Location: Olsson Hall, Rm 104 and Zoom
Add to Calendar 2022-12-02T09:00:00 2022-12-02T11:00:00 America/New_York PhD defense: Xiang Guo All Invited Committee Members: T. Donna Chen (Chair), ESE Arsalan Heydarian (Advisor), ESE Sara Lu Riggs, ESE Jeffrey Woo, School of Data Science Nan Li, Civil Engineering, Tsinghua University Meeting Information: Olsson Hall, Rm 104 and Zoom

All Invited

Committee Members:

  • T. Donna Chen (Chair), ESE
  • Arsalan Heydarian (Advisor), ESE
  • Sara Lu Riggs, ESE
  • Jeffrey Woo, School of Data Science
  • Nan Li, Civil Engineering, Tsinghua University

Meeting Information:

Title: Evaluating Cyclist Behavior in Different Roadway Designs through Immersive Virtual Environments and Psychophysiological Sensing

Abstract:

As a healthier and more sustainable way of mobility, cycling has been advocated by literature and policy. However, current trends in cyclist crash fatalities suggest deficiencies in current roadway design in protecting these vulnerable road users. The lack of cycling data is a common challenge for studying cyclists' safety, behavior, and comfort levels under different design contexts. The integration of human sensing technologies has greatly facilitated the human-centered design evaluation of road infrastructures. Another question of interest is cyclists' distraction and involvement in secondary tasks. Understanding cyclists' behaviors under the influence of distraction can provide evidence for interventions to address safety related issues. This study will focus on cognitive distraction as it is related to the most frequently reported secondary task during cycling, such as listening to music or talking in the earphones.

To understand cyclists' behavior in different contextual settings, an Immersive Virtual Environment (IVE) bicycle simulator - Omni-Reality and Cognition Lab Simulator (ORCLSim) is developed with the ability to collect the following data: cycling performance (speed, steering, braking, acceleration and lane position), eye Tracking (gaze direction, fixation), physiological responses (Heart rate, head movement, hand acceleration), video recording and stated preferences surveys (subjective rating). With the ORCLSim system framework, this proposal aims to study the effect of different roadway designs (external factors) and psychophysiological states (internal factors) on cycling behavior with the goals of: (1). Capturing and analyzing cyclist behavior and psychophysiological responses within an immersive virtual environment. (2).Validate the use of IVE-based bike simulator with multimodal data sources for cyclist study. (3). Evaluate alternative designs for cyclists with human-centered and data-driven methods. (4). Study the effect of cognitive distraction on cyclist behavior.

Specifically, in addition to the system setup integration and development, three experimental studies are conducted: (1). Benchmark the use of IVE for cyclist behavioral study. Cyclist behavior and psychophysiological responses will be compared between the real world environment and IVE to validate what set of information in IVE is representative of the real world (n=6). (2). Evaluation of different design alternatives in the IVEs. Three different roadway designs will be evaluated in the ORCLSim system framework to test if the cyclists' physiological response is different from their stated preferences and actual cycling behaviors (n=50). (3). Study the effect of cognitive distraction on cyclist behavior. Two types of cognitive distractions are tested: listening to music and talking on the phone with earphones. Both standardized secondary tasks (Mock phone conversation task) and actual secondary task (music listening) will be introduced in the experiment to test if standardized secondary tasks can be applied to simulate cyclists' cognitive workload and the effect of cognitive distraction on cycling behaviors (n=75). 

The results indicate that: (1). Most of the performance measurements have absolute validity, the IVE bike simulator can be further utilized for understanding cyclists’ behaviors. (2). It is important to track physiological metrics to better understand how different settings may impact the cyclists. Additionally, we showcased the importance of gaze tracking and heart rate data in capturing behavioral response to different events or roadway settings. (3). Bicycle infrastructure can meaningfully impact cyclist’s movement and psychophysiological responses. The protected bike lane design has the highest subjective safety rating, lowest cycling speed and highest lateral distance to the vehicle lane, indicating the potential for safer bicycling behavior with lower speeds and increased separation from vehicles; cyclists focus their gaze on the cycling task more in the separate and protected bike lane scenarios; creating separation zones for bicyclists (whether separate bike lane or protected bike lane) has the potential to reduce the stress level, as indicated by decreased HR changes compared to the shared bike lane. (4). Differences are found in cyclists' adaptive behaviors with different types of cognitive distractions. Talking on the phone is rated as the most distracting scenario, cyclists would keep a lower speed with less input power and less head movement variation. While listening to music, cyclists would have a significantly higher speed, lower standard deviation of speed, and higher input power.The introduction of bike lanes has the potential to stabilize the lateral lane position. Demographic information is also found to affect the behavioral and psychophysiological responses.