Study suggests that HUD tech may actually reduce driving safety
Cruising at speed down the highway with a heads-up display (HUD) constantly feeding data into your line of sight can make anyone feel like a jet pilot on the road; totally in control of your vehicle and primed to avert any potential danger that comes your way. However, recent studies by the University of Toronto show that the HUD multi-tasking method of vehicle piloting may well not provide the extra margin of safety that we think it does. In fact, according to the researchers, it could be downright dangerous.
With HUDs displaying digital images onto the inside of a windshield, drivers can be alerted about everything from a mundane incoming phone call to an imminent collision with a vehicle suddenly braking ahead. The trouble with this range of activity, say U of T academics in a recent study on driver reactions, is that the very nature of alerts for every occurrence is distracting. And, rather than being able to take in all of this information, drivers actually have to split their focus between watching the road and interpreting the cause of an alert.
"Drivers need to divide their attention to deal with this added visual information," said Department of Psychology professor Ian Spence, who led the research. "Not only will drivers have to concentrate on what’s happening on the road around them as they’ve always done, they’ll also have to attend to whatever warning pops up on the windshield in front of them."
To ascertain the effects of extraneous information in a driver’s line of sight, professor Spence and his team of students created two tests to measure the outcome. The first involved volunteers completing a number of computer-based tests in which they were required to say how many of a number of randomly organized spots were shown on a screen as quickly and accurately as they could.
Added to this, in some tests a black-outlined square arbitrarily appeared and the subjects were told to report whenever they saw this too. This secondary stimulus was shown at the same time as the spots, but did not appear in all of the trials.
When recounting the results, the researchers realized that when only spots were displayed (without the box), the accuracy rate of spots detected was high among the participants. However, if the box was introduced when a small number of dots were on the screen, participants failed to notice the box around one in fifteen times on average. Adding more dots at the same time that the box appeared had a greater effect on this figure, increasing the miss rate to an average of one in ten times.
Additionally, increasing the number of spots, regardless of whether the box made an appearance or not, also diminished the accuracy of the dot count.
These results led the team to believe that attending to a primary task is diminished proportionately to the number of secondary tasks presented to the participants. In the real world, this indicates that drivers may not be responding to multiple stimuli efficiently enough to distinguish between imminent danger and mere information prompts. In other words, drivers presented with multiple alerts vying for attention can lead to information overload.
"It would be necessary to distinguish, for example, between warnings of a collision and a recommendation to make a turn," said professor Spence. "Otherwise competing warnings may be more dangerous than no warning at all."
For the second computer-based test, the researchers told the volunteers to identify shapes that appeared at random in another field of spots. This time, a triangle, a square, or a diamond were displayed at different times together with the spots, but unpredictably and without warning, just as would occur in real-life situations with HUDs warning of external dangers.
As had happened with the first experiments using large numbers of dots, the participants in the second set often failed to see the shape at all. They also frequently misidentified the shape and counting the number of spots became much more difficult when a shape suddenly appeared.
"Observers made both judgements more slowly when the shape appeared among the spots by as much as 200 per cent," said professor Spence. "The two visual tasks interfered with each other and impaired both reaction speed and accuracy. Missed warnings and slowed reaction times present real threats to safety. Furthermore, this rivalry for the driver’s attention is most likely to occur when the driving environment is demanding."
The results of these studies were recently published in the journal PLOS ONE.
Source: University of Toronto