Automotive

Study suggests that HUD tech may actually reduce driving safety

Study suggests that HUD tech m...
Recent research suggests that HUD use may result in too much distraction for drivers and a subsequent loss of adequate concentration at times of crisis
Recent research suggests that HUD use may result in too much distraction for drivers and a subsequent loss of adequate concentration at times of crisis
View 1 Image
Recent research suggests that HUD use may result in too much distraction for drivers and a subsequent loss of adequate concentration at times of crisis
1/1
Recent research suggests that HUD use may result in too much distraction for drivers and a subsequent loss of adequate concentration at times of crisis

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 digitalimages onto the inside of a windshield, drivers can be alerted about everythingfrom a mundane incoming phone call to an imminent collision with a vehicle suddenlybraking ahead. The trouble with this range of activity, say U of Tacademics in a recent study on driver reactions, is that the very nature ofalerts for every occurrence isdistracting. And, rather than being able to take in all of this information,drivers actually have to split their focus between watching the road andinterpreting the cause of an alert.

"Drivers need to divide their attention to deal with this addedvisual information," said Department of Psychology professor Ian Spence, who led the research. "Notonly will drivers have to concentrate on what’s happening on the road aroundthem as they’ve always done, they’ll also have to attend to whatever warningpops up on the windshield in front of them."

To ascertain the effects ofextraneous information in a driver’s line of sight, professor Spence and histeam of students created two tests to measure the outcome. The first involvedvolunteers completing a number of computer-based tests in which they wererequired to say how many of a number of randomly organized spots were shown ona 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 ofa collision and a recommendation to make a turn," said professor Spence. "Otherwisecompeting 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 ofspots became much more difficult when a shape suddenly appeared.

"Observers made both judgements more slowly when the shape appeared amongthe spots by as much as 200 per cent," said professor Spence. "The two visual tasksinterfered with each other and impaired both reaction speed and accuracy. Missed warnings and slowed reaction times present real threats tosafety. Furthermore, this rivalry for the driver’s attention ismost 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

17 comments
Robert Walther
Obviously this heads up display is designed to give feedback as to what a (near future) self driving car is going to do. If the driver is ignorant of the rules, on the phone and has no clue as to how to react in an emergency situation, all of the tech display on earth is not going to help. Modern multi-lane, limited access, directionally separate driving is not conducive to teaching Grand Prix skills. Giving marginally competent operators more information becomes less useful as the amount of data increases. For 20+ computers analyzing billions of instructions per second and vehicle hardware reacting a thousand times faster than the human anchor, it is all in a day's accident free, self driving work.
zr2s10
So, I didn't see anything in there where they actually tested people using a heads up display while driving. Did I miss it? HUD info is complimentary to the task at hand, not just a random distraction, which is all these tests seemed to do. One of my cars has a HUD for just MPH and temp, and radio station when changing via steering wheel controls. You can also use a "night mode" that turns off the dash lighting. It greatly helps fatigue while driving at night, because my eyes can stay straight ahead. I wish my others cars had it, especially with the added info available these days. Especially one that could highlight deer at night from IR cameras. That's a big problem in my area.
Kpar
Military pilots have long suffered "information saturation", especially during combat. These selected people are highly trained to deal with multiple critical inputs and still are not immune. A major effort is still underway to simplify, as much as possible, the information needed to safely manage such an environment. Drivers (many poorly trained) are much more likely to be overwhelmed and distracted- after all, we can barely keep up with phone calls. Perhaps HUDs should be limited to "synthetic vision"- FLIR displays that enhance difficult to see hazards- and let the driver just have better senses.
cvx
This is too new. It takes time for the person to learn what's important and what is not. Situational awareness is different today with sensors doing the work that eyes and hearing used to.
FrankNitty II
A person has to learn instincts from a defensive and offensive driving aspect. All this is going to do is, make present drivers lazy and the future drivers come to rely on this technology. How about this...instead of creating technology that dums down the human for profit; create technology that will assist with humanity's development in life preservation and etc.
Bob Flint
In the first slide is the green arrow showing that if you achieve 124kph (speeding) & dangerous maneuver you can pass? But do it now? Most highway limits here are 100kph, occasionally 110, but not 124, might as well be on the safe side and do 140kph, oopps not what the screen recommended aaaaaaaaaaaahhh
Bob Flint
Ok never mind just saw the top of the speedometer, you are already speeding haha, NOW GET OUTTA THE WAY the car in front of you is much slower and has his brakes on!
JR
I agree with the study that too much information is distracting. But If I can select the information displayed it can be very helpful. I don't need to know what channel my radio is set to or what the outside temperature is though. Alerts like lane closings or TA can be very helpful as can extended sensors that warn of wildlife near/on the road. If I can't select what the HUD displays then it is detrimental.
4Freedom
Perhaps the engineers could limit the display to a safe amount of information, such as four types simultaneously, and let the user pick the four they would like to see? Personally, I wouldn't need anything fancy, just the stuff that's already on the dashboard, but up on the windshield instead.
Martin Hone
Obviously if information overload is a problem, reduce the amount of information. Driving is not a video game. Keep an eye out for the ever present dangers, or stay at home on your computer....