Calibrating Pedestrians' Trust in Automated Vehicles

Faas, Stefanie M.; Kraus, Johannes; Schoenhals, Alexander; Baumann, Martin

doi:10.1145/3411764.3445738

Cited by 44 publications

(22 citation statements)

References 74 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Distrust describes an insufficient level of trust in a system, leading to non-usage and, in turn, a loss of the advantages of the technical system ( Lee and See, 2004 ). In the context of eHMIs, distrust in the system would lead to pedestrians’ reduced willingness to use the provided information by eHMI signals ( Faas et al, 2021 ). Whereas, overtrust would result if the users’ trust exceeds the system’s capabilities.…”

Section: Introductionmentioning

confidence: 99%

“…The users’ overtrust in a system, as an attitude, leads to overreliance in the system’s capabilities as a behavioral aspect ( Lee and See, 2004 ). With regard to eHMIs in AVs, overtrust implies an overreliance in the eHMI signals that could lead to insufficient considerations of implicit communication signals that are provided by the vehicle’s driving behavior ( Faas et al, 2021 ). Hence, overtrust should be respected as an essential safety issue in eHMIs ( Tabone et al, 2021 ).…”

Section: Introductionmentioning

confidence: 99%

“…Previous studies that investigated eHMI malfunctions, realized the malfunctions by contradicting information of the provided eHMI signals and the vehicles’ driving behavior as implicit communication signals. First research results indicated a decline of participants’ trust after encountering invalid eHMI functions in street crossing scenarios ( Kaleefathullah et al, 2020 ; Faas et al, 2021 ). Faas et al (2021) reported that participants’ trust declined temporarily after experiencing a single eHMI malfunction.…”

Section: Introductionmentioning

confidence: 99%

“…First research results indicated a decline of participants’ trust after encountering invalid eHMI functions in street crossing scenarios ( Kaleefathullah et al, 2020 ; Faas et al, 2021 ). Faas et al (2021) reported that participants’ trust declined temporarily after experiencing a single eHMI malfunction. Therefore, the authors concluded that participants’ trust formation in eHMIs can be seen as a dynamic process that is based on previous experience during encounters with the system ( Faas et al, 2021 ).…”

Section: Introductionmentioning

confidence: 99%

“… Faas et al (2021) reported that participants’ trust declined temporarily after experiencing a single eHMI malfunction. Therefore, the authors concluded that participants’ trust formation in eHMIs can be seen as a dynamic process that is based on previous experience during encounters with the system ( Faas et al, 2021 ). In order to prevent potentially safety critical situations and the users’ overtrust, the effect of multiple eHMI malfunctions was investigated in the current study in a shared space setting, comprising ambiguous encounters between the involved traffic participants.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

The Effect of eHMI Malfunctions on Younger and Elderly Pedestrians’ Trust and Acceptance of Automated Vehicle Communication Signals

et al. 2022

View full text Add to dashboard Cite

To ensure traffic flow and road safety in automated driving, external human–machine interfaces (eHMIs) could prospectively support the interaction between automated vehicles (AVs; SAE Level 3 or higher) and pedestrians if implicit communication is insufficient. Particularly elderly pedestrians (≥65 years) who are notably vulnerable in terms of traffic safety might benefit of the advantages of additional signals provided by eHMIs. Previous research showed that eHMIs were assessed as useful means of communication in AVs and were preferred over exclusively implicit communication signals. However, the attitudes of elderly users regarding technology usage and acceptance are ambiguous (i.e., less intention to use technology vs. a tendency toward overreliance on technology compared to younger users). Considering potential eHMI malfunctions, an appropriate level of trust in eHMIs is required to ensure traffic safety. So far, little research respected the impact of multiple eHMI malfunctions on participants’ assessment of the system. Moreover, age effects were rarely investigated in eHMIs. In the current monitor-based study, N = 36 participants (19 younger, 17 elderly) repeatedly assessed an eHMI: During an initial measurement, when encountering a valid system and after experiencing eHMI malfunctions. Participants indicated their trust and acceptance in the eHMI, feeling of safety during the interaction and vigilance toward the eHMI. The results showed a positive effect of interacting with a valid system that acted consistently to the vehicle’s movements compared to an initial assessment of the system. After experiencing eHMI malfunctions, participants’ assessment of the system declined significantly. Moreover, elderly participants assessed the eHMI more positive across all conditions than younger participants did. The findings imply that participants considered the vehicle’s movements as implicit communication cues in addition to the provided eHMI signals during the encounters. To support traffic safety and smooth interactions, eHMI signals are required to be in line with vehicle’s movements as implicit communication cues. Moreover, the results underline the importance of calibrating an appropriate level of trust in eHMI signals. An adequate understanding of eHMI signals needs to be developed. Thereby, the requirements of different user groups should be specifically considered.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

The Effect of eHMI Malfunctions on Younger and Elderly Pedestrians’ Trust and Acceptance of Automated Vehicle Communication Signals

et al. 2022

View full text Add to dashboard Cite

show abstract

Measuring Pedestrians’ Gap Acceptance When Interacting with Vehicles - A Human Gait Oriented Approach

Theobald

Joisten

Abendroth

2022

Communications in Computer and Information Science

View full text Add to dashboard Cite

A significant variable describing the pedestrians' behavior when interacting with vehicles is gap acceptance, which is the pedestrians' choice of temporal and spatial gaps when crossing in front of vehicles. After a review of relevant approaches to measure gap acceptance used in studies, this paper presents a novel approach, which is suitable for the usage in field experiments and allows a natural crossing behavior of subjects. In particular, following a detailed analysis of forces exerted during human gait, an algorithm was developed that is capable of identifying the accurate temporal point at which subjects start crossing as the basis for calculating gap acceptance. Pretest results show the system's stability and reliability as well as the gait algorithm's robustness in determining the correct gap acceptance value. The human gait oriented approach can serve as a basis for designing interaction processes between pedestrians and automated vehicles that are a focus of current research efforts.

show abstract