Creating Appropriate Trust in Automated Vehicle Systems: A Framework for HMI Design

Ekman, Fredrick; Johansson, Mikael; Sochor, Jana

doi:10.1109/thms.2017.2776209

Cited by 136 publications

(70 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These results are consistent with the finding that anthropomorphic agents result in greater trust resilience and a higher resistance to breakdowns in trust (de Visser et al., ). Trust is a dynamic process, even on a short time scale of interaction (Ekman et al., ). Humans have a different attitude toward automated agents than to human agents in errors (Madhavan & Wiegmann, ).…”

Section: Conclusion and Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Anthropomorphizing information to enhance trust in autonomous vehicles

Niu

Terken

Eggen

2018

Hum Ftrs & Erg Mfg Svc

View full text Add to dashboard Cite

Trust is an essential condition for accepting and relying on autonomous vehicles. One of the wellstudied factors contributing to trust in automation systems is anthropomorphism. It is expected that anthropomorphism may also enhance trust in the field of autonomous vehicles. A study is presented that investigated the effect of anthropomorphic embodiment for information about the vehicle's maneuvers on people's trust in autonomous vehicles. In a driving simulator experiment deploying a between-subjects design, participants (N = 39) were exposed to varying amounts of information displaying upcoming actions of an autonomous vehicle (symbolic information or symbolic + anthropomorphic information). Each group rated trust and liking for the test condition against a reference condition where no information about the upcoming actions was provided through questionnaires. Symbolic + anthropomorphic information resulted in significantly more trust than symbolic information. Through one-sample tests, it was found that ratings for symbolic + anthropomorphic information were significantly different from no information, while symbolic information by itself was not. Ratings for perceived anthropomorphism were positively correlated with trust and liking. It is concluded that anthropomorphizing information may foster the perception of autonomous vehicles as social agents and enhance trust in those vehicles. K E Y W O R D Santhropomorphism, autonomous vehicles, human-vehicle interaction, information transparency, trust

show abstract

Section: Conclusion and Discussionmentioning

confidence: 99%

“…“Technical competence refers to the degree of user perception on the performance of the autonomous vehicles” (J. K. Choi & Ji, , p. 694). That means it is important to know how the user processes the information presented (Ekman, Johansson, & Sochor, ).…”

Section: Introductionmentioning

confidence: 99%

Anthropomorphizing information to enhance trust in autonomous vehicles

Niu

Terken

Eggen

2018

Hum Ftrs & Erg Mfg Svc

View full text Add to dashboard Cite

show abstract

“…However, these information needs were not quantified and differentiated according to the level of automation. References [4,11,13,16,18] report a degree of certainty (uncertainty information) with which the automation can cope with the current scenario as relevant. Additionally, it has been reported that the remaining automated travel distance/time is important [11,[19][20][21].…”

Section: Information Needs In Automated Drivingmentioning

confidence: 99%

“…NDRAs are thus regarded as one of the greatest benefits of automated driving [3]. However, previous publications have shown that the driver does not want to be completely excluded from the driving process in automated driving [4,5]. Therefore, the identification of desired driving-related information is indispensable to allow adequate human-machine interface (HMI) design.…”

Section: Introductionmentioning

confidence: 99%

Information Needs and Visual Attention during Urban, Highly Automated Driving—An Investigation of Potential Influencing Factors

et al. 2020

View full text Add to dashboard Cite

During highly automated driving, the passenger is allowed to conduct non-driving related activities (NDRA) and no longer has to act as a fallback at the functional limits of the driving automation system. Previous research has shown that at lower levels of automation, passengers still wish to be informed about automated vehicle behavior to a certain extent. Due to the aim of the introduction of urban automated driving, which is characterized by high complexity, we investigated the information needs and visual attention of the passenger during urban, highly automated driving. Additionally, there was an investigation into the influence of the experience of automated driving and of NDRAs on these results. Forty participants took part in a driving simulator study. As well as the information presented on the human–machine interface (system status, navigation information, speed and speed limit), participants requested information about maneuvers, reasons for maneuvers, environmental settings and additional navigation data. Visual attention was significantly affected by the NDRA, while the experience of automated driving had no effect. Experience and NDRA showed no significant effect on the need for information. Differences in information needs seem to be due to the requirements of the individual passenger, rather than the investigated factors.

show abstract

“…Finally, HMI usability is likely to be related to attitudes concerning computers [40] [16] and trust in technology and automation ( [41], [42], [43]) and these should be considered. Concerning older adults, they are more likely to use and develop trust with HMIs that are designed according to their abilities and needs ( [18], [44], [45]).…”

mentioning

confidence: 99%

Exploring the Usability of a Connected Autonomous Vehicle Human Machine Interface Designed for Older Adults

Morgan

Voinescu

Alford

et al. 2018

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

Users of Level 4-5 connected autonomous vehicles (CAVs) should not need to intervene with the dynamic driving task or monitor the driving environment, as the system will handle all driving functions. CAV humanmachine interface (HMI) dashboards for such CAVs should therefore offer features to support user situation awareness (SA) and provide additional functionality that would not be practical within non-autonomous vehicles. Though, the exact features and functions, as well as their usability, might differ depending on factors such as user needs and context of use. The current paper presents findings from a simulator trial conducted to test the usability of a prototype CAV HMI designed for older adults and/or individuals with sensory and/or physical impairments: populations that will benefit enormously from the mobility afforded by CAVs. The HMI was developed to suit needs and requirements of this demographic based upon an extensive review of HMI and HCI principles focused on accessibility, usability and functionality [1,2], as well as studies with target users. Thirty-one 50-88 year-olds (M 67.52, three 50-59) participated in the study. They experienced four seven-minute simulated journeys, involving inner and outer urban settings with mixed speed-limits and were encouraged to explore the HMI during journeys and interact with features, including a real-time map display, vehicle status, emergency stop, and arrival time. Measures were taken pre-, during-and post-journeys. Key was the System Usability Scale [3] and measures of SA, task load, and trust in computers and automation. As predicted, SA decreased with journey experience and although cognitive load did not, there were consistent negative correlations. System usability was also related to trust in technology but not trust in automation or attitudes towards computers. Overall, the findings are important for those designing, developing and testing CAV HMIs for older adults and individuals with sensory and/or physical impairments. IntroductionWe are moving at a rapid pace towards fully autonomous vehicles that handle all driving related functions without the need for user intervention. Level 5 fully connected autonomous vehicles (CAVs) are defined by SAE International [4] as systems that can handle 'the full-time performance by an automated driving system of all aspects of the dynamic driving task under all roadway and environmental conditions that can be managed by a human driver', whereas Level 4 highly automated vehicles are similar but only have such system capabilities for some driving modes and may require human intervention for some non-driving related actions. Both levels assume that the user is not an active driver and can therefore have eyes and mind off the road. This means they will not require driving input devices such as a steering wheel and pedals. Instead, user interaction with Level 4 and 5 CAVs will be through human-machine interface (HMI) dashboards providing vehicle-related (e.g., speed, time/distance to destination, local area information)...

show abstract

Creating Appropriate Trust in Automated Vehicle Systems: A Framework for HMI Design

Cited by 136 publications

References 22 publications

Anthropomorphizing information to enhance trust in autonomous vehicles

Anthropomorphizing information to enhance trust in autonomous vehicles

Information Needs and Visual Attention during Urban, Highly Automated Driving—An Investigation of Potential Influencing Factors

Exploring the Usability of a Connected Autonomous Vehicle Human Machine Interface Designed for Older Adults

Contact Info

Product

Resources

About