which are the result of a two-way process between the observer and his [her] environment. The environment suggests distinctions and relations, and the observer … selects, organizes and endows with meaning what he sees” (Lynch, 1960, p.6). Lynch claimed that spatial quality is the result of the clarity or legibility of the general view of the city. Spatial quality is also the degree to which different portions of a city can be identified, remembered, represented, and organized into a coherent and consistent whole. Coherence enhances preference in how elements are set together, and its effect in the understanding of the immediate environment (Kaplan & Kaplan, 1982). According to Ramadier and Moser (1998) the complexity of the urban structure, the differentiation of urban elements, and its visual characteristics are the major variables that affect legibility in terms of spatial representation. Other researchers claimed that spatial representation is not only related to Euclidean information of the environment, but
    
    City Games as a Framework for Studying Spatial Information...
    
    357
    
    also on cognitive processes of categorization, as well as on the formation of cluster hierarchies (Hirtle & Jonides, 1985; Kitchin, 1994). Kaplan, (1979) argued that the importance of legibility resides in the degree of uniqueness that allows people to categorize and represent the physical environment.
    
    Environmental Images, Cognitive Maps, and the De]ities
    Portugali (2002) claimed that a city as a representation of itself can be viewed as the result of a combination between spatial properties of the city and basic cognitive strategies that are active in the evolution and dynamics of cities. Designers dedicate vast efforts to generate representations endowed with powerful spatial features that are easy to remember. One of the reasons is that a well designed city has better chances to transfer a sharper image in the mental representations of each of its residents and users (Casakin, 2003). The relationship between cognitive mapping and urban design was studied by researchers like Aitken, Cutter, Foote and Sell (1989), Golledge and Stimson (1997), and Madinapour (1996). Recently, Halseth and Doddridge (2000) introduced the KIDSMAP, a project based on cognitive maps whose aim was to identify ways to collect information regarding interests of children in their environment. But the seminal work of Lynch (1960) was the first to use the understanding on cognitive processes for better urban design. Lynch’s urban image theory (Lynch, 1960) aided to understand how city residents perceive and cognize their cities. Other researchers such as Greene (1992) studied the legibility of cognitive maps as a fundamental foundation for planning. The underlying order derived from a legible city allows people to understand and become familiar with the environment, move, spatially orientate, find their way, and communicate among each other. The extent to which the legibility of a city can be enhanced by certain distinctive environmental elements and particular spatial configurations is a topic of large interest and debate. Urban designers still have a contribution to make in order to improve the design of cities, and to help shape the cognitive maps of its inhabitants. In the
    
    c:\prsnfinal24/chapter 15
    
    358
    
    From Negations to Negotiations
    
    current study, the concept of legibibility is considered to study the evolution of city form, and its effect on spatial quality. Design Artifacts and the Evolution of Cities Cities are considered to be large-scale design artifacts. They are the collective outcome of a synergetic process under which thousands of participants act locally in a relatively independent manner (Casakin and Portugali, 2005; Portugali, 1999). Although the synergetic actions carried out during the development of a city seem to be chaotic and rather un-coordinated, the design output is supposed to be an ordered artifact. Despite vast energy is spent in the planning and design of cities, the general spatial structure emerges spontaneously, that is through self-organization (Portugali, 1997; 1999; 2004). Characteristic in self-organizing processes is that their basic component entities are individuals, each of which is considered as a self-organizing system. In these processes, participants are individual agents (e.g. urban designers) responsible for both their own specific self-organization process, and the collective selforganization process of the city. On the other hand, the selforganizing process of the city as a whole plays a role in the specific self-organization process of each individual agent. While studying self-organizing processes, Portugali and Casakin (2002) established a main distinction between design artifacts, which can be designed in a pre-determined way, and those that cannot be designed in a pre-determined way. The former are referred as ‘engineerable’ design artifacts, (e.g.; tools, buildings, bridges, etc.) of which the designer can successfully predict its final form and behavior. The latter are referred as self-organizing design artifacts, where the city is the best example. One of the reasons that a city cannot be designed a priori result from the fact that a city is a large and complex artifact. These aspects do not enable designers acting in it to fully control its development, emerging structure, and final shape (Portugali, 1999). The size difference between small and large artifacts entails qualitative differences with respect to internal and external representation. For example, it is not possible to perceive or internally represent a city in its totality, and these influence the
    
    City Games as a Framework for Studying Spatial Information...
    
    359
    
    process of external representation. A consequence of this is that internal representations of cities are frequently schematized, and distorted. The role of internal representations in the design of cities was not acknowledged in most design cognition studies. Small-scale objects (Smith, 1999) captured the attention of studies in cognitive science, but very large-scale objects such as cities were overlooked. This is one of the reasons why many cognitive processes and properties that are prominent in the domain of cities and other large objects have been repeatedly disregarded (Portugali, 2002).
    
    City Games as a Tool for Representing and Analyzing the Construction and Evolution of Cities
    Designers are fluent in creating, changing, and updating their representations about the design artifact in progressive development. The design of large-scale artifacts such as cities is a complex design activity, where the interplay between internal and external representations is not always fully understood. City Games were suggested by Portugali (1996b) as a tool to illustrate and study synergetic, and self-organized cognitive process, (referred to in the Synergetic Inter-representation Networks theory- SIRN) while constructing internal representations (cognitive maps) about the city. City Games are viewed as a new type of the so-called Bartlett scenarios (1932/1961), (Haken and Portugali, 1996; Hatna, et al; 2001). In classic Bartlett scenario a participant is shown a figure and is requested to memorize it. Afterwards, he/she is asked to externally represent it. This externalization is given to another participant, and so forth. The typical outcome of these scenarios is that after several strong fluctuations in the externalizations, the figure becomes stable and does not dramatically change from iteration to iteration. The importance of the effect of enslavement is referred in the externalization and representation of sequential iterations during the construction of a city. City Games approach provide an empirical framework for simulating and visualizing the evolution of cities. It allows analyzing the way designers experience and perceive
    
    c:\prsnfinal24/chapter 15
    
    360
    
    From Negations to Negotiations
    
    the city, remember, acquire knowledge, and modify its spatial structure. During the game, participants take action in accordance to two main aspects: (i) their internal representations constructed in the mind; (ii) the dynamic situation in the playing ground, which progressively evolves and changes while the game develops. During this process, subjects internalize the external situation, and locate buildings on the playground depending on their own interpretation of the emerging spatial order. City Games embrace all the components of the self-organizing process given by: a sequential interplay between internal and external representations, the emergence of a complex artifact, and a typical synergetic process of self-organization (Haken and Portugali, 1996).
    
    Empirical Study
    In this work, City Games are tested empirically to assess the visual quality of the city, and the cognitive processes involved in its construction. City Games may contribute to gain insight in the way people, design students in particular, interact with their representations during the spatial evolution of cities. We aim to explore how spatial patterns emerge during the process in which cities constantly evolve. We investigate design strategies considered while constructing a city, and spatial qualities that contribute to the legibility of the final outcome.
    
    Research Goals
    The first goal is to study what is the emergent structure of an ‘ideal’ city represented in the playground. We want to explore whether there is a coherent image of the city shared by students, what are the spatial features of such image, and how it affects the quality and legibility of the structure of the city. We aim to study if students, considered to be self-organizing systems, organize and locate physical objects such as buildings according to: (i) a continuous pattern where no individual clusters can be distinguished, (ii) a random pattern where individual clusters are difficult to visualize, or (iii) a pattern composed of legible clusters. In addition, we will explore possible relationships between the spatial characteristics of the
    
    City Games as a Framework for Studying Spatial Information...
    
    361
    
    emergent structure of the city, and the existence of legible clusters organized by functions (i.e., dwellings, services, and industries). The second goal is to study cognitive strategies used by students to construct the city, and their effect on the evolution of the spatial structure. A question to be addressed is whether the principle of enslavement proposed by Bartlett (1932/1961) and by Haken and Portugali (2003) guides and controls the design of the city. In particular, we aim to verify whether students construct the image of their cities by: (i) acting globally (i.e., through a cluster- simultaneity development strategy); or (ii) acting locally (i.e., by means of a cluster-after-cluster development strategy). We also intend to assess possible effects of cognitive strategies on the spatial features, and the functional organization of the final outcome.
    
    Description of the Empirical Study
    Subjects. 30 third year undergraduate students belonging to urban studies volunteered their time to participate in the experiments. The design task. The task consisted in the design of an ideal city within a playing ground simulating a real physical environment. Students were provided with a collection of mock-ups, representing buildings with variable size, and geometry (Figure 1). They were requested to label each building according to three different functional categories dealing with dwelling, industry, and business. Thereafter, students were asked to construct the city by locating the mock-ups within the playing ground delimited in the floor (Figure 2).
    
    Figure 1: Example of 1:100 Mock-ups Provided to Students.
    
    c:\prsnfinal24/chapter 15
    
    362
    
    From Negations to Negotiations
    
    Figure 2: Illustration of a Typical City Game Design Process
    
    Procedure. Students were provided with a sheet with general instructions, and a description of task requirements. They were given a set of fifty two 1:100 mock-ups randomly distributed near the playing ground. Thereafter, participants were requested to design the city by locating the mock-ups one after the other. The construction of the city was video recorded and analyzed a posteriori. The topological location of mock-ups in the playing ground was registered using Cartesian coordinates, and their position was measured by considering an eight sector polar grid. The task was concluded when all the available mock-ups were located in the playing ground. Each individual design session lasted around 20-30 minutes. Statistical Analyses. Data from the experiments related to cluster organization in the city was submitted to Nearest Neighbor Test. This test enabled to analyze and classified the pattern of spatial information in a city according to three different categories of cluster organizations. The first category of spatial arrangement is concerned with highly legible cluster organizations. It reflects a clear intention of the designer to develop the structure of the city according to strong cluster formations. This category stands for values that are close to 0. The second category corresponds to a random order of the city structure that difficulty the legibility of clear clusters. It is referred by values around 1. The third category corresponds to a
    
    City Games as a Framework for Studying Spatial Information...
    
    363
    
    continuous ordered structure that as in the previous category, does not makes possible the legibility of clusters. It is represented by values that approach 2.14. Neither in the second category nor in the third one, spatial order of the city reflects a clear intention of the designer to create cluster organizations. In order to study the way that students design their cities, we analyzed how cluster organizations developed during the design process. For this purpose we verified the number and kind of iterations generated during the process among the clusters. Iterations were considered to be representations that substantially differ to those of the conventional cognitive map. In this study, we use this concept to analyze sequences generated in the process of creating and developing clusters. Cluster heterogeneity is referred as the number of discontinuities produced in each cluster (Figure 3). As a cluster grows in number of buildings more moves are expected. In order to measure cluster heterogeneity we divided the number of iterations to construct a cluster by the number of buildings of the same cluster. Two buildings were considered to belong to a same cluster if there was a maximum distance of 40cm between their boundaries. Data obtained from the experiment was submitted to Nearest Neighbor, Pearson R square, and T-Tests for statistical analyses.
    
    (a)
    
    (b)
    
    c:\prsnfinal24/chapter 15
    
    364
    
    From Negations to Negotiations
    
    Results
    Spatial characteristics of the emergent city. First, we analyzed the extent to which cities were organized according to cluster organizations. What we wanted to know was whether the design of cities is based on: (1) an intended order based on clear cluster organizations; (2) a random order that does not make possible the legibility of clear clusters; or (3) a continuous ordered structure that neither makes possible the legibility of clusters. Results showed significant differences between the first category and the rest. An intended order of the city based on clear cluster organizations was found. (Table 1).
    
    Table 1 Organization of Cities According to Clusters Subjects Nearest neighbor analysis 30 Minimum .181 Maximum .657 Mean .404 Std. Dev. 1.44
    
    Spatial organization of the city and functions. Further to the analysis of the spatial structure of the city, we assessed relationships between emerging clusters, and their functional organization (i.e., dwellings, services, and industries). Each cluster was labeled according to its dominant function, defined as the most iterated function per cluster. The sum of dominant functions through the total number of clusters was divided by the total number of buildings in the city. Results indicated that the average of dominant functions per cluster in all cities constructed by students was about 85%. City development and cluster organization. In order to learn how cities evolved through time, we investigated strategies used for its construction. Two main cognitive strategies were observed. In the first one, clusters were sequentially developed one after the other (cluster-after-cluster strategy), and in the second one clusters were simultaneously developed (cluster-simultaneity strategy). A
    
    City Games as a Framework for Studying Spatial Information...
    
    365
    
    significant relationship was found between the cognitive strategies used to construct the city, and the number of clusters produced. Cities that were organized by means of the cluster-simultaneity strategy used to be structured by a large number of cluster configurations. On the contrary, when cities were constructed by means of the cluster-after-cluster strategy, the spatial structure was characterized by a reduced number of clusters. This Pearson R square correlation was found to be statistically significant (p < .001) (Figure 4).
    
    Figure 4: Relationship Between Number of Clusters and Development of Clusters in the City.
    
    City development and functions. Another objective was to analyze if the strategy used to construct the city has an effect on the functional composition of the clusters. Results showed that clusters belonging to cities constructed by cluster-simultaneitystrategy were structured by a significantly dominant urban function. On the contrary, clusters of cities developed by cluster-after-cluster strategy were constituted by mixed functions, with no dominant one (Figure 5). The Pearson R square correlation was statistically significant (p < .002) (Figure 6).
    
    c:\prsnfinal24/chapter 15
    
    366
    
    From Negations to Negotiations
    
    (a)
    
    (b)
    Figure 5 Example of two cities illustrating the strategy used during the design process, and the effect on the functional composition of clusters: 5a Homogeneous/continuous cluster development, characterized by the existence of mixed functions; 5b Heterogeneous/discontinuous cluster development, characterized by the existence of a dominant function.
    
    City Games as a Framework for Studying Spatial Information...
    
    367
    
    Figure 6 Relationship Between Cognitive Strategies used to Develop the city, and Cluster Functional Composition.
    
    City development and spatial distances between buildings. Finally, we investigated the relationship between the cognitive strategy applied by students during the design process, and the density of the city. Density was defined as distance between buildings sequentially located in the playing ground. Statistically significant differences were found between the group of students who constructed cities by locating buildings at short distances, and those locating buildings at large distances (t = -10.523; df = 25.209; p< .001) (Table 2, and Figures 7-8).
    Table 2 Sequential Distances Between Buildings During the Design Process
    
    Subjects Group 1 Group 2 16 14
    
    Mean Distance 6240 8796
    
    Std. Deviation 583.12 749.79
    
    c:\prsnfinal24/chapter 15
    
    368
    
    From Negations to Negotiations
    
    Figure 7 Distribution of Distances Between Buildings Sequentially Located in the Playing Ground.
    
    (a)
    
    (b)
    
    Figure 8 Example of two cities illustrating sequential distances between buildings: 8a close iterations (mean distance = 5619). 8b distant iterations (mean distance = 10217).
    
    City Games as a Framework for Studying Spatial Information...
    
    369
    
    A significant relationship was also found between the cognitive strategy used to develop the city, and the distances between iterations. While cities constructed by means of cluster-after-cluster strategy were correlated with short distance iterations, cities developed by the cluster-simultaneity strategy were correlated with long distance iterations. Statistical analysis showed the significance of Pearson R square for this relationship (p < .003) (Figure 9).
    
    Figure 9 Relationship Between Distances Between Buildings, and the Development of Clusters in the City.
    
    Conclusions The design of cities is an intricate process that is characterized by an interactive dialogue between the designer internal, and external representations (Goldschmidt, 1999; Schon, 1983; Goel, 1995). Due to its scale and complexity, this process is never fully internalized. The location on the playing ground of 1:100 mock-ups enabled to represent and visualize spatial relationships of the physical environment that otherwise could not be captured in their entirety by direct perception.
    
    c:\prsnfinal24/chapter 15
    
    370
    
    From Negations to Negotiations
    
    In complex design tasks, the ability to successfully deal with spatial knowledge, and to internalize it is related to the tools used for representation (Lawson, 1994; Rodgers et al, 2000). City Games allowed studying spatial representations of large-scale artifacts during the front edge stages of the design process. As an external memory, City games proved to be successful to overcome cognitive limitations in the processing and inter-representation of complex spatial information (See The magic number seven, Miller, 1956). A major contribution of City Games is that they enabled gaining an insight in the dynamic interplay of complex spatial representations in constant change and evolution. From empirical results we found that rather than developing cities through a random, continuous, or undifferentiated pattern of clusters, students tend to design the structure of their cities based on cluster organizations characterized by a high level of legibility. Students purposefully create spatial structures in the city according to clearly identifiable cluster formations. By analyzing the design process, we observed two main cognitive strategies used for the construction of cities. In the first strategy, students acted locally by means of a cluster-after-cluster cognitive strategy. They developed one group of related buildings after the other, while the general structure of the city constantly changed. When using this strategy, the spatial characteristic of the city resulted in a low number of large clusters, with mixed urban functions, and buildings located at short distances from each other. While design actions were limited to constructing an area of the city, the general spatial structure remained unclear until the final stages of the process. We propose that students using the cluster-after-cluster strategy had a weak mental representation of the city. As a result they were unable to foresee how the final city would look like. In contrast to the first strategy, the cluster-simultaneity cognitive strategy allowed students to act globally. Clusters evolved simultaneously, and in general no dramatic changes occurred in the overall structure of the city during the process. Cities constructed under this strategy were seen to have large numbers of small clusters
    
    City Games as a Framework for Studying Spatial Information...
    
    371
    
    characterized by a dominant urban function, located at larger distances from each other. After initial design iterations a recognizable urban order emerged from the playing ground. Students internalized the pattern of this emerging order, and were likely to locate the subsequent buildings to strengthen this order. It is suggested that participants who applied cluster-simultaneity strategy had a strong mental representation that enabled to: (i) exercise some control over the design process, and (ii) predict from the earlier stages of the process how the emerging structure of the city will look like. These phenomena support the self-organization found in Bartlett scenarios principle and suggested in the SIRN theory by Haken, and Portugali (1996) and by Portugali (1996b; 2002). Accordingly, after strong fluctuations emerge in the process of creating an object, the general structure of external representation becomes stable, and thus no substantial changes take place.
    
    Future Research
    In a future research City Games will be used to explore aspects related to the inter-personal spatial representations that have exceeded the scope of the present research. We will investigate the dynamics of self-organization when different designers interact simultaneously to construct a city over the same playground. A main focus will be set to assess how design actions are enslaved by decisions of others, and how these affect the spatial quality of the city.
    
    References
    Aitken, S.C., Cutter, S.L., Foote, K.E., and Sell, J.L. (1989) Environmental perception and behavioral geography. In G.L. Gaile, and C.J. Willmott (Eds.), Geography in America, Columbus: Merrill, pp. 218-238. Bartlett, F. C. (1932/1961) Remembering, Cambridge University Press, Cambridge.
    
    c:\prsnfinal24/chapter 15
    
    372
    
    From Negations to Negotiations
    
    Casakin, H. and Portugali, J. The design and evolution of cities as selforganizing systems, Proceedings of the 6th International Conference of the European Academy of Design, Design + system + evolution. University of the Arts, Bremen, Germany. March, 2005. Casakin, H. Representing spatial information through a collective map of the urban environment: implications for design, Proceedings of the Annual Meeting of the Israeli Geographical Association. Bar Ilan University, Ramat Gan, Israel. December, 2003. Foo, P., Warren, W.H., Duchon, A., and Tarr, M.J. (2005) Do humans integrate routes into a cognitive map? Map-versus landmark-based navigation of novel shortcuts. Journal of Experimental Psychology: Learning Memory and Cognition, vol. 31, no.2, pp.195-215. Goel, V. (1995) Sketches of Thought MIT Press, Cambridge, MA. Golledge, R.G. and Stimson, R.J. (1997) Spatial Behavior: A Geographic Perspective. The Guilford Press, New York Goldschmidt, G. (1991) The dialectics of sketching. Creativity Research Journal, vol. 4, no. 2 pp. 123-143. Haken, H. and Portugali, J. (1996) Synergetics, inter-representation networks and cognitive maps. In J. Portugali (Ed.), The Construction of Cognitive Maps Kluwer academic publishers, Dordrecht, pp. 45-67. Haken, H. and Portugali, J. (2003) The face of the city is its information. Journal of Environmental Psychology, vol. 23, no.4, pp.385-408. Halseth, G. and Doddridge, J. (2000) Children’s cognitive mapping: a potential tool for neighborhood planning. Environment and Planning B, vol. 27, pp. 565 – 582. Hatna, E., Casakin, H., and Portugali, J. Characterization and classification of cognitive maps: the use of city games, Proceedings of the Annual Meeting of the Israeli Geographical Association, Tel Aviv University, Tel Aviv, Israel. December, 2001. Jeffery, J. and Burgess, N. (2006) A metric for the cognitive map: found at last? Trends in Cognitive Sciences, vol. 3, pp. 1-3. Kaplan, S. (1983) A model of person environment compatibility. Environment
    
    City Games as a Framework for Studying Spatial Information... and Behavior, vol. 15, pp. 311-332.
    
    373
    
    Kaplan S. and Kaplan, R. (1982) Humanscape: environments for people. Ulrich’s Books, Michigan. Kitchin, R. M. (1994) Cognitive maps: what are they and why study them? Journal of Environmental Psychology, vol. 14, pp. 1-19. Konar, A. and Chakraborty, U.K. (2005) Reasoning and unsupervised learning in a fuzzy cognitive map. Information Sciences, vol. 170, no. 2-4, pp. 419-441. Hirtle, S. C. and Jonides, J. (1985) Evidence of hierarchies in cognitive maps. Memory and Cognition, vol. 13, pp. 208-217. Lawson, B. (1994) Design in Mind. Butterworth-Heinemann Ltd, London. Lynch, K, (1960) The image of the City, MIT Press, Cambridge, MA. Lowenthal, D. (1967) Environmental perception and behavior, University of Chicago, Chicago. Madinapour, A. (1996) Design of urban space. John Wiley & Sons, Chichester. Milgram, S. and Jodelet, D. (1976) Psychological maps of Paris. In H. Proshansky (Ed.), Environmental Psychology 2nd edition. Holt, Rinehart &Winston, New York. Miller, G. (1956) The magic number seven, plus or minus two: some limits on our capacity for processing information. The Psychological Review, vol. 63, pp. 81-97. Nassar, J. (1998) The evaluative image of the city. Sage Publications, London. Portugali, J. (Ed.) (1996a). The Construction of Cognitive Maps. Kluwer academic publishers, Dordrecht. Portugali, J. (1996b) Inter-representation networks and cognitive maps. In J. Portugali (Ed.), The Construction of Cognitive Maps, Kluwer academic publishers, Dordrecht, pp.11-43. Portugali, J. (1997) Self-organizing cities. Futures vol. 29, pp. 353-380. Portugali, J. (1999) Self-Organization and the City Springer, Heidelberg.
    
    c:\prsnfinal24/chapter 15
    
    374
    
    From Negations to Negotiations
    
    Portugali, J. (2002) The seven Basic Propositions of SIRN (Synergetic Inter-Representation Networks). Nonlinear Phenomena in Complex Systems, vol. 5, no 4, pp 428-444. Portugali, J. (2004) Towards a cognitive approach to urban dynamics. Environment and Planning B: Planning and Design, vol. 31, pp. 589-613. Portugali, J. and Casakin, H. SIRN (Synergetic Inter-Representation Networks): an approach to design, Proceedings of 2002 International Conference on Common Ground Design, London, U K, July, 2002, pp. 884-902. Portugali, J. and Haken, H. (1992) Synergetics and cognitive maps. In J. Portugali (Ed.) Geography, Environment and Cognition, Special Issue of Geoforum, vol. 23, no 2, pp. 111-130. Ramadier, T. and Moser, G. (1998) Social legibility, the cognitive map, and urban behavior. Journal of Environmental Psychology, vol. 18, pp. 307-319. Rodgers, P., Green, G., and McGown A. (2000) Using concept sketches to track design progress. Design Studies, vol. 21, no.5, pp. 451-464. Soini, K. (2001) Exploring human dimensions of multifunctional landscapes through mapping and map-making. Landscape and Urban Planning, vol. 57, no. 3-4, pp. 225-239. Smith, B. (1999) Agglomerations. In C. Freksa and D. M. Mark (Eds.), COSIT’99, Spatial Information Theory- Cognitive and Computational Foundations of Geographic Information Science, Berlin, Springer, pp.214-238. Schon, D. (1983) The Reflective Practitioner: How Professionals Think in Action, Basic Books, New York. Tolman, E. (1948) Cognitive maps in rats and men. Psychological Review, vol. 56, pp. 144-155.
    
    PART II

Add Comment

To add a comment, you need to login or signup (you can signup without leaving this page)