With well-being at the core of travel experience in hybrid mobility environments, this talk will present findings on how prosocial behaviors in interdependent traffic scenarios influence rider well-being, cognition, and behavior. The results demonstrate that prosocial behaviors enhance cognitive and affective well-being and maximize payoffs in terms of time and speed management, while asocial behaviors lead to increased emotional strain and inefficiency. These insights highlight the need to incorporate prosocial dynamics into user-centered transportation systems to foster safer and more efficient mobility experiences.
SooYeon Kim is a Ph.D. student in the Industrial and Systems Engineering Department at the University of Wisconsin-Madison. Her research primarily centers on Human-Automation Interaction, focusing on modeling driver behavior, emotions, and cognition to enhance driver experience and safety, with the aim of proposing user-centered transportation systems.
Since the last decade, the flexibility provided by micro-mobility has led to its increased usage in urban areas. Automated systems with artificial intelligence (AI) may be integrated into e-scooter systems to reduce human workload and improve usability. However, the impact of varying levels of individual skill and automation on user accessibility and subjective well-being in shared e-scooter systems remains unclear. The goal of this study is to evaluate two main aspects of accessibility—individual skill level and automation level—in shared e-scooters through subjective well-being (SWB). A national online survey was conducted through Qualtrics™, collecting 400 responses. Participants rated four control-sharing scenarios (Baseline, Human-led, System-led, Balanced) with SWB-related measurements. Results show that the scenario with higher individual skill levels received higher cognitive and affective SWB on most measures, and the scenario with higher automation levels received higher cognitive SWB on only one measure. These findings provide an initial understanding of the expectations for shared control between humans and automated systems in micro-mobility and will guide the development of next-generation micro-mobility solutions.
Dr. Gaojian Huang is an Assistant Professor in the Department of Industrial and Systems Engineering at San Jose State University and a Research Associate at the Mineta Transportation Institute. He is also the director of the Behavior, Accessibility, and Technology (BAT) Lab. He earned master's degrees in Safety Management from Indiana University Bloomington in 2016 and Cognitive Psychology from Purdue University in 2020, and obtained his Ph.D. in Industrial Engineering from Purdue University in 2021. His research focuses on human behavior and performance modeling, and technology accessibility in complex systems, with an emphasis on aging, surface transportation, and multimodal human-machine interfaces.
Harmonious traffic interactions will become more important for on-road satisfaction and experiences, as safety and efficiency become more established. Prosocial behaviors are a critical aspect of the future hybrid society, thus the detection and modeling of these behaviors serve as fundamental stepstones for the intelligent vehicle services towards harmonious interactions. In this talk, we introduce the ongoing progress to detect, classify, and predict prosocial behaviors and intentions. We demonstrate the user study designs, in both simulator study and naturalistic traffic data collection, where we observe participants conduct and ignore prosocial behaviors. We collected the traffic scenes, vehicle information, as well as multimodal human response data. These data include eye gaze, semantic segmentation, and peripheral physiological data. The machine learning models demonstrate accurate detection of prosocial behaviors, and their degrees of satisfaction, effort, and obligations. We also investigate deep learning methods to predict prosocial intentions, to locate more readily helpers. The model demonstrates promising performance. With these advances, we picture future intelligent vehicles to understand human intentions and act accordingly, while encouraging and reinforcing harmonious behaviors at the same time.
Dr. Zhaobo Zheng received his Ph.D. in mechanical engineering from Vanderbilt University in 2021. He received his B.Sc. also in mechanical engineering from Xi'an Jiaotong University in 2015. Now he is a senior scientist at Honda Research Institute USA, Inc, in San Jose, California. His research interests include human computer interaction (HCI), multimodal signal processing, machine learning and deep learning with applications in intelligent vehicles and educational devices.
Introducing automated vehicles (AVs) on roads may challenge es¬tablished norms as drivers of human-driven vehicles (HVs) interact with AVs. Our study explored drivers’ decisions in game-theoretical scenarios amid mixed traffic using an online survey study. We ma¬nipulated factors including interaction types (HV-HV vs. HV-AV), scenario types (chicken game vs. public goods game), vehicle driv¬ing styles (aggressive vs. conservative), and time constraints (high vs. low). The quantitative results showed that human drivers tended to "defect" more, that is, not cooperate, against vehicles with con¬servative driving styles. The effect of vehicle driving styles was pronounced when interacting with AVs and in chicken game sce¬narios. Drivers exhibited more "defection" in public goods game scenarios and the effect of scenario types was weakened under high time constraints. Only drivers with moderate driving styles "defected" more in HV-AV interaction. Our qualitative findings pro¬vide essential insights into how drivers perceived conditions and formulated strategies for decision-making.
Na Du is an Assistant Professor in the Department of Informatics and Networked Systems at the School of Computing and Information, University of Pittsburgh. Her research focuses on enhancing human performance, safety, and well-being by integrating human factors and data analytics in the analysis, design, and evaluation of technology systems. Her key areas of interest include human factors in smart cities, computational modeling of human behavior, and human-centered design.
This work explores the potential of robots in emotion regulation by examining how different regulation types and empathic responses affect human emotions, decision-making, and trust in various emotional contexts. Human-subject experiments were conducted to evaluate the effects of emotion regulation across emotional scenarios. This work aims to inform the design of emotion regulation modules in robots, contributing to more effective human-robot interactions and advancing the development of emotionally intelligent social robots capable of supporting users in real-time across diverse emotional experiences.
Jackie Ayoub is a Human Factors Researcher and Data Scientist at Honda Research Institute's Ann Arbor office. She holds a Ph.D. in Human Factors, with a research focus on emotion, emotion regulation, and fostering effective interactions between humans and robots to build strong, collaborative teams. Her work integrates human factors and data science to enhance human-robot teaming and improve overall performance and well-being in technological systems.
This research explores the social dimensions of human-robot interactions, focusing on prosocial behaviors exhibited by both humans and autonomous agents. Building on prior work that primarily examined direct reciprocity using economic games, this study introduces a spatial environment where prosocial interactions are a secondary task. Participants either interact with a human (a replay of a previous participant) or a robot across multiple rounds, each featuring a single prosocial opportunity. The results indicate that participants pass on prosocial behavior equally to humans regardless of whether they were helped by a human or a robot. However, participants are less likely to pass on prosocial behavior to robots. Over time, there is an increase in overall prosocial behavior. Future research aims to explore prosocial dynamics within human groups and investigate tipping-point dynamics, where coordinated robotic behavior could influence broader population behavior.
Xinyue Hu is a second-year PhD student in Cognitive Science at the University of California, Irvine, working under the supervision of Dr. Mark Steyvers. She earned her bachelor's degree from Vassar College and her MS degree from the University of Copenhagen, both in Cognitive Science. Her research focuses on multi-agent interactions between human and machine intelligence.
Investigating methods to improve prosocial behavior has been a recent topic of interest for researchers studying human-automation and human-robot interaction. However, scientists have yet to uncover why certain features of technology increase prosocial behavior in humans. While researchers assume, based on the Media Equation (Reeves & Nass, 1996), that social influences in prosocial behavior remain consistent across domains, I will discuss differences between the human-automation (H-A) domain and the human-human (H-H) context. I also present a novel design framework and model that examines why specific characteristics in agents promote prosocial behavior in human users. The framework, Robots and Agents as Persuasive Prosocial Actors (RAPPA), combines principles from persuasive social computing, the Robot Social Influence model (Erel et al., 2024), and findings from the limited work within H-A prosocial behavior research. The theoretical model argues that anthropomorphism, social intelligence, and adaptiveness increase a human’s relatability, or sense of belonging, to the technology, which strengthens the agent’s influence on a human. This social influence can then persuade humans to behave prosocially. The framework is then connected back to theories of human behavior such as the theory of planned behavior (Ajzen, 1991) and social learning theory (Bandura, 1971). RAPPA serves to enhance the understanding of how H-A prosocial behavior develops and provide scientists with a valuable reference for future work. I will discuss a series of applications and research topics, including for my dissertation, that encourage researchers the use and study of the framework in in-vehicle agents.
Sidney Scott-Sharoni is a fifth-year PhD student in the Sonification Lab under Professor Bruce Walker at Georgia Tech. She holds a BS in psychology from Old Dominion University and a MS in psychology from Georgia Tech. Her research interests include prosocial behavior in human-agent interactions, customization of interfaces in automated vehicles, trust in automation, and information presentation in UIs.