A Brave New Roadway: Are Autonomous Vehicles the Solution to Improving the Safety of Aging Road Users?

Walter R. Boot, Ph.D., Department of Psychology at Florida State University

First, the good news with respect to crashes: It’s getting better. Over the past 20 years we’ve seen reductions in fatal crash rates, with declines greater for older drivers. This is due in part to the improved crashworthiness of vehicles which differentially benefits older drivers due to age-related increases in fragility. Declines may also be related to increases in the health of older adults, allowing older drivers to better survive crash forces and recover from crash injuries. The fact that older adult crash rates are declining even for property-damage only crashes, however, suggests that declines in crash risk represent more than just a decrease in injury severity, but that the crash involvement of older drivers may be declining as well. The bad news, however, is that even if there is only a small increase in crash risk associated with advancing age, we can expect age-related crashes and injuries to persist as the number of older road users increases dramatically in the United States and around the world. By the year 2050 it’s predicted that 1 in 5 individuals in the U.S. will be 65 years of age or older, with the 85+ age group being one of the fastest growing population segments of the United States. This makes increases in crash risk faced by older road users an important and urgent problem.

One reason older drivers are at greater risk is increased fragility (some argue that this is the primary reason for increased risk). However, normal age-related changes in perceptual, cognitive, and motor abilities, as well as changes related to age-related disease processes, can result in a mismatch between the abilities of the aging driver and the demands of the driving task. As we age our ability to see small details and extract visual information from the periphery decline, we become more susceptible to glare, and we have more difficulty seeing under low light conditions. Our ability to react quickly, juggle multiple tasks, remember and mentally manipulate information, as well as a number of other cognitive abilities important for driving also decline. This mismatch between the abilities of the aging road user and the demands of the driving environment has the potential to make crashes more likely. It can also make driving uncomfortable, causing aging road users to limit the amount of driving they do, or even cease driving altogether. Driving cessation is an important issue to consider. A lack of mobility can result in social isolation, which can negatively impact health and well-being if suitable transportation alternatives are not available.

Researchers with the Institute for Successful Longevity (ISL) and the Center for Accessibility and Safety for an Aging Population (ASAP Center) at Florida State University have explored a number of solutions to reduce the mismatch between the aging road user and the demands of the driving environment to enhance the safety, comfort, and mobility of aging road users. These studies (funded by the Department of Transportation and the Florida Department of Transportation) have found, for example, that changing the geometry of an intersection, introducing new signals, and providing drivers with education in the form of “tip cards” are all strategies that can result in safer turning maneuvers for drivers of all ages. Changing the driving environment to realign task demands with the abilities of the road user has proven a fruitful strategy. However, another strategy that is being discussed more and more frequently these days has been to offload the driving task from the driver altogether (or most of it) to the vehicle itself.

The evidence is convincing that autonomous vehicles are coming to our roadways. Google’s fleet of autonomous vehicles, as of 2015, has driven over a million miles on public roads. This year, Ford CEO Mark Fields predicted that fully autonomous vehicles will be available to consumers by the year 2020. Tesla CEO, Elon Musk, has made similar predictions, suggesting that in the near future (2023) you may be able to fall asleep in your car and wake up at your destination. Daimler’s chairman Dieter Zetsche goes one step further, suggesting that by the year 2025 cars may be sold with optional steering wheels. While there are reasons to be skeptical of some of these predictions, it’s a safe bet that consumer autonomous vehicles will be on the roadway in the near future (but perhaps not as soon as 2020). The question raised here is whether autonomous vehicles will be an easy solution to improving the safety and mobility of aging road users.

Autonomous vehicles, no doubt, hold great promise for improving the safety and mobility of not just older adults, but also individuals with disabilities. Mismatches between the demands of the driving task and the ability of the road user mostly become irrelevant when the vehicle itself takes over the driving task. However, if we propose autonomous vehicles as a solution to age-related crash risk, benefits may be linked to whether or not older adults adopt this new technology. With many forms of technology there is a digital divide associated with age. For example, while 84% of adults in the United States use the Internet, only 58% of adults 65 years of age or older do so. Smartphone, tablet, and the adoption of other newer technologies lag substantially for older adults. Given this general trend, overcoming barriers to technology adoption may be key, which may include understanding attitudinal barriers to the adoption of autonomous vehicle technology. Adoption often depends on perceived benefits and costs (both financial costs, and the effort it costs to learn something new), and it would be useful to anticipate some of these factors that influence adoption in advance of autonomous vehicles being sold commercially. Financial costs may exceed what a retired older adult is willing or able to pay, with estimates of automation adding up to $10,000 on top of the cost of a new vehicle. The decision to adopt a new form of technology is a complex one, and understanding factors that influence this decision with respect to autonomous vehicles may be crucial in determining whether or not older adults will benefit from them.

Another issue that needs to be considered is that no automated system is perfect. For an excellent discussion of this issue by Raja Parasuraman and Christopher Wickens, see their appropriately titled article “Humans: Still Vital after All These Years of Automation.” It is likely that at some point automation will fail and the human driver will need to reengage with the driving task (though hopefully this is a rare occurrence). This, unfortunately, may reintroduce the opportunity for age-related crash risk. Will older adults be able to detect and react to automation failures as quickly as their younger counterparts? Will driving skill degradation occur after extended periods of automated driving, and might there be age-related differences in skill degradation? Will older adults experience a greater “task switch” cost when suddenly they are required to take control of the vehicle after an automation failure? All of these are issues that need to be understood if the goal is to eliminate age-related crash risk. Age-related risk may also arise if automation technology (and the training to use this technology) isn’t designed taking into account age-related changes in cognition and perception. If the design of automated vehicles, the displays that control them, and the training required for their use aren’t designed well for an aging population, this again may put older drivers at a disadvantage.

In sum, autonomous vehicles hold great promise, and have the potential to substantially reduce the more than 30,000 roadway fatalities that occur each year in the United States, and also reduce the differential crash risk of older drivers. Predictions of autonomous vehicles offered to consumers by 2020 may be optimistic, but it is likely that they will be available at some point in the near future. However, there are technological challenges, as well as a host of legal and regulatory issues that need to be resolved before we are likely to see many autonomous vehicles on the road. When we do, it may be tempting to think of autonomous vehicles as the solution to age-related crash risk, but it is unlikely to be a perfect or easy solution. Rather than make the researchers who study the issues related to older adult crash risk obsolete, it is likely that the transition to semi-autonomous and autonomous vehicles will keep them as busy as ever. Much of their focus may shift from the driver’s interactions with the roadway to his or her interactions with the vehicle’s automation system, with a special emphasis on recovery from automation failures. Research in this area will help ensure that autonomous vehicles are the optimal solution to preserving the safety and mobility of the rapidly expanding older adult population in the United States and around the world.