On-road trust and perceived risk in Level 2 automation

To encourage appropriate use of driving automation, we need to understand and monitor driver’s trust and risk perception. We examined (1) how trust and perceived risk are affected by automation, driving conditions and experience and (2) how well perceived risk can be inferred from behaviour and physiology at three levels: over traffic conditions, aggregated risk events, and individual risk events.30 users with and without automation experience drove a Toyota Corolla with driving support. Safety attitude, subjective ratings, behaviour and physiology were examined.Driving support encouraged a positive safety attitude and active driver involvement. It reduced latent hazards while maintaining saliently perceived risks. Drivers frequently overruled lane centring (3.1 times/minute) and kept their feet on or above the pedals using ACC (65.8% of time). They comfortably used support on curvy motorways and monotonic and congested highways but less in unstable traffic and on roundabouts. They trusted the automation 65.4%, perceived 36.0% risk, acknowledged the need to monitor and would not engage in more secondary tasks than during manual driving.Trust-in situation reduced 2.0% when using automation. It was 8.2% higher than trust-in-automation, presumably due to driver self-confidence. Driving conditions or conflicts between driver and automation did not affect trust-in-automation.At the traffic condition level, physiology showed weak and partially counter-intuitive effects. For aggregated risk events, skin conductance had the clearest response but was discernible from baseline in  < 50%. Pupil dilation and heart rate only increased with strong braking and active lane departure assist. For individual risk events, a CNN classifier could not identify risk events from physiology. We conclude that GSR, heart rate and pupil dilation respond to perceived risk, but lack specificity to monitor it on individual events.     

Alternative options for dealing with automation failures: Automated stopping vs. Taking over manual control

Previous research has investigated the driver’s ability to resume control after an automation failure. This study considers situation criticality, safety and difficulty between take-over and automated stop scenarios. Findings are reported of an experimental study investigating different strategies to take back control of an automated road vehicle. The study had a mixed factorial design with automation being the between-participants factor with two levels, SAE Level 1 vs. SAE Level 3. Urgency (High vs Low) and Take-Over Mode (Manual, Automated Stop) were the within participants factors. Dependent variables were: i) Minimum Time-to-Collision (TTC), ii) Self-reported Situation Criticality and Perceived Safety and iii) Percentage of Gazes to [the] Road Centre. Overall, 36 participants took part, with half assigned to either level of the Automation factor. The automated driving experience was designed to fail during participation. This required the driver to take-over manual control or have the system bring the vehicle to an automated stop. Results revealed that during manual take-over of control, measures of time-to-collision, perceived criticality and safety were worse and fixations to the forward view increased when compared to the Automated stop condition. Manual take-over of control in a high-urgency situation revealed higher criticality (shorter time-to-collision and lower perceived safety) and more difficulty (increasing fixations to the road centre) than in a low-urgency situation. The study points to an automated stopping manoeuvre being the preferable strategy to manual take-over of control.     

Effects of explanation types and perceived risk on trust in autonomous vehicles

Despite technological advances, trust still remains as a major issue facing autonomous vehicles. Existing studies have reported that explanations of the status of automation systems can be an effective strategy to increase trust, but these effects can differ depending on the forms of explanations and autonomous driving situations. To address this issue, this study examines the effects of explanation types and perceived risk on trust in autonomous vehicles. Three types of explanations (i.e., no, simple, and attributional explanations) are designed based on attribution theory. Additionally, four autonomous driving situations with different levels of risk are designed based on a simulator program. Results show that explanation type significantly affects trust in autonomous vehicles, and the perceived risk of driving situations significantly moderates the effect of the explanation type. At a high level of perceived risk, attributional explanations and no explanations lead to the lowest and highest values in trust, respectively. However, at a low level of perceived risk, these effects reverse.     

How to identify the take-over criticality in conditionally automated driving? An examination using drivers’ physiological parameters and situational factors

Autonomous vehicles and advanced driver assistance technology are growing exponentially, and vehicles equipped with conditional automation, which has features like Traffic Jam Pilot and Highway Assist, are already available in the market. And this could expose the driver to a stressful driving condition during the takeover mission. To identify stressful takeover situations and better interact with automated systems, the relationship and effect between drivers’ physiological responses, situational factors (e.g., takeover request [TOR] lead time, takeover frequencies, and scenario types), and takeover criticality were investigated.34 participants were involved in a series of takeover events in a simulated driving environment, which are varied by different TOR lead time conditions and driving scenes. The situational factors, drivers’ skin conductance (SC), heart rate (HR), gaze behaviors, and takeover criticality ratings were collected and analyzed. The results indicated that drivers had a higher takeover criticality rating when they experienced a shorter TOR lead time level or at first to fourth take-overs. Besides, drivers who encountered a dynamic obstacle reported higher takeover criticality ratings when they were at the same Time to collision (TTC). We also observed that the takeover situations of higher criticality have larger driver’s maximum HR, mean pupil size, and maximum change in the SC (relative to the initial value of a takeover stage). Those findings of situational factors and physiological responses can provide additional support for the designing of adaptive alert systems and environmental soothing technology in conditionally automated driving, which will improve the takeover performances and drivers’ experience.     

What happens when drivers of automated vehicles take over control in critical lane change situations?

According to legislation, take-overs initiated by the driver must always be possible during automated driving. For example, when drivers mistrust the automation to handle a critical and hazardous lane change, they might intervene and take over control while the automation is performing the maneuver. In these situations, drivers may have little time to avoid an accident and can be exposed to high lateral forces. Due to lacking research, it is yet unknown if they recognize the criticality of the situation and how they behave and perform to manage it. In a driving simulator study, participants (N = 60) accomplished eight double lane changes to evade obstacles in their lane. Time-to-collision and traction usage were varied to establish different degrees of objective criticality. To manipulate these parameters as required, participants were triggered to take over control by an acoustic cue. This setting shows what might happen if drivers disable the automation and complete the maneuver themselves. The results of the experiment demonstrate that drivers rated objectively more critical driving situations as more critical and responded to the hazard very fast over all experimental conditions. However, their behavior was more extreme with respect to decelerating and steering than necessary. This impaired driving performance and increased the risk of lane departures and collisions. The results of the experiment can be used to develop an assistance system that supports driver-initiated take-overs.     

Effects of forward collision warning and repeated event exposure on emergency braking

Many experimental studies use repeated lead vehicle braking events to study the effects of forward collision warning (FCW) systems. It can, however, be argued that the use of repeated events induce expectancies and anticipatory behaviour that may undermine validity in terms of generalisability to real-world, naturalistic, emergency braking events. The main objective of the present study was to examine to what extent the effect of FCW on response performance is moderated by repeated exposure to a critical lead vehicle braking event. A further objective was to examine if these effects depended on event criticality, here defined as the available time headway when the lead vehicle starts to brake. A critical lead vehicle braking event was implemented in a moving-base simulator. The effects of FCW, repeated event exposure and initial time headway on driver response times and safety margins were examined. The results showed that the effect of FCW depended strongly on both repeated exposure and initial time headway. In particular, no effects of FCW were found for the first exposure, while strong effects occurred when the scenario was repeated. This was interpreted in terms of a switch from closed-loop responses triggered reactively by the situation, towards an open-loop strategy where subjects with FCW responded proactively directly to the warning. It was also found that initial time headway strongly determined response times in closed-loop conditions but not in open-loop conditions. These results raise a number of methodological issues pertaining to the design of experimental studies with the aim of evaluating the effects of active safety systems. In particular, the implementation of scenario exposure and criticality must be carefully considered.     

Investigating driver reactions to movements of autonomous vehicle in permitted right turn through driving simulator experiments

Traffic safety has always been a hot topic for human-driven (HDV) and autonomous vehicles (AV) mixed flow. The conflict between permitted right-turn vehicles (PRT) and opposing through vehicles (TH) at signalized intersections (left-handed traffic) is extraordinarily critical. AVs with aggressive behaviors are able to accept short gap time without losing safety. However, such a turning maneuver may lead to dangerous feelings and cause unexpected reactions of approaching drivers. This study aims to investigate and model drivers’ reactions in TH movements to PRT AVs considering the trust degree of drivers to AVs. Questionnaire surveys and driving simulator experiments were conducted for 41 participants. Results reveal that the right turn timing of PRT AV will significantly influence drivers’ reactions. Basically, TH drivers will brake with a high probability under the situation of small expected post encroachment time (PET). It is also found that female drivers and drivers with low trust in AVs are more vigilant to PRT AVs than other drivers. Based on this finding a two-layer model for reproducing TH drivers’ reactions to PRT AVs is proposed. The first layer is to determine the braking decision and the second layer is to calculate the parameters of braking behaviors (brake lag, braking time, and speed drop). The significance and coefficients variables in these models proved that the trust in AV will influence drivers’ decisions and braking behaviors (brake lag and braking time). The more the drivers trust AVs, the smaller the expected PET to AVs they can accept for passing without braking, and the more gently they will brake (longer brake lag and shorter braking time) due to the cutting in of PRT AVs. This effect will become significant after drivers have experienced several interactions with AVs.     

Safety at first sight? – Manual drivers’ experience and driving behavior at first contact with Level 3 vehicles in mixed traffic on the highway

Soon, manual drivers will interact with conditionally automated vehicles (CAVs; SAE Level 3) in a mixed traffic on highways. As of yet, it is largely unclear how manual drivers will perceive and react to this new type of vehicle. In a driving simulator study with N = 51 participants aged 20 to 71 years (22 female), we examined the experience and driving behavior of manual drivers at first contact with Level 3 vehicles in four realistic driving scenarios (highway entry, overtaking, merging, introduction of a speed limit) that Level 3 vehicles may handle alone once their operational domain extends beyond driving in congested traffic. We also investigated the effect of an external marking via a visual external human–machine interface (eHMI), with participants being randomly assigned to one of three experimental groups (none, correct, incorrect marking). Participants experienced each driving scenario four times, twice with a human-driven vehicle (HDV), and twice with a CAV. After each interaction, participants rated perceived driving mode of the target vehicle as well as perceived safety and comfort. Minimum time headways between participants and target vehicles served as an indicator of safety criticality in the interactions. Results showed manual driver can distinguish CAVs from HDVs based on behavioral differences. In all driving scenarios, participants rated interactions with CAVs at least as safe as interactions with HDVs. The driving data analysis showed that manual driver interactions with CAVs were largely uncritical. However, the CAVs’ strict rule-compliance led to short time headways of following manual drivers in some cases. The eHMI used in this study neither affected the subjective ratings of the manual drivers nor their driving behavior in mixed traffic. Thus, the results do not support the use of eHMIs on the highway, at least not for the eHMI design used in this study.     

Personality and demographic differences in the perceived risks of potentially timid driving behaviours

Timid driving behaviours can be described as overly cautious and hesitant driving behaviours. Little research has examined behaviours that potentially resemble timid driving and how these behaviours are perceived by other drivers. This is despite the potential for these behaviours to be perceived in a way that leads to angry and aggressive retaliatory behaviours in some drivers (e.g., in anger-prone drivers). We conducted an online survey examining the perceived road safety risks of several behaviours that could potentially result from timid driving and their relationships with driver personality (trait anxiety, trait driving anger), behaviour (anxious driving, angry driving), and demographic (age, gender, annual mileage) background. Drivers (N = 439, M_age = 49.41 ± 5.59 years, aged 18–89) perceived excessively cautious and unpredictable braking behaviours as posing moderate levels of risk. Multiple linear regression analyses also indicated higher perceived risks of slow and excessively cautious behaviours in older, male, and anger prone drivers. No meaningful associations were found between driver characteristics and the risks of unpredictable braking behaviours. These results suggest that safety campaigns to reduce aggressive behaviour may benefit from targeting the perceptions of other drivers’ behaviours.     

Trust Development: Testing a New Model in Undergraduate Roommate Relationships

Interpersonal trust is a vital component of social relationships. In this study the roles of parental attachment, perceived similarity of trustee to self, and social exchange processes in trust development were investigated longitudinally with randomly assigned, same-sex undergraduate roommates during emerging adulthood. A total of 214 first-year students completed weekly self-report measures during the first 5 weeks of the fall semester. Perceived similarity measured the second week and social exchange with roommates across the 5 weeks predicted participants’ trust in their roommate, with social exchange mediating the relation between perceived similarity and trust. Results highlight interrelations of social exchange and trust in established relationships.     

Assessing safety critical braking events in naturalistic driving studies

Naturalistic driving studies permit the study of driving behaviour during every day driving. Such studies have a long duration and rare events such as near-crashes and even crashes do occur during the period of the study. This fact gives an opportunity to study events that are otherwise difficult to find. However, the vast amount of data recorded within these naturalistic driving studies demands a huge amount of manual work to identify hazardous situations. This paper concerns the development and validation of a new method, based on critical jerk, to identify safety critical braking events during car driving. The method was compared with one of today’s most used method, which is based on the longitudinal acceleration measure. Both methods were applied on near-crash data from the 100-car naturalistic driving study previously carried out by the Virginia Tech Transportation Institute (VTTI). The data included 637 near-crashes. The results from the analyses showed that the critical jerk method performed approximately 1.6 times higher overall success rate than the method based on the longitudinal acceleration measure. In addition, a positive correlation was found between driver’s safety critical braking event and crash involvement. The conclusion is that the critical jerk method is capable of detecting safety critical braking events and may also be used for assessing high risk drivers.     

Developing human-machine trust: Impacts of prior instruction and automation failure on driver trust in partially automated vehicles

To prompt the use of driving automation in an appropriate and safe manner, system designers require knowledge about the dynamics of driver trust. To enhance this knowledge, this study manipulated prior information of a partial driving automation into two types (detailed and less) and investigated the effects of the information on the development of trust with respect to three trust attributions proposed by Muir (1994): predictability, dependability, and faith. Furthermore, a driving simulator generated two types of automation failures (limitation and malfunction), and at six instances during the study, 56 drivers completed questionnaires about their levels of trust in the automation. Statistical analysis found that trust ratings of automation steadily increased with the experience of simulation regardless of the drivers’ levels of knowledge. Automation failure led to a temporary decrease in trust ratings; however, the trust was rebuilt by a subsequent experience of flawless automation. Results showed that dependability was the most dominant belief of drivers’ trust throughout the whole experiment, regardless of their knowledge level. Interestingly, detailed analysis indicated that trust can be accounted by different attributions depending on the drivers’ circumstances: the subsequent experience of error-free automation after the exposure to automation failure led predictability to be a secondary predictive attribution of drivers’ trust in the detailed group whilst faith was consistently the secondary contributor to shaping trust in the less group throughout the experiment. These findings have implications for system design regarding transparency and for training methods and instruction aimed at improving driving safety in traffic environments with automated vehicles.     

Understanding driver anger and aggression: attributional theory in the driving environment

Two studies tested the applicability of Weiner's (1995, 1996, 2001, 2006) attributional model of social conduct to roadway environments. This model highlights the role of inferences of responsibility after making causal judgments for social transgressions. Study 1 employed written scenarios where participants were asked to imagine themselves driving on a major highway. The degree of controllability and intentionality of the driving act was manipulated experimentally by altering the specific event-related details provided to the participants. Study 2 extended this research to life events by having participants complete online driving diaries every 2 days, identifying their most negative/upsetting encounter with another motorist. The most anger-provoking event was selected from among 4 diary entries and participants were asked to respond to a questionnaire similar to that used in Study 1. Path analyses in both studies generally supported predictions derived from Weiner's model; the association between perceived controllability, intentionality, and dispositional locus of causality of the negative driving event and subsequent anger was mediated by perceptions of responsibility. Additional results in Study 2 suggested that low perceived controllability, intentionality, and dispositional locus of causality were associated with reduced perceived responsibility, which, in turn, facilitated feelings of sympathy. Anger was associated with aggressive responses to the offending driver, whereas sympathy was associated with prosocial responses. Recommendations were offered for improved driver safety, including the development of attributional retraining programs to combat self-serving attributional biases, teaching novice drivers about both formal and informal roadway communication, and the promotion of forgiveness among drivers     

Perceived racism and vascular reactivity in black college women: moderating effects of seeking social support.

This quasi-experimental study explored the association of perceived racism and seeking social support to vascular reactivity in a college sample of 110 Black women. Perceived racism and seeking social support were assessed via self-report, and vascular reactivity was measured before and during a standardized speaking task. Hierarchical regression analyses indicated that perceived racism was positively related to changes in systolic blood pressure. These analyses also indicated that seeking social support moderated the relationship between perceived racism and systolic blood pressure changes. This interaction effect persisted after controlling for several potential confounders. Follow-up regression analyses showed that perceived racism was positively associated with reactivity among participants who were low in seeking social support. A significant relationship was not observed between perceived racism and systolic blood pressure changes among participants who were high in seeking social support. Perceived racism and seeking social support were not significantly associated with changes in diastolic blood pressure. These findings highlight the importance of examining psychosocial factors that may mitigate the hypothesized relationship between perceived racism and reactivity.     

General and Familiar Trust in Websites

When people rely on the web to gather and distribute information, they can build a sense of trust in the websites with which they interact. Understanding the correlates of trust in most websites (general website trust) and trust in websites that one frequently visits (familiar website trust) is crucial for constructing better models of risk perception and online behavior. We conducted an online survey of active Internet users and examined the associations between the two types of web trust and several independent factors: information technology competence, adverse online events, and general dispositions to be trusting or cautious of others. Using a series of nested ordered logistic regression models, we find positive associations between general trust, general caution, and the two types of web trust. The positive effect of information technology competence erases the effect of general caution for general website trust but not for familiar website trust, providing evidence that general trust and self-reported competence are stronger associates of general website trust than broad attitudes about prudence. Finally, the experience of an adverse online event has a strong, negative association with general website trust, but not with familiar website trust. We discuss several implications for online behavior and suggest website policies that can help users make informed decisions about interacting with potentially risky websites.     

Perceived trustworthiness of online shops

• Trust has been found to be crucial for consumer behaviour towards online shops. However, existing studies on the role of trust in electronic commerce are mainly based on ad hoc scales to measure trustworthiness, merely rely on self‐reported consequences of trust such as intention to buy, and focus on low‐risk products.
• In a web‐based study, 634 participants interacted with a provider of medical goods, that is a simulated online pharmacy. The study develops and employs a psychometrically sound scale for assessing perceived trustworthiness of online shops. Moreover, it examines the impact of trustworthiness on both consumers' intended and actual behaviour towards online shops.
• Results show that trustworthiness promotes both intention to buy and actual financial risk taking. Perceived risk was not found to moderate the relationship between trustworthiness and intention to buy. Instead, trustworthiness partially mediated the influence of perceived risk on intention to buy. The results from the scale development challenge multidimensional conceptualizations of trust; comparing this finding with other studies suggests that the duration of the relationship might moderate the dimensionality of trust.

Copyright © 2008 John Wiley & Sons, Ltd.     

Distracted Driving Among College Students: Perceived Risk Versus Reality

Although the rate of alcohol-impaired driving among adolescents has declined in the past two decades, distracted driving has become a major public safety concern. The present study compared perceptions of accident risk and social norms related to cell phone use while driving (CPWD), as well as alcohol-impaired driving, with self-reported behavior among a sample of 726 college students. Results indicated that although participants perceived sending text messages while driving as posing a similar accident risk as driving while legally intoxicated, they were much more likely to text behind the wheel. Furthermore, participants perceived their peers as being more accepting of and having more liberal views toward CPWD than their own, suggesting that one factor underlying the discrepancy between perceived risk and risk exposure may be the level of social acceptability attributed to texting while driving. Future interventions may benefit from focusing not only on risk perception, but on social norms, legal consequences, and adaptive alternatives.     

The potential of gamification for user education in partial and conditional driving automation: A driving simulator study

Drivers must establish adequate mental models to ensure safe driver-vehicle interaction in combined partial and conditional driving automation. To achieve this, user education is considered crucial. Since gamification has previously shown positive effects on learning motivation and performance, it could serve as a measure to enhance user education on automated vehicles. We developed a tablet-based instruction involving gamified elements and compared it to instruction without gamification and a control group receiving a user manual. After instruction, participants (N = 57) experienced a 30-minute automated drive on a motorway in a fixed-base driving simulator. Participants who received the gamified instruction reported a higher level of intrinsic motivation to learn the provided content. The results also indicate that gamification promotes mental model formation and trust during the automated drive. Taken together, including gamification in user education for automated driving is a promising approach to enhance safe driver-vehicle interaction.     

Driver responses to motorcycle and lead vehicle braking events: The effects of motorcycling experience and novice versus experienced drivers

In many countries, motorcyclists are over-represented in traffic collision fatalities and injuries compared to vehicle registrations. Why drivers may violate the right-of-way of motorcyclists traveling as lead vehicles in front of drivers is empirically examined in two studies that were conducted with a moderate-fidelity driving simulator. The purpose of the first study was to determine if drivers, who also held a motorcycle license (N = 16), drove cars differently than regular drivers (N = 16) around motorcycles. The two groups did not differ on responses to motorcycling braking events, which was consistent with previous research on car following. The second study compared the driving performance of sixteen novice teenage drivers (M = 16.2 years of age) to 15 experienced drivers (M = 32.9) over the span of six monthly simulator sessions. Novice drivers’ perception response times (PRT) to the braking events were significantly longer than those of the experienced drivers. PRTs to motorcycle and lead vehicle braking events decreased over sessions. For all participants, PRTs to the motorcycle events were longer than to the car events. The implications of these results for motorcyclists and drivers with different levels of experience are discussed.     

Increasing following headway with prompts, goal setting, and feedback in a driving simulator

We evaluated the effects of prompting, goal setting, and feedback on following headway of young drivers in a simulated driving environment and assessed whether changes produced in following headway were associated with reductions in hard braking when drivers were and were not using cell phones. Participants were 4 university students. During baseline, drivers spent half of the time talking on cell phones while driving. At the start of the intervention, drivers were prompted to increase following headway while on the cell phones and were provided a specific target for following headway. Drivers were given feedback on increasing following headway when on cell phones at the end of each session. The intervention package was associated with an increase in following headway and a decrease in hard braking when participants were on and off the cell phones. Cell phone use did not affect any of the measures.