Understanding pedestrian behavioural adaptation to urban thermal environments is critically important to attain this goal. Combining well-controlled experiments with human participants and computational methods inspired by behavioural ecology and decision theory, we examine the effect of sun exposure on route choice in a tropical city.Д¬urrent research in pedestrian behaviour lacks controlled experimentation, which limits the quantitative modelling of such complex behaviour.

We find that the distance walked in the shade is discounted by a factor of 0.86 compared to the distance walked in the sun, and that shadows cast by buildings have a stronger effect than trees. The discounting effect is mathematically formalised and thus allows quantification of the behaviour that can be used in understanding pedestrian behaviour in changing urban climates. The results highlight the importance of assessment of climate through human responses to it and point the way forward to explore scenarios to mitigate pedestrian heat stress. The global process of climate change poses a significant threat to urban populations. The population growth, happening mostly in the cities 1-the areas mainly contributing to the climate change 2 and strongly affected by higher temperatures 3-results into an increasing number of people being exposed to excessive heat. This in turn challenges many aspects of modern society: public health 4, 5, human- 6 and economic development 7, mental health 8 and social relations 9. It is therefore critically important to understand the ways to improve urban thermal environments and the human response to these improvements.

Modelling urban climate at the pedestrian level allows to predict the thermal environment 10, 11, 12, 13. Thermal perception and acceptance studies conducted through surveys throughout the world 14, 15, 16, 17 connect the microclimate and comfort of the people in it. These developments allow to evaluate and introduce the design and planning measures to improve pedestrian thermal environments through green 18, 19, 20 and built 21, 22 shading infrastructure, orientation of buildings 23 as well as smart path planning 24, 25. Yet, to evaluate the implication of these design and planning solutions for urban residents, it is necessary to gain a quantitative understanding of pedestrian behavioural responses to varying urban microclimates.ДЄttendance of urban areas and occupation of sun and shade 26, 27, duration and intensity of activities in urban parks 28 and preference for a sun-lit side of a street 29 have all been found to have an association with climate. While these rather qualitative observations confirm the presence of a pronounced human behavioural response to thermal environments, they do not provide insights into mechanisms underlying such behaviour, which are crucial for development of the quantitative model of pedestrian behavioural response to urban thermal environments. Research on crowd dynamics 30, 31 and human navigation in urban environments 32 provides mechanistic models of pedestrian motion, which are able to reproduce the aggregate dynamics or distribution of pedestrian flows in the environment.

These proposed mechanisms, however, remain hypothetical, requiring formal testing of them through controlled experimentation and quantitative characterisation of pedestrian choices in natural environments. Current literature lacks such studies, owing primarily to complexity of the problem of pedestrian decision making in complex urban environments.Д¬hoice modelling has been successfully employed to uncover the decision processes in many domains-from finance and behavioural economics to transportation, marketing, food preferences and animal behaviour 33, 34, 35, 36.

Bus simulator 18 sun visor control