Mental Workload of Voice Interactions with 6 Real-World Driver Interfaces (original) (raw)
Related papers
Mental Workload of Common Voice-Based Vehicle Interactions across Six Different Vehicle Systems
Founded in 1947, the AAA Foundation in Washington, D.C. is a not-for-profit, publicly supported charitable research and education organization dedicated to saving lives by preventing traffic crashes and reducing injuries when crashes occur. Funding for this report was provided by voluntary contributions from AAA/CAA and their affiliated motor clubs, from individual members, from AAA-affiliated insurance companies, as well as from other organizations or sources. This publication is distributed by the AAA Foundation for Traffic Safety at no charge, as a public service. It may not be resold or used for commercial purposes without the explicit permission of the Foundation. It may, however, be copied in whole or in part and distributed for free via any medium, provided the AAA Foundation is given appropriate credit as the source of the material. The AAA Foundation for Traffic Safety assumes no liability for the use or misuse of any information, opinions, findings, conclusions, or recommendations contained in this report. If trade or manufacturer's names are mentioned, it is only because they are considered essential to the object of this report and their mention should not be construed as an endorsement. The AAA Foundation for Traffic Safety does not endorse products or manufacturers.
Proceedings of the 7th International Driving Symposium on Human Factors in Driver Assessment, Training, and Vehicle Design : driving assessment 2013, 2013
Interaction with a voice-command interface for radio control, destination entry, MP3 song selection, and phone dialing was assessed along with traditional manual radio control and a multi-level audio-verbal calibration task (nback) on-road in 60 drivers. Subjective workload, compensatory behavior, and physiological indices of cognitive workload suggest that there may be both potential benefits and cautions in the implementation of a representative production level interface.
Evaluating Demands Associated with the Use of Voice-Based In-Vehicle Interfaces
Proceedings of the Human Factors and Ergonomics Society ... Annual Meeting, 2016
This panel addresses current efforts associated with the evaluation of demands associated with the use of voice-based in-vehicle interfaces. As generally implemented, these systems are perhaps best characterized as mixed-mode interfaces drawing upon varying levels of auditory, vocal, visual, manual and cognitive resources. Numerous efforts have quantified demands associated with these systems and several have proposing evaluation methods. However, there has been limited discussion in the scientific literature on the benefits and drawbacks of various measures of workload; appropriate reference points for comparison (i.e. just driving, visual-manual versions of the task one is looking to replace, etc.); the relationship of demand characteristics to safety; and practical design considerations that can be gleamed from efforts to date. Panelists will discuss scientific progress in the topic areas. Each panelist is expected to present a brief perspective followed by discussion and Q&A. Figure 1: Re-conceptualization of DVI demands: (a) "traditional view point" of visual-auditory-cognitive-psychomotor dimensions; (b) proposed conceptualization of modern multi-modal system demands
Additional Findings on the Multi-Modal Demands of “Voice-Command” Interfaces
SAE Technical Paper Series, 2016
This paper presents the results of a study of how people interacted with a production voice-command based interface while driving on public roadways. Tasks included phone contact calling, full address destination entry, and point-of-interest (POI) selection. Baseline driving and driving while engaging in multiple-levels of an auditoryvocal cognitive reference task and manual radio tuning were used as comparison points. Measures included self-reported workload, task performance, physiological arousal, glance behavior, and vehicle control for an analysis sample of 48 participants (gender balanced across ages 21-68). Task analysis and glance measures confirm earlier findings that voice-command interfaces do not always allow the driver to keep their hands on the wheel and eyes on the road, as some assume. Self-reported workload, task completion time, glance metrics, and error rates varied significantly across the tasks, highlighting the importance of evaluating a particular design characteristic for specific tasks and exercising caution in generalizing across this class of user interfaces. For example, the "one-shot" voice-command structure in the study vehicle was associated with very low workload and error rates for phone contact calling, but much higher values for destination address entry and POI selection. Total task and eyes-off-road time were also relatively high for the latter tasks. However, mean single glance duration was nominally lower and the impact on major steering wheel reversals was less during the voice command tasks than during manual radio tuning, suggesting the importance of considering the comprehensive demand of an interface relative to other tasks in which the driver might engage.
The Effects of Voice System Design Components on Driver Workload
Transportation Research Record: Journal of the Transportation Research Board, 2018
In-vehicle voice control systems are standard in most new vehicles. However, despite auditory-vocal interaction allowing drivers to keep their hands on the steering wheel and eyes on the forward roadway, recent findings indicate the potential for these systems to increase levels of workload and lead to lengthy interaction times. Although many studies have examined the distraction potential of interacting with in-vehicle voice control systems, more research is needed to understand the relationship between different system design components and workload. In this study, we investigate the role of system delay, system accuracy, and menu depth in determining the overall level of demand and interaction times on eight different 2017 model-year vehicles. Voice system accuracy was measured via playback of a pre-recorded sample of voice commands through a studio monitor mounted near the headrest. Menu depth and system delay were calculated by measuring, respectively, the number of interaction...
Transportation Research Part F: Traffic Psychology and Behaviour, 2019
Given the promise for auditory-vocal systems to be less distracting and safer to use than their visual-manual counterparts, automotive manufacturers are introducing an increasing number of voice assistant-like interfaces in vehicles. However, recent studies suggest using auditory-vocal systems can be mentally taxing for drivers, and require long interaction times. Low accuracy and menu complexity are believed to negatively affect the usability of the system, but a systematic analysis of the role of different design components on driver workload and user experience within a real-world setting is missing. This study investigates the role of voice system design components in determining levels of driver workload with participants driving twelve on-road vehicles and interacting with onboard infotainment systems. Menu depth, delay times and system accuracy are recorded for twelve different auditory-vocal systems, and one aggregate score is assigned to each system. Total delay time and menu depth were found to be significant predictors of task duration time and mental workload. Longer delay times also had a direct effect on lower usability ratings, measured by the System Usability Scale and sentiment analysis. Delay times recorded on seven additional vehicles were used to validate the regression model for task duration. This study provides findings of primary importance for researchers and automotive manufacturers to be used in the assessment and development of in-vehicle auditory-vocal systems.
The effect of cognitive load on disfluencies during in-vehicle spoken dialogue
2008
In-vehicle spoken dialogue systems are gaining increased interest by the automotive industry. They enable the driver to perform secondary tasks (i.e. tasks not related to driving the vehicle) without having to take her eyes off the road or her hands from the steering wheel. Dialogue systems also enable the driver to speak in a natural way, without having to memorize commands or navigate through a menu structure. It is however crucial to take the cognitive load of the driver into consideration, in order to be able to adapt the dialogue system accordingly. This paper presents a user study where spoken dialogues between drivers and passengers have been analysed to find out how spontaneous speech is affected by driving and carrying out other activities that increase the cognitive load of the user. The results indicate systematic changes in specific disfluency rates as the cognitive load increases.
This report assesses the extent to which key findings from our initial on-road study on driver interaction with a production version, invehicle voice command system replicate, as well as considering whether two differing approaches to introducing drivers to the driver vehicle interface (DVI) impact their pattern of interaction, including driving behavior. An analysis sample of 64 participants, equally balanced by gender across the four age groupings (18-24, 25-39, 40-54, and 55+) specified in the NHTSA (2013) visual-manual guidelines for DVI assessment, was evaluated during manual radio tuning, voice-command assisted radio tuning, and voice-command assisted navigation system interaction consisting of full destination address entry and route cancelation under actual highway driving conditions. The MIT AgeLab auditory presentation / vocal response n-back cognitive demand reference task was also presented. No statistically significant main effects of self-guided vs. structured training condition were found across the tasks as a whole, although clear advantages were evident in tasks requiring memorization of complex command syntax. The basic pattern of results seen in the first study (considering self-reported workload, physiological arousal, driving performance metrics, and glance behavior) largely replicated.
Cognitive Workload Using Interactive Voice Messaging Systems
Proceedings of the Human Factors and Ergonomics Society ... Annual Meeting, 2016
The current research sought to understand the sources of cognitive distraction stemming from voice-based in-vehicle infotainment systems (IVIS) to send and receive textual information. Three experiments each evaluated 1) a baseline single-task condition, 2) listening to e-mail/text messages read by a "natural" prerecorded human voice, 3) listening to e-mail/text messages read by a "synthetic" computerized text-tospeech system, 4) listening and composing replies to e-mail/text messages read by a "natural" voice, and 5) listening and composing replies to e-mail/text messages read by a "synthetic" voice. Each task allowed the driver to keep their eyes on the road and their hands on the steering wheel, thus any impairment to driving was caused by the diversion of non-visual attention away from the task of operating the motor vehicle.