In-car Tasks Research Articles

Impact of In-Air Gestures on In-Car Task’s Diver Distraction

As in-vehicle information systems (IVIS) grow increasingly complex, the demand for innovative artificial intelligence-based interaction methods that enhance cybersecurity becomes more crucial. In-air gestures offer a promising solution due to their intuitiveness and individual uniqueness, potentially improving security in human–computer interactions. However, the impact of in-air gestures on driver distraction during in-vehicle tasks and the scarcity of skeleton-based in-air gesture recognition methods in IVIS remain largely unexplored. To address these challenges, we developed a skeleton-based framework specifically tailored for IVIS that recognizes in-air gestures, classifying them as static or dynamic. Our gesture model, tested on the large-scale AUTSL dataset, demonstrates accuracy comparable to state-of-the-art methods and increased efficiency on mobile devices. In comparative experiments between in-air gestures and touch interactions within a driving simulation environment, we established an evaluation system to assess the driver’s attention level during driving. Our findings indicate that in-air gestures provide a more efficient and less distracting interaction solution for IVIS in multi-goal driving environments, significantly improving driving performance by 65%. The proposed framework can serve as a valuable tool for designing future in-air gesture-based interfaces for IVIS, contributing to enhanced cybersecurity.

On the relationship between occlusion times and in-car glance durations in simulated driving

Drivers have spare visual capacity in driving, and often this capacity is used for engaging in secondary in-car tasks. Previous research has suggested that the spare visual capacity could be estimated with the occlusion method. However, the relationship between drivers’ occlusion times and in-car glance duration preferences has not been sufficiently investigated for granting occlusion times the role of an estimate of spare visual capacity. We conducted a driving simulator experiment (N = 30) and investigated if there is an association between drivers’ occlusion times and in-car glance durations in a given driving scenario. Furthermore, we explored which factors and variables could explain the strength of the association. The findings suggest an association between occlusion time preferences and in-car glance durations in visually and cognitively low demanding unstructured tasks but that this association is lost if the in-car task is more demanding. The findings might be explained by the inability to utilize peripheral vision for lane-keeping when conducting in-car tasks and/or by in-car task structures that override drivers’ preferences for the in-car glance durations. It seems that the occlusion technique could be utilized as an estimate of drivers’ spare visual capacity in research – but with caution. It is strongly recommended to use occlusion times in combination with driving performance metrics. There is less spare visual capacity if this capacity is used for secondary tasks that interfere with the driver’s ability to utilize peripheral vision for driving or preferences for the in-car glance durations. However, we suggest that the occlusion method can be a valid method to control for inter-individual differences in in-car glance duration preferences when investigating the visual distraction potential of, for instance, in-vehicle infotainment systems.

Refining distraction potential testing guidelines by considering differences in glancing behavior

Driver distraction is a recognized cause of traffic accidents. Although the well-known guidelines for measuring distraction of secondary in-car tasks were published by the United States National Highway Traffic Safety Administration (NHTSA) in 2013, studies have raised concerns on the accuracy of the method defined in the guidelines, namely criticizing them for basing the diversity of the driver sample on driver age, and for inconsistent between-group results. In fact, it was recently discovered that the NHTSA driving simulator test is susceptible to rather fortuitous results when the participant sample is randomized. This suggests that the results of said test are highly dependent on the selected participants, rather than on the phenomenon being studied, for example, the effects of touch screen size on driver distraction. As an attempt to refine the current guidelines, we set out to study whether a previously proposed new testing method is as susceptible to the effects of participant randomization as the NHTSA method. This new testing method differs from the NHTSA method by two major accounts. First, the new method considers occlusion distance (i.e., how far a driver can drive with their vision covered) rather than age, and second, the new method considers driving in a more complex, and arguably, a more realistic environment than proposed in the NHTSA guidelines. Our results imply that the new method is less susceptible to sample randomization, and that occlusion distance appears a more robust criterion for driver sampling than merely driver age. Our results are applicable in further developing driver distraction guidelines and provide empirical evidence on the effect of individual differences in drivers’ glancing behavior.

Impacts of Touch Screen Size, User Interface Design, and Subtask Boundaries on In-Car Task's Visual Demand and Driver Distraction

• In-car UI design factors’ visual distraction potential should be better understood • First study on UI features’ visual demand while controlling visual demands of driving • Automotive-targeted application decreases glance durations and visual distraction • Larger screen size has relatively small decreasing impact on glance durations • Subtask boundaries enable adjustment of glancing behavior in relation to driving situation • Visual demand of a task is not equal to visual distraction Visual distraction by secondary in-car tasks is a major contributing factor in traffic incidents. In-car user interface design may mitigate these negative effects but to accomplish this, design factors’ visual distraction potential should be better understood. The effects of touch screen size, user interface design, and subtask boundaries on in-car task's visual demand and visual distraction potential were studied in two driving simulator experiments with 48 participants. Multilevel modeling was utilized to control the visual demands of driving and individual differences on in-car glance durations. The 2.5” larger touch screen slightly decreased the in-car glance durations and had a diminishing impact on both visual demand and visual distraction potential of the secondary task. Larger relative impact was discovered concerning user interface design: an automotive-targeted application decreased the visual demand and visual distraction potential of the in-car tasks compared to the use of regular smartphone applications. Also, impact of subtask boundaries was discovered: increase in the preferred number of visual or visual-manual interaction steps during a single in-car glance (e.g., pressing one button vs. typing one word) increased the duration of the in-car glance and its visual distraction potential. The findings also emphasize that even if increasing visual demand of a task – as measured by in-car glance duration or number of glances – may increase its visual distraction potential, these two are not necessarily equal.

Context-sensitive distraction warnings – Effects on drivers׳ visual behavior and acceptance

In this study, we investigated the effects of context-sensitive distraction warnings on drivers' in-car glance behaviors and acceptance. The studied prototype warning application functions on a smart phone. The novelty of the application is its proactive and context-sensitive approach to the adjustment of warning thresholds according to the estimated visual demands of the driving situation ahead. In our study, novice and experienced drivers conducted in-car tasks with a smart phone on a test track with and without the warnings. The application gave a warning if the driver's gaze was recognized to remain on the smart phone over a situation-specific threshold time, or if the driver was approaching a high-demand part of the track (an intersection or a tight curve). Glance metrics indicated a significant increasing effect of the warnings on glance time on road while multitasking. The effect varied between 5% and 30% increase depending on the in-car task. A text message reading task was the most visually demanding activity and indicated the greatest effect of the warnings on glance time on road. Driving experience did not have an effect on the efficiency of the warnings. The proposed gaze tracking with current smart phone technology proved to be highly unreliable in varying lighting conditions. However, the findings suggest that location-based proactive distraction warnings of high-demanding driving situations ahead could help all drivers in overcoming the inability to evaluate situational demands while interacting with complex in-car tasks and to place more attention on the road. Furthermore, survey results indicate that it is possible to achieve high levels of trust, perceived usefulness, and acceptance with these kinds of context-sensitive distraction warnings for drivers. Context-sensitive distraction warnings had a positive effect on drivers' behaviors.The warnings significantly increased glance time on road, especially while reading a text message.The level of driving experience did not seem to affect the efficiency of the warnings.Location-based warnings worked more reliably than glance duration based warnings.The context-sensitive warning application was well accepted by the drivers.

Modeling visual sampling on in-car displays: The challenge of predicting safety-critical lapses of control

In this article, we study how drivers interact with in-car interfaces, particularly by focusing on understanding driver in-car glance behavior when multitasking while driving. The work focuses on using an in-car touch screen to find a target item from a large number of unordered visual items spread across multiple screens. We first describe a cognitive model that aims to represent a driver׳s visual sampling strategy when interacting with an in-car display. The proposed strategy assumes that drivers are aware of the passage of time during the search task; they try to adjust their glances at the display to a time limit, after which they switch back to the driving task; and they adjust their time limits based on their performance in the current driving environment. For visual search, the model assumes a random starting point, inhibition of return, and a search strategy that always seeks the nearest uninspected item. We validate the model׳s predictions with empirical data collected in two driving simulator studies with eye tracking. The results of the empirical study suggest that the visual design of in-car displays can have a significant impact on the probability of distraction. In particular, the results suggest that designers should try to minimize total task durations and the durations of all visual encoding steps required for an in-car task, as well as minimize the distance between visual display elements that are encoded one after the other. The cognitive model helps to explain gaze allocation strategies for performing in-car tasks while driving, and thus helps to quantify the effects of task duration and visual item spacing on safety-critical in-car glance durations.

Joint Uncertainty Decoding With Predictive Methods for Noise Robust Speech Recognition

Model-based noise compensation techniques are a powerful approach to improve speech recognition performance in noisy environments. However, one of the major issues with these schemes is that they are computationally expensive. Though techniques have been proposed to address this problem, they often result in degradations in performance. This paper proposes a new, highly flexible, approach which allows the computational load required for noise compensation to be controlled while maintaining good performance. The scheme applies the improved joint uncertainty decoding with the predictive linear transform framework. The final compensation is implemented as a set of linear transforms of the features, decoupling the computational cost of compensation from the complexity of the recognition system acoustic models. Furthermore, by using linear transforms, changes in the correlations in the feature vector can also be efficiently modeled. The proposed methods can be easily applied in an adaptive training scheme, including discriminative adaptive training. The performance of the approach is compared to a number of standard schemes on Aurora 2 as well as in-car speech recognition tasks. Results indicate that the proposed scheme is an attractive alternative to existing approaches.

Navigational conversation impairs concurrent distance judgments

Dual-task performance as it relates to driving, such as tuning a radio or manipulating a cellular phone, forces drivers to divide their attention between the traffic demands and the in-car task. The present study investigated how concurrent spatial or non-spatial cognitive distractions mediated proximity judgments using vehicular stimuli. Utilizing a modified version of the task employed by [Elias, L.J., Robinson, B. in press. Drive on the right side of the road: perceptual asymmetries for judgments of automobile proximity. International Journal of Neuroscience.] the current study examined how mental navigation (spatial distraction) affected accuracy and response time for depth judgments on vehicular stimuli in each visual field. These were compared to a control condition in which no distraction was present, as well as when a semantic (non-spatial) distraction was present. We found that conversation of a navigational nature (i.e., spatial distraction) most negatively impacted accuracy and response time when processing dynamically changing vehicle proximity. Further, these deleterious effects appeared to be uniform throughout the visual field. Findings are related to driving while being distracted, with particular emphasis on the role of cerebral lateralization in dual-task performance.

Aging and Time-Sharing in Highway Driving

Proper time-sharing--visual attention allocation--between the road view ahead and other targets is an essential requirement for safe driving, along with other visual and attentional performance. Earlier on-road research has shown that neurologic problems (Alzheimer disease, brain injury) impair time-sharing during in-car tasks. This study analyzed age effects on time-sharing performance. Thirty participants in three age groups (mean age 22, 34, and 67 years) drove an instrumented car a trip of 350 km and performed an in-car visual search task with either a motor (keying) or vocal response. The frequency and duration of glances at the in-car targets, total time eyes off the road during task, speed, and lateral displacement of the car were recorded. The participants were also tested on a battery of cognitive tasks during the midway break. The elderly used a longer total time looking at the in-car display and they traveled a longer distance with eyes away from the road. The number of long (>2 sec) glances and the car's lateral displacement on the road were larger among the elderly than the young drivers. The difference between the older and younger participants was larger when a motor (keying) response was required. The age effects were mediated by cognitive performance (best by the Trail Making A test) rather than by vision parameters. Older drivers have difficulties in time-sharing in highway driving already at the age of 65 to 70 years.

Time-sharing strategies in driving after various cerebral lesions

Primary objective: To analyse time-sharing strategies in patients with cerebral lesions when they performed everyday in-car tasks in real-life highway driving. Research design: A case-control study. Methods and procedures: Thirteen male patients with brain damage and 11 healthy controls participated. The frequency and duration of glances at the in-car tasks, total time of eyes-off the road during tasks, speed and lateral displacement of the car were recorded. Main outcomes and results: Long glances away from the road and consequent large lateral displacements were characteristic for patients with anterior damage, while the patients with more posterior lesions used frequent short glances at the in-car task. Conclusion: Two different time-sharing strategies that appeared to be related to different aetiology were found in the patients, indicating different control of the driving task. It is proposed that standardized on-road dual tasks should be used for detecting problems in allocation of attention during driving.

Detection thresholds in car following situations and peripheral vision: implications for positioning of visually demanding in-car displays

This on-road study examined the effect of the positioning of an attentionally demanding in-car task on the driver's ability to detect the approach of a decelerating car ahead. Twelve participants aged between 19 and 27 years with an average of 18 170 km life-time driving experience drove 20 and 40 m behind a vehicle which decelerated from 50 km/h at ~0.7 m/s2 without braking. Detection thresholds for nine locations of a LED display were compared with thresholds when focusing on the car ahead. A strong inverse relationship was found between time-to-collision (TTC) and eccentricity of the task to the normal line of sight, with TTC decreasing from 6 to 8 s at 0° eccentricity to 4 s at 90°. The results suggest that there are optimal locations, in terms of detecting the deceleration of a car ahead, for positioning attentionally demanding in-car devices, and that there are some differences in detection thresholds for similar eccentricities in the vertical and horizontal peripheries of the eye.

Driving experience and perception of the lead car's braking when looking at in-car targets

Perception of the lead car's braking was measured on-road when subjects of various levels of driving experience were looking at a digital display located at the lower part of the windscreen, at the speedometer level, or in the mid-console. The brake lights of the lead car were either working normally or switched off. The results indicated that the detection of the lead car's brake lights, in daylight, is substantially impaired when a following driver is looking at the speedometer area and brake lights do not contribute to detection at all when he/she is looking at a target in the mid-console. Driving experience did not influence performance in detecting a closing headway in peripheral vision, in contrast to improvement in lane-keeping found in an earlier study. It is suggested that such differential ability in using peripheral vision for lane and distance-keeping may mislead experienced drivers when they follow another vehicle and perform certain in-car tasks.

Neuropsychological Correlates of Duration of Glances at Secondary Tasks While Driving

The aim of this study was to find out the neuropsychological measures correlating with overlong glances at secondary in-car tasks while driving. Fifteen. patients with brain damage (without clear neurological or neuropsychological restriction on driving a car) and 11 healthy participants drove a route of 126 km and performed a series of secondary tasks while driving on a highway in an instrumented compact car. Four videocameras allowed detailed analysis of glances during in-car tasks. Neuropsychological measures focused on executive functions, memory, visuospatial skills, and fine motor skills. Moreover, patients' emotional self-evaluation and relatives' evaluation of patients' competencies were included. The proportion of overlong glances away from the road during in-car tasks was greater among the patients than. the healthy drivers. The long glances of the patients correlated strongly with motor and visuospatial deficits, cognitive inflexibility, emotional symptoms, and relatives' evaluations of patients' impaired sensomotor abilities. The results suggest that the frequency of overlong glances was increased by 2 factors: (a) impaired motor and visuospatial skills that evidently caused difficulties in the manipulation of the equipment of the secondary tasks, and (b) impairments of executive functions that were likely to decrease the ability to control the risks related to long glances at the in-car tasks. The slowing of speed during secondary tasks was on the average rather slight and not significantly more pronounced among patients than control drivers, indicating that patients failed to reduce their speed and thus the risk related to prolonged glances at in-car tasks.

Driving experience and time-sharing during in-car tasks on roads of different width

Earlier research has not found consistent differences between the way novices and experienced drivers allocate time between road and in-car tasks when average time spent on the latter is being considered. This study addresses the question of whether novices occasionally fail in supervisory control more often than experienced drivers. The influence of driving experience and road type on the allocation of visual attention was investigated by analysing the distribution of glance duration at in-car tasks for 23 experienced (life-time driving experience 50 000 – 2 000 000 km) and 24 inexperienced (life-time driving experience 400 – 15 000 km) drivers. The duration and number of glances were recorded while the participants changed a radio cassette, dialled a mobile phone or tuned the radio, while driving on a highway or motorway. The experienced drivers allocated their visual attention more adequately, here interpreted as attributed to their greater driving experience. The glance duration of the novices showed a larger variance, due to a greater percentage of short, possibly ineffective, and long, risky glances. None of the experienced drivers took glances longer than 3 s at the in-car task, but 29% of the novices did. The novices' long glances were also associated with large lateral displacement of the car. Drivers took longer glances on a motorway than on a highway, which supports the notion that drivers accommodate their glances to the time margins dictated by different traffic situations.

Novice and Experienced Drivers' Looking Behavior and Primary Task Control While Doing a Secondary Task

This study is addressed to time-sharing and primary task control during a secondary task as a function of driving experience. After about 1.5 h of test driving, when well-adapted to the experimental car, 23 novices (less than 5,000 km of driving) and 26 experienced drivers (more than 150,000 km) were asked to change a cassette in a cassette player on an ordinary two-lane road. The task was replicated three times. The results showed no difference between novice and experienced drivers in time-sharing (glance length at the in-car task and at the road), lateral position-keeping (lateral displacement as a function of time at in-car task) or control in relation to oncoming traffic. The only difference occurred in speed control, experienced drivers keeping their speed level constant while novices slowed down somewhat during the secondary task. These data showed, in a supervised experimental setting, a similar linear relationship between time spent on an in-car task and lateral displacement both for novice and experienced drivers, and a similar median time gap of about 2 s to an oncoming vehicle at the moment when both novice and experienced drivers shifted their gaze from the in-car task to the road.

An Accuracy Analysis of Techniques for Measuring the Durations of In-Car Manual Tasks

Human factors research of automobile driver behavior often calls for timing in-car manual tasks. The present study was designed to compare the accuracy, bias, and consistency of various techniques for measuring in-car manual task durations. Additionally, this research was intended to reveal how closely these techniques approach the preciseness of the frame-by-frame video analysis method, which is time-consuming and expensive to perform. Six subjects were required to use an electronic stopwatch to measure “hand-off-wheel” times for 30 driver tasks. Each subject performed this procedure three times: while sitting as an observer in the back seat of a research vehicle, while watching a real-time video recording of task performance, and while watching a one-sixth real-time video recording of task performance. Timing Method (three levels), Duration of in-car task (three levels), and Subject (six levels) served as independent variables. Dependent measures gathered were raw timing error (a measure of response bias), absolute timing error (a measure of response accuracy), and squared timing error (a measure of response consistency). Timing error was obtained by subtracting the measured time for a particular task from the “true” task time obtained by using the frame-by-frame video analysis technique. Analysis of the data indicated a significant effect of Method on response bias. Specifically, use of the slow-motion video technique resulted in overestimation of in-car task durations, and use of the two real-time techniques resulted in estimates of task durations that were either equal to or less than the true durations. Significant effects of Subject, Gender, and Subject x Method were also revealed. The results suggest that the on-road timing technique should be used in the future, since this procedure requires little in terms of cost and implementation time, and errors are small when compared with the frame-by-frame technique.