Authors

You Li, Ruiyu Yan, Zhicheng Gao

Research Question

How do environmental conditions during subway commutes, specifically noise, air quality, temperature, and crowding, affect real-time physiological stress levels in commuters? While the previous work on neighborhood-level conditions' impact on stress revealed important associations, it also underscored a critical gap: the absence of real-time, data-driven understanding of how commuting environments contribute to stress. This year, we shift our focus to the NYC subway system, a vital yet often overlooked setting for daily urban stress. Our goal is to generate actionable insights to inform public health-oriented transit design and policies that promote mental well-being in dense urban settings.

Background

The E²HW Lab adopts a multidisciplinary, data-driven approach to explore the complex interplay between human activity and the environment, with a focus on its implications for public health. Through the use of geospatial analysis and data science, the lab investigates how environmental factors influence health outcomes, aiming to generate evidence-based solutions that support the creation of equitable and health-promoting environments. By integrating diverse areas of expertise, the E²HW Lab is committed to advancing health equity and driving meaningful change in the environments that shape individual and community well-being.

This project builds on a 2025 capstone project on urban stress, which received CUSP's Capstone of the Year award. That project highlighted the significant role of urban environments on stress among residents in New York, NY and Accra, Ghana. This year's project focuses on a significant yet understudied role of commuting environments in shaping real-time stress levels, focusing specifically on the NYC subway system - a daily reality for millions of urban residents.

Methodology

Leveraging wearable technology and environmental sensors, the project aims to investigate how subway commuting conditions affect individual-level physiological stress responses. Commuters are equipped with Empatica EmbracePlus watches to collect real-time physiological stress data (e.g., heart rate variability, electrodermal activity). These readings are paired with data from air quality monitors, noise sensors, temperature sensors, and wearable cameras used to assess crowding during commutes.

After the data is collected, machine learning techniques are applied to analyze the relationships between environmental stressors, such as noise, particulate matter, temperature, and crowd density, and physiological indicators of human stress. These approaches enable individual variability to be modeled and key patterns identified in how commuting conditions impact mental well-being.

The ultimate goal is to provide actionable, data-driven insights that can inform urban transportation policy and design, contributing to healthier, more equitable commuting environments in cities like New York.

Deliverables

• Analytical Report summarizing key findings on the relationship between subway commuting environments and physiological stress, using data collected from wearable sensors
• Data Visualizations and Interactive Dashboard showcasing stress patterns across different subway lines, environmental exposures, and commute conditions
• Technical Appendix or Code Repository containing data processing scripts, modeling workflows (including models), and documentation for reproducibility
• Draft of the manuscript that will later be submitted to the academic journal

Data Sources

This project relies on primary data collected by the capstone team through a field study involving subway commuters in NYC. Participants are equipped with the following wearable devices:

• Empatica EmbracePlus smartwatches to collect physiological stress data, including heart rate variability, electrodermal activity, skin temperature, and motion
• Portable air quality monitors to capture real-time levels of pollutants such as PM2.5 and CO2
• Noise sensors to measure ambient sound exposure during commutes
• Wearable cameras to visually assess crowding and density in subway cars and stations