Human Way-Finding in an Urban Environment: A Performance Analysis of a Computational Process Model

In this paper the characteristics and performance are described of a computational process model (CPM) of human way-finding that is based on psychological models of cognition and experimental data on human way-finding. The model comprises two modules, one for representing objectively a suburban environment and the other for representing the cognitive processes involved in navigation. The CPM is employed to simulate and investigate how spatial knowledge that has been acquired and stored is retrieved and processed in order to plan a path to a given goal. The results of the experiments carried out with NAVIGATOR are compared with the results of experiments involving young adults in real environments. We analyze the performance of NAVIGATOR in terms of its ability to find the goal, the amount of time taken to reach a given goal, the differences in performance during way-finding, and the errors in navigation, as a function of several parameters. The parameters are intended to represent key psychological components of human perceptual and cognitive systems and are derived from experimental studies of human way-finding behavior. The main value of this model lies in its use as a device for investigating and understanding human spatial performance, in terms of accurate performance and errors that arise in tasks of way-finding, and the variations among individuals in terms of performance. A set of computational experiments are used to examine the differences in search strategies between ‘individuals’. The effects of various parameters on way-finding performance are analyzed. Errors in way-finding are classified and are compared with Norman's and Reason's classification of errors in human performance. The CPM appears to provide a fruitful model for investigating human way-finding behavior.