Linking cognition, spatial syntax and network use - a bivariate ordered probit for link memory fluency and use frequency

Abstract

Understanding the role of spatial cognition and mental maps in route choice behavior is critical for efficient and behaviorally rigorous transport simulation models. We investigate the relationship between spatial network characteristics, collective memory, and link consideration in navigation. Using a survey involving spatial recall, spatial memory strategies, and navigation tasks, we quantify link use fluency and collective street memory. Network data and spatial syntax indices (centralities) are incorporated to analyze their association with recalled link frequencies and link use in route choice. Results reveal that collective spatial memory is related to road hierarchy, points of interest, and spatial syntax indices, while link consideration depends on collective street memory, shortest path consideration, and navigation landmarks. These findings validate the hypothesis that mental maps mediate between urban networks and route choice, and the usefulness of leveraging spatial syntax to generate link availability weights in choice set generation and recursive route choice models.