Washington -- Undertaking complex mental tasks can reduce a driver's ability to detect visual targets by as much as 30 percent. New research on driving in real traffic confirms that mental workload can interfere with the capacity to detect visual targets, discriminate among them, and select a response. Higher-level mental tasks take attentional resources away from the road, resulting in those all-too-familiar post-accident reports: "I didn't expect it" or "I saw it too late." These findings appear in the June issue of the Journal of Experimental Psychology: Applied.
Already knowing that external distractions divert drivers, the authors – at the Dirección General De Tráfico, Spain's Public Administration for Traffic Safety and the Universidad Complutense in Madrid -- studied internal distractions produced by the driver's own thoughts or cognitive activity unrelated to the task of driving. "It is easy to understand how one cannot see because of not looking," they explain, "but it is less obvious to explain how one looks but does not see."
The two-psychologist team studied 12 adults who drove for about four hours on the highway north from Madrid. They used a standard Citröen with an unobtrusive eye-tracking system that allowed them to study gaze and ocular fixation for signs of attention and distraction as drivers performed certain mental tasks. Drivers were additionally tested by an automatic system that periodically flashed spotlights in the driver's visual field, to which drivers responded using buttons placed ergonomically near the steering wheel. The researchers analyzed how drivers scanned the road scene, including use of the speedometer and mirrors. Researchers also measured how often the drivers glanced at the flashing spotlights, to identify them before they responded.
The effects of the performance of several mental tasks were compared with the effects of ordinary driving (the control condition). After being asked to attend and keep the information in mind, drivers listened to recorded audio messages with either abstract or concrete information (acquisition task). Next, drivers had to freely generate a reproduction of what they had just listened to (production task).
Although the more receptive tasks – listening and learning -- had little or no effect on performance, there were significant differences in almost all of the measures of attention when drivers had to reproduce the content of the audio message they had just heard.
Drivers also performed other tasks, either live or by phone. One was mental calculus (mentally changing between Euros and Spanish pesetas) either with an experimenter in the car, talking to the driver, or with the driver speaking by hands-free phone. One was a memory task (giving detailed information about where they were and what they were doing at a given day and time). Both tasks produced remarkable distraction effects.
"When performing complex mental tasks," the authors say, "the percentages of detected targets and/or correct responses decreased significantly." Drivers glanced at the targets less frequently, and gave a higher percentage of responses without directly looking in their direction. Also during these tasks, the targets (if looked at) were detected later and glanced at for less time, which would cause poorer identification.
Say the authors, "Some tasks showed a reduction in detection probability of almost 30 percent with respect to the control condition, something that is practically meaningful as an estimate of the increased risk of distraction errors hypothetically leading to traffic conflicts or accidents."
Driver errors appeared to derive from deficient target perception and/or identification rather then from a problem with decision rules and/or response performance. In other words, mentally distracted drivers still know how to drive, but don't see things well or fast enough to safely use their skills.
In the experimental variation that examined the impact of hands-free phone conversation, message complexity made the difference. The relative safety of low-demand phone conversation -- if hands-free and voice-operated --appeared to be about the same as that of live conversation. Thus, the authors say that because they demand too much attention, "Complex conversations, whether by phone or with a passenger, are dangerous for road safety." The participants only answered phone calls; the study did not include dialing.
The authors urge further study to assess the impact of another result, the spatial gaze concentration accompanied by a marked reduction in the speedometer and rearview mirror inspection, which "can be interpreted as a reduced situation awareness but also as an example of a balance to optimize visual resources."
The results are important, the authors conclude, for "the evaluation of the potential impact of in-car devices, the improvement of intelligent vehicle-user interfaces, and the issue of how to present information to drivers with minimal interference."
The findings also confirm that the risk of endogenous distraction (internal to the driver, as in complex thought) is at least as relevant as exogenous distraction (external, as in a crying child in the back seat or an accident scene on the road).
The authors stress that because distraction cannot be directly submitted to surveillance and enforcement, safety measures "such as removing potential external distractors or restricting the use of in-vehicle devices, could be effective if users and legislators feel committed to the importance of attentional control for road safety." Without such commitment, they conclude, drivers will again adapt and underestimate the importance of attention and the risk of distraction, including their own mental activity while driving.
Article: "Mental Workload While Driving: Effects on Visual Search, Discrimination and Decision Making," Miguel Angel Recarte Goldarecena, Ph.D., Universidad Complutense, and Luis Miguel Nunes González, Ph.D., Dirección General de Tráfico; Journal of Experimental Psychology: Applied, Vol. 9, No. 2.
Materials provided by American Psychological Association. Note: Content may be edited for style and length.
Cite This Page: