Abstract

The level of life quality can be closely tied to the quality of our surrounding environments. Though the psychological benefits of the natural environment have been exhaustively studied, we still have a limited understanding of the mechanism behind the built environment’s role in psychological well-being. Given that urbanization is an ongoing phenomenon, how to create the urban environment healthier and happier for people who spend time in the place should be understood. For this reason, this study is designed to understand the impact of the urban environment on people’s psychological well-being throughout a three-part empirical study. Before these studies, I proposed a theoretical framework that can explain the underlying mechanism behind the relationship between the built environment and psychological well-being by adopting two key concepts, walkability and third place.In the first empirical study, I delved into the role of third places for the psychological well-being of people by conducting a survey. This study found that third places can be psychologically restorative places and have stress-relieving effects. By serving as a resting place for contemporary people, third places were found to be the most popular resting place for them. Also, this study found that third places should be easily accessible, have enough space with chairs and tables, and provide openness for people to frequent the places. In the second study, I tested the impact of accessible (i.e., numbers of third places) and walkable neighborhood design on community-wide psychological well-being. This study measured psychological well-being by translating tweets into the level of mood and collected neighborhoods’ sociodemographic characteristics throughout the City of Los Angeles. Using multiple linear regression with ordinary least squares, this study assessed the impact of walkability and accessibility on community-wide psychological well-being. These research findings showed that walkability and accessibility can raise the level of psychological well-being of people. Also, the number of third places was more crucial for low-walkable communities. Lastly, the last study focused on providing an in-depth discussion on the applicability of prediction models and deep neural networks (DNN) in urban planning and policy to create healthy urban environments. To that end, this study developed two prediction models by using deep neural network (DNN): Binary mood classification model and Crime regression models. This study’s findings showed that DNN has a great potential in urban planning and policy to develop advanced prediction models using big data. However, this study also showed that prediction models can be more applicable when the output data is objective and concrete and can be explained by spatial patterns.