A spatial model of cognitive distance in cities

ABSTRACT

Spatial cognition is fundamental to the behaviour and activity of humans in urban space. Humans perceive their environments with systematic biases and errors, and act upon these perceptions, which in turn form urban patterns of activity. These perceptions are influenced by a multitude of factors, many of them relating to the static urban form. Yet much of geographic analysis ignores the influence of urban form, instead referring most commonly to the Euclidean arrangement of space. In this paper, we propose a novel spatial modelling framework for estimating cognitive distance in urban space. This framework is constructed from a wealth of research describing the effect of environmental factors on distance estimation, and produces a quantitative estimate of the effect based on standard GIS data. Unlike other cost measures, the cognitive distance estimate integrates systematically observed distortions and biases in spatial cognition. As a proof-of-concept, the framework is implemented for 26 cities worldwide using open data, producing a novel comparative measure of ‘cognitive accessibility’. The paper concludes with a discussion of the potential of this approach in analysing and modelling urban systems, and outlines areas for further research.

KEYWORDS:

• Spatial cognition
• GIScience
• cognitive geography
• cognitive distance
• accessibility

Acknowledgments

This work was partly supported by a Fellowship award from The Alan Turing Institute, United Kingdom.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data and code availability statement

The code supporting the findings of this study are available on Figshare at the link https://doi.org/10.5522/04/11777736. The data used in this study are freely available from OpenStreetMap, and can be accessed via methods described at https://wiki.openstreetmap.org/wiki/Downloading_data or through using the code accompanying this paper.

Notes

1. Described at https://elevation-api.io

2. Further details on classification: https://wiki.openstreetmap.org/wiki/Key:highway

3. Further details on classification: https://wiki.openstreetmap.org/wiki/Key:amenity

4. Calculated according to the shortest metric distance path.

Additional information

Notes on contributors

Ed Manley

Ed Manley is Professor of Urban Analytics in the School of Geography, University of Leeds, and Turing Fellow at the Alan Turing Institute for Data Science and Artificial Intelligence. He is a co-author of the book ‘Agent-based Modelling and Geographical Information Science’, published by Sage, and chairs the GIScience Research Group at the Royal Geographical Society. He received his EngD from University College London, UK in 2014, his MSc from University of Leeds, UK, and a BSc from University of East Anglia, UK.

Gabriele Filomena

Gabriele Filomena is a PhD candidate at the Institute for Geoinformatics, University of Münster, Germany. He is interested in studying how the interaction between people and the urban environment gives shape to and is mediated by cognitive representations of space. Before undertaking doctoral studies in 2017, he obtained a master’s degree in Cognitive Science at the University of Turin, Italy, and a MRes in Spatial Data Science and Visualisation at University College London, UK.

Panos Mavros

Panos Mavros is a researcher at the Future Cities Laboratory (FCL) of the Singapore-ETH Centre, where he is the Project Coordinator of the project Cognition, Perception and Behaviour in Urban Environments. Panos studied Architecture at the National Technical University of Athens, Greece and Digital Media at the University of Edinburgh, UK. He completed his PhD at The Bartlett Centre for Advanced Spatial Analysis (CASA) at University College London.