Linda Ng Boyle

Generative artificial intelligence (AI) tools are rapidly reshaping how academic research is conceived, produced, and evaluated, raising questions regarding authorship, peer review, graduate training, and the preservation of scholarly skills. The human factors and ergonomics (HFE) community, with its expertise in automation reliance, trust, and human-centered system design, is uniquely positioned to examine this disruption. This panel brings together HFE academics spanning career stages and perspectives on AI to discuss the trade-offs of integrating AI across the academic research cycle. Moderated discussion will address the boundaries of authorship in iterative human-AI collaboration, implications for the peer review ecosystem, risks of cognitive atrophy and deskilling, and adaptations needed in doctoral training. Attendees will gain a richer vocabulary for navigating the dilemma that now defines academia: publish or perish, with or without AI.

While the design of autonomous vehicle systems often centers on efficiency and reliability, user experience and trust play a critical role in adoption and long-term use. This paper presents a foundational step in that direction by introducing an Interactive Reinforcement Learning algorithm, termed Trust-Aware Reward Shaping, or TARS in short, to incorporate trust dynamics into vehicle automation design. The TARS algorithm leverages real-time trust as an additional reward signal during training of the automated trajectory generation system, with lane change as the use case. It, therefore, enables customization based on the user-preferred driving style (aggressive, neutral, or conservative) and trust dynamics (vigilant, confident, or skeptical). Consequently, the algorithm successfully generates emergent behavior that dynamically responds to trust changes, such as a more aggressive style when users show over-trust tendency, and a more neutral style when users demonstrate under-trust or skeptical behaviors. It also predicts real-world driving behaviors observed in naturalistic driving data, and outperforms non-customized, expert-coded, and game-play trust-aware RL methods, as well as baseline inverse RL methods, especially for aggressive and neutral driving styles.

There are many conversations among researchers associated with the limited utility of the SAE Levels of Automation (LoAs) for defining behavioral-related research questions. The goal of this workshop is to examine these issues and to present a complementary framework to describe the interactions between people and automation, to account for the differences in capabilities between the human operator and the autonomous system, and to match the user’s expectation to system capabilities. This workshop will identify considerations for research and design that is neglected in the SAE framework. It will help address the interaction between human and automation in a more adequate way for human-centered designs.

In the US, light- to heavy-duty vehicles account for the largest portion of total greenhouse gas emissions. A rapid transition to electric vehicles (EVs) is critical in achieving the Net Zero goal of cutting carbon emissions. Approximately 11% of new car sales are EVs but adoption of this technology is still slow. The panel will present some of the EV adoption challenges, including charging infrastructure, charging behavior, changes in driving behavior and risk outcomes, crashworthiness, and other unintended consequences. Different research methods are used to address these challenges. The benefits of the approaches and findings for EV infrastructure stakeholders, regulators and policymakers, and researchers will be discussed. Lastly, the panel will review existing research gaps and emerging topics that still need to be addressed in order to increase EVs adoption while also supporting EVs’ safe operation.

Future technological innovation will continue to transform the future of human-system interactions (HSI). While immersive sensors, smarter devices, and integrated systems will drive innovations in HSI, the fundamental elements of human information processing must still be considered to ensure the appropriate use of the technology in the context that the human operator will be using the technology. BOHSI has organized a panel that brings together prominent researchers to discuss emerging trends in technology and the implication for future research needs in domains where increasingly autonomous systems are becoming more accessible, but supervisory control by the human is still needed. The implications for the future of work and the safety and security of systems that are complex and continually evolving are also discussed).

This study examined multitasking behaviors of drivers in environments that include large numbers of pedestrians and cyclists, using video and vehicle data from the second Strategic Highway Research Program (SHRP2). The study includes 15 sites in both Seattle, WA, and Tampa, FL, U.S., (nine pedestrian and six cyclist locations), including three marking and signal types for crosswalks and two types for bike treatments. A total of 1,458 SHRP2 traversals with time-series data and forward videos were extracted with face/dash videos for about 50% of these traversals. Forward video coding was conducted for all daytime traversals starting from one block before to one block after the selected site. Face/dash video was coded for all traversals with pedestrians or cyclists. A matched set of traversals without pedestrians or cyclists were also coded. The final data set included 458 traversals with coded data on multitasking behavior and the multimodal environment. Mixed-effect binary logistic regression models were used to examine the associations of pedestrian/cyclist presence and the facility type with drivers’ multitasking behavior. The findings show that the presence of pedestrians/cyclists and facility types could be related to drivers’ multitasking behavior. The findings can provide the foundation for future studies that examine safety for non-motorists with respect to infrastructure design, signage, and policies. There is also the potential to provide insights into assistive driving systems within automated vehicles, which are discussed in this paper.

Vehicle automation seeks to enhance road safety and improve the driving experience. However, a standard system does not account for variations in users and driving conditions. Customizing vehicle automation based on users’ preferences aims to improve the user experience and adoption of the technologies. This study introduces a systematic paradigm that starts with naturalistic driving data to identify the driving behaviors and styles for a customized automated lane change system. The driving behaviors are first extracted using Multivariate Functional Principal Component Analysis (MFPCA) with minimum prior expert knowledge. The driving styles are identified by clustering the extracted driving behaviors. An Inverse Reinforcement Learning (IRL) algorithm is then used to train the automated lane change system from grouped demonstrations of the identified driving styles to capture the preferences of a group of drivers with a similar driving style. The performance of the proposed customized automated lane change system is compared to (1) a non-customized system trained on all the sample trips, (2) customized systems built on expert-coded reward functions, and (3) customized systems trained using a Generative Adversarial Imitation Learning (GAIL) algorithm. The results show that our method outperforms all the other systems with respect to the prediction accuracy of the lane change actions. Additionally, our method gains insights on the representative behaviors of different driving styles to enable customization of automated lane change systems.