A tool to predict perceived urban stress in open public spaces

By Professor Martin Knöll, Department of Architecture, Technische Universität Darmstadt, Germany

I love people-watching. Most urban designers do. The more people in a public space, the better. The more diverse, active, extroverted, connected, playful, and affectionate people are with each other, the more there is to see and do. In this sense, the more insane a crowd is, the more livable and healthy is a city. My research on people-centered urban design does not aim to eliminate insanity from everyday life in cities. I rather seek to better understand which factors of the urban environment are stressful to pedestrians and particularly vulnerable people and how urban design can better support them in using public space.

Figure 1 shows participants rating environmental properties in open public spaces (OPS) using a smartphone app (Halblaub Miranda, Hardy, & Knöll, 2015). Photo credit: Marianne Halblaub Miranda.

Current city living has been related to various manifestations of stress and a higher risk for mental health problems (Lederbogen, et al., 2011). Lederbogen et al. (2013) have named a set of influencing factors for urban social stress including infrastructure, socio-economic factors, noise and environmental pollution. These remain open questions:

• Which factors of the built environment are critical in influencing subjectively perceived urban stress (PUS)
• How do these factors interact? 
• How can these factors be addressed by planning and urban design to maximize pedestrian comfort?

 
This blog entry reports on a few findings in two recent publications, in which my colleagues and I from TU Darmstadt have introduced a framework of environmental factors and spatial analysis tools shown to be useful to describe and even predict PUS in open public spaces (OPS). In a first step, environmental properties have been constructed for a sample of OPS in the city of Darmstadt, Germany, using the space syntax framework (Hillier & Hanson, 1984). These were paired to users’ ratings of spatial qualities such as loudness and subjectively perceived safety and stress (figure 2). Isovist vertice density has been shown to be weakly associated to users’ ratings of safety (r=.365, p=.09, Pearson), while global and citywide integration of a street segment have been shown closely related to PUS (r=0.432, p=0.04, Pearson) (Knöll, Neuheuser, Li, & Rudolph-Cleff, 2015). City-wide integration of streets has been closely related to actual amount of car traffic. Our findings may underline the importance of traffic calming, speed limits and walkability for measures to reduce pedestrians’ perceived stress levels.

Figure 2 shows an isovist with a high vertices density (vertex number / isovist area2) is shown, which is weakly associated to users' ratings of an OPS as "safe". The map of Darmstadt on the right shows global integration (r=n) values of its street segments (red indicates high global integration). They are significantly related to ratings of OPS as “max. stressful“ (Knöll, Neuheuser, Li, & Rudolph-Cleff, 2015).

In a recent journal article, the data has been analyzed using different types of multivariate models with the aim to predict ratings of PUS with a highly explained variance and significance (Knöll, Neuheuser, Cleff, & Rudolph-Cleff, 2017). Open space typologies (park, square, courtyard, streets) were found to be the best predictors for PUS, followed by building coverage ratio, isovist vertices numbers and syntactical characteristics (Figure 3). The isovist characteristics in particular, revealed interesting new insights and research questions. For example, visibility, as the relative size of the area that can be overseen from a given point in a space, was found positively related to users ratings of perceived stress. This was contrary to findings previously reported in indoor spaces. And it is somewhat surprising, since visibility of pedestrians by car drivers is key to reduce traffic injuries and improve overall actual pedestrian safety. In other words, people seem to feel most stressed in those areas of busy squares and streets where, in theory, they should be safest from car traffic. We conclude that to further study the isovist characteristics and their relation to percieved stress should be a priority in future research. Visibility and the geometric shape (“complexity”) of urban environments are factors that can be influenced by urban design measures such as street furniture, trees and facades (Knöll, Neuheuser, Cleff, & Rudolph-Cleff, 2017).

Figure 3 lists environmental factors found related to ratings of perceived stress in open public spaces (Knöll, Neuheuser, Cleff, & Rudolph-Cleff, 2017).

A model has been presented that uses a combination of environmental properties and achieves a predictive power of R2=54.6% (Knöll, Neuheuser, Cleff, & Rudolph-Cleff, 2017). These results are a first attempt to predict more complex emotions such as perceived urban stress by analyzing factors of the built environment and using standard planning tools such as GIS and Space Syntax. They extend existing models that have predicted tranquility in green spaces (Watts, Pheasant, & Horoshenko, 2014) or activities and spatial experience in streetscapes (Bielik, Schneider, Kuliga, Valasek, & Donath, 2015). The framework may be useful to architects and neuroscientists alike, who seek to identify or visualize urban configurations likely to be perceived as stressful and seek to further investigate pedestrian comfort by pairing environmental factors with geo referenced, psychophysiological effects.

References

Bielik, M., Schneider, S., Kuliga, S., Valasek, M., & Donath, D. (2015). Investigating the effect of urban form on the environmental appraisal of streetscapes. In K. Karimi, L. Vaughan, K. Sailer, G. Palaiologou, & T. Bolton (Ed.), Proceedings of the 10th International Space Syntax Symposium (pp. 119:1-13). London: Space Syntax Laboratory, The Bartlett School of Architecture, University College London.

Franz, G., & Wiener, J. M. (2008, April 3). From space syntax to space semantics: a behaviourally and perceptually oriented methodology for the efficient description of the geometry and the topology of environments. Environment and Planning B: Planning and Design , XXXV, pp. 574-92.

Halblaub Miranda, M., Hardy, S., & Knöll, M. (2015). MoMe: a context-sensitive mobile application to research spatial perception and behaviour. In Department of Geosciences and Natural Resource Management (ed.), Human mobility, cognition and GISc. Conference proceedings. November 2015, (p. 29-30). Copenhagen.

Hillier, B., & Hanson, J. (1984). The Social Logic of Space. Cambridge: University Press.

Knöll, M., Neuheuser, K., Cleff, T., & Rudolph-Cleff, A. (2017). A tool to predict perceived urban stress in open public spaces. Environment and Planning B: Urban Analytics and City Science.

Knöll, Martin, Environmental factors and tools to analyze perceived stress in open spaces. ANFA 2016: CONNECTIONS – BRIDGESYNAPSES. 23rd-24th September 2016, La Jolla, CA: Academy of Neuroscience for Architecture. Video

Knöll, M., Neuheuser, K., Li, Y., & Rudolph-Cleff, A. (2015). Using space syntax to analyze stress perception in open public space. In K. Karimi, L.  Vaughan, K. Sailer, G. Palaiologou, & T. Bolton (Ed.), Proceedings of the 10th International Space Syntax Symposium (pp. 123:1-15). London: Space Syntax Laboratory, The Bartlett School of Architecture, University College London.

Lederbogen, F., Haddad, L., & Meyer-Lindenberg, A. (2013, December). Urban social stress - Risk factor for mental disorders. The case of schizophrenia. Environmental Pollution .

Lederbogen, F., Kirsch, P., Haddad, L., Streit, F., Tost, H., Schuch, P., et al. (2011, June 23). City living and urban upbringing affect neural social stress processing in humans. Nature , 474, pp. 498-501.

Watts, G., Pheasant, R., & Horoshenko, K. (2014). Predicting perceived tranquillity in urban parks and open spaces. Environment and Planning B: Planning and Design , 38 (4), pp. 585-94.
 
Acknowledgments

I am grateful to my co-authors Dr. Annette Rudolph-Cleff, Professor of Design and Urban Development, Yang Li, PhD Candidate, both Department of Architecture, Katrin Neuheuser, PhD Candidate, Dept. of Human Sciences, TU Darmstadt, and Dr. Thomas Cleff, Professor of Quantitative Methods for Business and Economics at Pforzheim University, Germany.

About the Author

Martin Knöll is Junior Professor and head of the Urban Health Games research group in the Department of Architecture at Technische Universität Darmstadt. He is a registered architect and holds a PhD from Universität Stuttgart, Germany, on health-oriented urbanism. Dr. Knöll leads the research group in building new transdisciplinary collaborations between urban designers and health experts to address global health challenges such as social inclusion, physically inactive lifestyles, diabetes and urban stress. Previously, Dr. Knöll was visiting scholar at the Lansdown Centre for Electronic Arts at Middlesex University, London, UK, and held a post-doc position in the Department of Architecture at Karlsruhe Institute of Technology, Germany.

Contact @MartinKnoell