Effects of an 'Expert Mode' Voice Command System on Task Performance, Glance Behavior & Driver Physiology

Abstract

Multi-function in-vehicle interfaces are an increasingly common feature in automobiles. Over the past several years, these interfaces have taken on an ever-greater number of functions and the ways in which drivers interact with information have become more complex. Parallel with these technical developments, interest in ensuring that these systems minimize demand placed upon the driver has also increased. Voice command capability has become a popular and desirable feature, as interacting with a vehicle interface through auditory/vocal interactions is often hypothesized to allow the driver to keep their eyes on the road and hands on the wheel. However, research has shown that production level voice command systems may still impart considerable visual demands on the driver [18]. These demands might be due in part to screen displays associated with extensive confirmatory dialogue and the driver's desire for visual confirmation that commands were accurately recognized. This study extends this work by comparing the default mode of a production voice system with an "Expert" mode which streamlines tasks by removing several confirmatory steps. We found that, although the use of the Expert mode significantly reduces overall task completion time, it has no appreciable effect on the amount of visual engagement; drivers still glance off the road for durations that are consistent with the Default mode. Implications for interface design and driver safety are discussed.

A Link Between Trust in Technology and Glance Allocation in On-Road Driving

“…The driver's face was video-recorded at 30Hz. This recording was used to manually code the driver's eye glances, as described in Reimer et al (2014).…”

“…Sawyer et al (2016) and Reimer et al (2014) showed that despite reduced glances off-road relative to use of a visualmanual human-machine interface (HMI), voice-based interactions with HMIs often have a significant visual component. Four glance metrics were considered, as detailed in Table 1.…”

“…When drivers interact with an in-vehicle device, they divide their glances between driving related locations, such as the road and mirrors, and the device (Reimer et al, 2014). During glance time off-road, drivers may miss events on-road.…”

Toward an Antiphony Framework for Dividing Tasks into Subtasks

“…This approach is more time intensive, and suffers from the entropy that human decision-making and response time may inject. Double coding, and use of interrater reliability scores is highly advised (as explored in the context of glance coding in Reimer et al, 2014). Subtask data can, in some cases, involve subsecond epoch lengths, and it is important to consider the synchronicity of clocks responsible for coding various aspects of human and machine performance (for an expanded discussion, see Sawyer et al, 2015).…”

Modeling of Stimulus-Response Secondary Tasks with Different Modalities while Driving in a Computational Cognitive Architecture

“…Many experimental studies have investigated the effect of different types of stimulus-response tasks on driving performance (e.g., Angell et al, 2006;Shutko, Mayer, Laansoo, & Tijerina, 2009;Young, Hsieh, & Seaman, 2013;Reimer et al, 2014b). According to the literature, visual and auditory modalities are two of the most frequently used information presentation channels in the in-vehicle secondary tasks, whereas manual and speech (or verbal) input techniques are the most common responding methods.…”