|
Article Excerpt
INTRODUCTION
Use of a cellular phone while driving creates a situation that may be unlike that of other driving distractions in its effect on driving. Although demographic data are limited on the number of accidents that have occurred while a driver used a cellular phone, the data that are available suggest that cellular phone use while driving increases accident risk. (For a review, see Cain & Burris, 1999.) Even in the laboratory, the negative effects of conversation on tasks that mirror driving can be measured reliably. A good example is recent work by Strayer and Johnston (2001), who found that participants performing a simulated driving task while conversing on a cellular phone were more likely to miss traffic signals and had longer reaction times to critical events than did participants who simply listened to a radio broadcast.
Much of the debate thus far has centered on finding direct evidence to prove that on-road cellular phone activity increases accident risk. The purpose of the current research is not to provide this proof but to explore the remaining question: What is the mechanism that leads to a decrease in performance on tasks such as driving that require the processing of visual information with a concurrent conversational task?
A brief review of the current literature suggests numerous possible causes of driving detriments with a concurrent speech task. For example, work by Lee, Caven, Haake, and Brown (2001) suggests that reaction time to the braking of a leading vehicle slows by 30% with a concurrent speech task. Work by Strayer, Drews, and Johnston (2003) and Strayer and Johnston (2001) suggest that conversation limits the ability of drivers to attend to foveal visual information, resulting in missing information that may be critical for active control. Recarte and Nunes (2000) observed participants driving in real traffic in an instrumented car that included an eye tracker to measure gaze. The drivers performed a verbal or spatial imagery task, such as repeating words with a particular letter or performing mental rotation of letters and responding to questions. These tasks led to a reduction in size of the spatial region observers inspected during driving (see also Crundall, Underwood, & Chapman, 1999, 2002; and Janelle, Singer, & Williams, 1999). Recarte and Nunes (2003) have also found negative effects of concurrent tasks on detection time for targets. Clearly, detriments to driving performance while performing a concurrent verbal task are attributable to a number of factors.
The purpose of the current work is to examine an additional possible mechanism that has been previously shown to relate to accident risk. Although it has been noted that eye movements change and foveal attention lapses if a conversational load is added, it is not known whether the distribution of attention changes as well. This is of interest because, as we will review, it has been shown that in older drivers, a reduction in the extent of spatial attention can account for a large proportion of the increase in accident risk.
The distinction between people's visual field and what they can actually process in the visual field was noted in early work by Mackworth (1965). He demonstrated that the addition of visual noise in a reading task resulted in reduced recognition accuracy for information in the periphery. Further research expanded on this finding, resulting in the distinction between the field of view that is determined by the physical nature of the observer's visual system and the "functional" (Sanders, 1970) or "useful" (Mackworth, 1976) field of view, which is a restriction of a person's field of view imposed by limits in his or her visual information processing capacity. For example, Williams (1982, 1985, 1995) showed that information that is above the threshold for detection and presented in the periphery can be missed if the observer is performing a concurrent mental task or making a judgment about information presented foveally. Clearly, what people see is not limited by their eyes but by cognitive operations they must simultaneously perform.
This effect is important for the consideration of how conversation may influence accident risk because reductions in the functional field of view have been linked to accident risk, at least in older drivers. This account arises from the observation that the increase in accident rates among older drivers seems to be linked to an increase in the likelihood of an accident at an intersection or a failure to yield the right of way (Carr, Jackson, Maddden, & Cohen, 1992; Morgan & King, 1995; Ryan, Legge, & Rosman, 1998). These accidents involve a failure to detect critical information in the visual periphery.
The attentional account of this increased risk in older adults is supported by a number of studies showing that decreases in the region of attention in which they process information is linked to an increase in accident risk (e.g., Ball, Owsley, Sloane, Roenker, & Bruni, 1993; Owsley, 1994; Owsley et al., 1998; Owsely, Ball, Sloane, Roenker, & Bruni, 1991; Sekuler & Ball, 1986). The study of Owsley et al. (1998), for example, tested 294 older drivers (age 55-87 years) on a variety of sensory measures (acuity, contrast sensitivity, stereoacuity, visual field sensitivity, and glare) and also assessed of the drivers' functional field of view. They found that a 40% reduction in the size of the functional field of view results in an increase in crash risk of about 2.3 times. Decreases in the functional field of view have also been linked to age-related driving risks in drivers with Alzheimer's disease (Reinach, Rizzo, & McGehee, 1997) as well as to increased driving risks when noise levels are above optimal (Scialfa, Kline, & Lyman, 1987).
Although one must be cautious in generalizing a mechanism used to account for older adult risk to risk induced by conversation, it is not unreasonable to consider such a possibility, especially when one understands how limits to the functional field of view are produced. To produce limits to the functional field of...
|
