2017 model-year automobiles provide a variety of features and functions that allow motorists to perform a plethora of secondary tasks unrelated to the primary task of driving. Many of these InVehicle Information Systems (IVIS) involve complex multimodal interactions to perform the secondary tasks. Surprisingly little is known about how these complex multimodal IVIS interactions impact a driver’s workload. Given the ubiquity of these systems, the current research sought to address three interrelated questions concerning this knowledge gap. First, which task types are most distracting and what are the sources of distraction (e.g., visual/manual vs. cognitive)? Second, what is the workload associated with different modes of IVIS interaction (e.g., center stack, auditory vocal, center console)? Third, we directly compared the IVIS interactions supported by different OEMs to determine the bases for any differences in the workload associated with their use. Depending on the availability of the IVIS features in each vehicle, our testing involved an assessment of up to four task types (audio entertainment, calling and dialing, text messaging, and navigation) and up to three modes of interaction (e.g., center stack, auditory vocal, and the center console). Three additional tasks were used to facilitate the assessment of the visual/manual and cognitive demands of the IVIS interactions. The first was the single-task baseline where participants drove the vehicle without any secondary-task interaction. The second, a cognitive referent task, was a concurrent auditory/vocal N-back secondary task that placed a high level of cognitive demand on the driver without imposing any visual demands (Mehler, Reimer, & Dusek, 2011). The third, a visual referent task, was a concurrent Surrogate Reference Task (SuRT, ISO TS 14198), which placed a high level of visual/manual demand on the driver. The N-back and SuRT tasks were adjusted so that they were equivalent in difficulty when compared with the single-task baseline (i.e., Cohen’s d was 1.423 and 1.519, respectively). For each of the 30 vehicles tested in the current research, 24 participants were evaluated as they drove on a residential road with a posted speed limit of 25 mph. After familiarization with the road, the vehicle, the IVIS tasks, and the modes of interaction, testing commenced, with the order of the testing conditions counterbalanced across participants. A number of performance measures were obtained while participants performed the tasks including primary-task measures, secondary-task measures, and subjective measures. Primary-task measures included GPS data and video recording of the driver and the driving environment. Secondary-task measures were obtained using two variants of the Detection Response Task (DRT, International Organization for Standardization #17488). Subjective measures were obtained at the end of each condition using the NASA Task Load Index. The data collected from each participant provided a measure of cognitive demand, a measure of visual/manual demand, a subjective workload measure, and a measure of the time it took to complete the different tasks. These metrics were evaluated separately and also combined to provide an overall demand score for the different tasks, modes of interaction, and vehicles. These metrics were standardized relative to the high demand cognitive referent (i.e., the N-back task had a rating of 1.0) and the high demand visual referent (i.e., the SuRT task had a rating of 1.0). Using this integrated metric, task types, modes of interaction, and vehicles that had a rating between 0.0 (the demand associated with the single-task baseline) and 1.0 were easier than the high-demand referent and those with ratings greater than 1.0 were harder than the high-demand referent. This procedure also provided a metric for directly comparing different tasks, different modes of interaction, and different vehicles. Our analysis found that the IVIS task types differed in terms of visual and cognitive demand, with the audio entertainment task type being equivalent to the calling and dialing task type (the two most universal of IVIS tasks available in all 2017 model-year automobiles we tested). Text messaging, an IVIS feature found in 22 out of 30 vehicles we tested, was associated with a significantly higher level of demand than the former task types. Most demanding of all was destination entry for navigation, an IVIS feature that was available in 12 out of 30 of the vehicles we evaluated. The navigation task type had an overall demand that was more than two times that of the high demand referent. Second, we found that the overall workload associated with each mode of IVIS interaction was greater than the high workload referent. Interactions using the center stack were significantly less demanding than auditory vocal interactions, which were less demanding than center console interactions. Interestingly, using voice-based commands to control IVIS functions resulted in lower levels of visual demand than the SuRT task. However, the benefits of reduced visual demand were offset by longer interaction times. Auditory vocal interactions took significantly longer than any other IVIS interaction (an average of 30 seconds in our testing). Finally, our analysis found surprisingly large differences between vehicles in the overall demand of IVIS interactions. Seven of the 30 vehicles received an overall rating significantly below 1.0 (i.e., a moderate level of overall demand). Eleven of the 30 vehicles received a score that did not differ from the high demand referent (i.e., a high overall demand score). Twelve of the 30 vehicles scored significantly above the high demand referent (i.e., a very high overall demand score). On the whole, vehicles in the latter category tended to have higher levels of demand on cognitive, visual, and subjective measures as well as longer interaction times. The vast majority of the IVIS features and functions in the vehicles we evaluated were unrelated to the task of driving (or, in the case of destination entry to support navigation, could have been performed before the vehicle was in motion). Many had cumbersome human-machine interfaces with design inconsistencies that lead to high levels of workload. In fact, many IVIS interactions were associated with high levels of cognitive and visual demand with long interaction times. For example, 83% of the vehicles with a very high overall demand offered destination entry for navigation while the vehicle was in motion, an IVIS task we found to produce high levels of workload. Our research provides empirical evidence that the workload experienced by drivers systematically varied as a function of the different tasks, modes of interaction and vehicles that we evaluated. Our objective assessment suggests that many of these IVIS features are too distracting to be enabled while the vehicle is in motion. This is troublesome because motorists may assume that features that are enabled when they are driving are safe and easy to use. Greater consideration should be given to what interactions should be available to the driver when the vehicle is in motion rather than to what IVIS features and functions could be available to motorists. (Author/publisher)
Samenvatting