U.S. traffic fatalities started rising 2 years ago after several years of declines. Psychologists around the world are looking for ways to improve traffic safety.
By Stephanie Pappas Date created: June 1, 2022 17 min readVol. 53 No. 4
Print version: page 46
Pappas, S. (2022, June 1). With traffic deaths on the rise, psychologists are being called on to make driving safer. Monitor on Psychology, 53(4). https://www.apa.org/monitor/2022/06/feature-traffic-safety
In the early months of the coronavirus pandemic, when roads were eerily clear, an underground subculture of street racers did the furthest thing from staying at home: They began launching attempts to speed from New York to Los Angeles in record time.
These races, called Cannonball Runs, date back to the 1970s, but empty roads enabled audacious driving. Drivers beat each other’s New York to LA records at least three times in 2020, averaging over 100 mph during their trips, with some reaching max speeds of 175 mph, according to Road & Track.
These Cannonball racers are extreme, but they are hardly alone in taking a riskier approach to the road during the pandemic. Though vehicle miles traveled decreased by 11% in the United States in 2020, traffic fatalities rose 6.8%, according to the National Highway Traffic Safety Administration (NHTSA). An estimated 38,824 people died. The trend continued in the first nine months of 2021, with deaths rising 12% compared with the same period in 2020. This was the biggest percentage jump in the year-to-year 9-month statistics ever recorded. These numbers represent a reversal of the decreasing trend in traffic deaths seen between 2016 and 2019, according to NHTSA data, and they are all the more striking considering that economic recessions like the one in 2020 typically reduce traffic fatalities. They’re also uniquely American, as most other high-income countries have reported fewer traffic deaths since the pandemic began.
The rising fatalities seemed to be caused by what University of Utah cognitive neuroscientist David Strayer, PhD, calls the “four horsemen of death.” Together, they are speed, impairment, distraction, and fatigue, the human foibles behind more than 90% of vehicle crashes. All, experts say, can be worsened by relentless cycles of pandemic stress.
“People’s brains are not perceiving information and processing emotion in the way that they did prior to the pandemic,” said Kira Mauseth, PhD, a clinical psychologist at Seattle University who studies disaster behavioral health. “People might be a little bit more impulsive, they’re a little bit less regulated, they might not be considering consequences.”
Strayer’s four horsemen aren’t going to disappear as pandemic stresses ease, though. They are persistent contributors to crash deaths and injuries, and the only question is to what level they’ll continue to kill. Around the world, psychologists are working to understand who is most at risk and why, studying everything from basic perceptual processing to cognitive biases to the way the environment can make matters better (or worse). The federal government is funding these efforts with the U.S. Department of Transportation’s National Roadway Safety Strategy, unveiled in January, which highlights the need for more research into interventions against risky behavior (U.S. Department of Transportation, 2022).
Research is starting to hint at some of the reasons for the initial jump in fatalities in 2020. A February 2022 report found that the people who reduced their driving the most during the lockdown phase of the pandemic were disproportionately middle-age and female, a relatively safe group of drivers (Tefft, B. C., et al., Self-Reported Risky Driving in Relation to Amount of Driving During the COVID-19 Pandemic (PDF, 148KB), AAA Foundation for Traffic Safety, 2022). The 4% of drivers who started driving more during this period were largely young and male, the demographic statistically most likely to engage in risky driving behavior. The American Automobile Association (AAA) also found that those who drove more during the early pandemic were more likely than average to report recent risky driving behaviors, such as driving without a seat belt or speeding.
Speed on the road is among the deadliest of the four main human foibles that cause accidents, contributing to an estimated 11,258 vehicle-related fatalities in 2020, according to NHTSA data. These dangers are not evenly distributed: In urban areas, high-speed commuter routes often cut through low-income neighborhoods and neighborhoods with high proportions of residents of color. An analysis of Washington, D.C., traffic deaths between 2014 and 2021, for example, found that the two poorest wards, both of which are majority Black, had half the city’s road fatalities despite being home to only a quarter of the population (Lazo, L., et al., Washington Post, February 23, 2022).
Psychologists have found both perceptual and cognitive biases that nudge people toward unsafe speeds, said Ola Svenson, PhD, a psychologist and head of the Risk Analysis, Social and Decision Research Unit at Stockholm University in Sweden. Drivers overestimate how much time they’ll save by speeding and grossly underestimate the increased accident risk at higher speeds (Applied Cognitive Psychology, Vol. 23, No. 4, 2009). They also underestimate how long it takes to stop at high speeds. In one study, Svenson and his team presented participants a scenario in which a child runs in front of a car driving 18 mph. At that speed, the driver can slam on the brakes and just avoid hitting the child. What then, the participants were asked, would happen if the driver were going 25 mph in this scenario?
The typical estimate, Svenson said, is that the car would hit the child while it was traveling at a speed of 12 mph. But this is a dramatic underestimate—in fact, the car would be going 23 mph when it hit the child (Accident Analysis & Prevention, Vol. 45, 2012). The researchers discovered that they could reduce this bias by providing information about how far a car travels during the split second it takes a driver to react, a finding with implications for public education campaigns (Accident Analysis & Prevention, Vol. 58, 2013).
Because many of these processes occur without much conscious thought, researchers also advocate for engineering the environment to reduce the likelihood of a crash. Crashes almost never have a single cause, says Ann Williamson, PhD, an emeritus professor at the University of New South Wales in Australia who has long studied human factors in road safety, and some roadways make it difficult for drivers to even discern what constitutes the correct behavior and what constitutes risky behavior. Visually cluttered roads, confusing signage, and broad thoroughfares that practically beg drivers to stomp on the accelerator can encourage behaviors that raise risk.
“We need to acknowledge that, yes, people will make errors, but when those errors are made by the system itself, we need to change the system,” Williamson said. Design choices like medians, trees, and cycle lanes can obstruct drivers’ views of the horizon and move their focus close to the front of their cars, encouraging more cautious driving. (See “Curbing the Need to Speed,” April 2018 Monitor).
Situational factors can increase risk-taking, and the early pandemic seems to have set the stage. Under stressful circumstances, “people are going to do things that might be considered risky or out of character to feel good, to feel alive,” Mauseth said.
Lack of enforcement also played a role during the early pandemic. In Australia, police reduced speed checks to focus on enforcing travel bans between states, says Mark King, PhD, a traffic psychologist at the Queensland University of Technology Centre for Accident Research & Road Safety (CARRS-Q). Natalie Watson-Brown, PhD, also at CARRS-Q, happened to be surveying young drivers about road safety behavior when the pandemic first hit and saw an increase in respondents saying they were more likely to break the law because they knew they were less likely to be caught.
Street racers and stunt drivers also likely saw an opportunity during lockdowns. Evelyn Vingilis, PhD, a developmental and clinical psychologist in the Department of Family Medicine at Western University in Ontario, Canada, has long studied these particularly egregious forms of rule-breaking. It’s hard to know how often street racing causes collisions or fatalities, given that racers aren’t likely to admit to racing, Vingilis said, and most jurisdictions don’t include street racing as a standard code on collision report forms. But Vingilis’s research finds that the practice is fairly common around the world, especially among young men, with between 38% and 69% of adolescent males and men in their early 20s reporting recently racing other cars on the road (Traffic Injury Prevention, Vol. 10, No. 2, 2009).
The evidence so far suggests that education doesn’t reduce racing, Vingilis said—after all, the danger can be part of the thrill. Strict penalties appear to be key to reducing this type of risky behavior. In 2007, Ontario introduced new legislation that made driving more than 50 kilometers an hour (31 mph) over the speed limit a serious offense, punishable by an immediate license suspension, car impoundment, and heavy fine or jail time. After that law was put in place, “we saw, on average, about 58 fewer young men per month were injured and killed in motor vehicle collisions,” Vingilis said.
Aggression on the road often overlaps with risk-taking in other ways, such as a willingness to drive while impaired. Daniel Bradford, PhD, a clinical psychologist at Oregon State University, has found that alcohol dampens people’s reactions, such as anxiety, to unknown stressors far more than it does to known stressors (Clinical Psychological Science, online first publication, 2022). This has potential implications for both the pandemic and drunk driving, Bradford said. Many of the stressors people faced during the height of the coronavirus crisis were unknown: Will you be exposed to the virus if you go shopping today? If you catch COVID-19, will it be a fever or death? And alcohol’s unique ability to suppress stress about unknowns may have been one reason that alcohol consumption rose in the United States in 2020 (Pollard, M., et al., JAMA Network Open, Vol. 3, No. 9, 2020). Likewise, the dangers one might face while driving drunk are highly variable, Bradford said. You might get pulled over or crash, but you also might get home without incident. If alcohol dampens reactions to these kinds of unknown stressors, it could contribute to decisions to get behind the wheel after imbibing.
Drunk driving has declined precipitously over the decades, but it still played a role in 11,654 traffic deaths in the United States in 2020, according to NHTSA data. Marie Claude Ouimet, PhD, a research psychologist at the Université de Sherbrooke in Quebec and director of the Road Safety Research Network of Quebec, and her colleagues have found that although there is overlap, there are differences between people with a history of drunk driving compared with those who drive recklessly in other ways (PLOS ONE, Vol. 11, No. 2, 2016). Speeders tend to be high in sensation-seeking, risk-taking, disinhibition, and poor decision-making. Repeated driving while impaired (DWI) offenders are high in disinhibition and alcohol misuse. People with a history of both DWI and other reckless driving are high in substance misuse and sensation-seeking, low in agreeableness, and tend to be reward-sensitive.
Ouimet and her colleagues are trying to use this information to tailor interventions to repeat offenders. They’ve found, for example, that brief motivational interviewing reduces future arrests in young and early middle-age drivers with a DWI arrest compared with a typical education and advice intervention (Alcoholism, Clinical and Experimental Research, Vol. 37, No. 11, 2013). The researchers are currently working to evaluate Quebec’s current program of targeting interventions to subgroups of offenders.
“For a long time, a lot of the thinking was that all of the individuals who drink and drive have a severe alcohol use disorder. What we see in reality is that is not necessarily true,” Ouimet said. Some DWI offenders struggle with periodic binge-drinking and poor decision-making, she said; for others, drunk driving is part of a larger constellation of reckless behavior. Specific targeted treatment may promote behavior change in different subgroups of offenders (Alcoholism, Clinical and Experimental Research, Vol. 43, No. 2, 2019).
The pandemic has been an exhausting experience for many, especially for front-line workers pulling grueling shifts. The resulting fatigue can be as dangerous as driving intoxicated, said the University of New South Wales’s Williamson. “When you’ve done a double [shift] and you’ve suddenly found that it’s been 16, 17, 18 hours since you’ve last slept and it is 2 o’clock, or 3 or 4 in the morning, you’re really asking for problems,” Williamson said. “It is highly risky.”
The NHTSA reported 633 drowsy-driving fatalities in 2020, but identifying fatigue as the cause of an accident can be difficult unless the driver clearly fell asleep at the wheel. Drew Dawson, PhD, a cognitive psychologist at CQUniversity Australia, has suggested a new way of classifying accidents based on a sliding scale of likelihood that fatigue played a role (Sleep Medicine Reviews, Vol. 42, 2018).
Research suggests that people know fatigue is bad for their driving abilities, but they don’t always respond effectively. A qualitative study of nurses who worked night shifts found most responded to drowsiness with less effective strategies, such as listening to music in the car (Smith, A., et al., International Journal of Nursing Studies, Vol. 112, 2020). Few tried the more effective option of napping before leaving work.
It’s not just about sleep, though, Williamson said. Human brains are poorly designed to pay close attention to monotonous tasks over extended periods of time. This can be a problem for long-distance truck drivers, some of whom saw the rules governing their hours of service relaxed in 2020 because of the need to keep essential goods moving. The challenge, Williamson said, is finding activities that effectively engage the brain without distracting the driver from the road. Researchers in Israel have found that trivia seems to fit the bill, increasing alertness and reducing driving deterioration compared with other mental tasks (Oron-Gilad, T., et al., Accident Analysis & Prevention, Vol. 40, No. 3, 2008). Such findings have led Australian authorities to put up trivia-game road signs on barren stretches of highway. (Sample question: “What is a monotreme?”)
Still, task-related driving fatigue is a difficult problem to crack, Williamson said. Brain games get drivers only so far down the road, and long-distance driving is still mentally exhausting. “I don’t think we’re there yet,” she said. “We’ve still got some work.”
While some drivers are struggling with fatigue on the roads during the pandemic, many are also likely contending with higher-than-usual levels of distraction. Distracted driving contributed to 3,142 deaths in 2020, according to the NHTSA.
Many front-line workers, especially in medicine, have reported intense stress on the job. Research indicates that work stress can spill over into the commute, leading to rumination while driving and, subsequently, riskier driving behaviors (Journal of Occupational Health Psychology, No. 25, Vol. 4, 2020). “If you’re thinking about work too much, you’re not paying as much attention to what’s happening on the road,” said Katrina Burch, PhD, an industrial and organizational (I/O) psychologist at Western Kentucky University, who led the research.
The relationship between work stress and road safety can be quite complex, said Charles Calderwood, PhD, an I/O psychologist at Virginia Tech who just launched a project looking at work stress, strain reactions, and commuting behavior with funding from the National Institute for Occupational Safety and Health. “Challenge stressors,” or difficult but attainable and satisfying problems people solve in their workday, may make people more alert or energetic on the drive home, while more negative stressors might not, Calderwood said.
Stress can also act indirectly on driving. For example, pandemic stress might cause poor sleep, which in turn could cause more crashes, Calderwood said. “Our work tries to be really dynamic in how we look at these processes over time,” he said.
It’s not just work stress, said Jing Feng, PhD, a cognitive psychologist at North Carolina State University who studies attention and driving. When a driver’s mind wanders, response time slows, meaning that unexpected situations can rapidly turn deadly, she said (Proceedings of the Human Factors and Ergonomics Society Annual Meeting, Vol. 59, No. 1, 2016). COVID-19 made many ordinary tasks more stressful, potentially increasing the amount of mental distraction drivers experience. “I think this pandemic has likely intensified some of the challenges,” Feng said.
Those challenges were stark to begin with. Not only do drivers carry around pure distraction in the form of smartphones, but cars are also now increasingly preloaded with features that divert attention from driving, said the University of Utah’s Strayer. “You have this explosion of technology,” he said. “The cars are coming equipped with touchscreen displays, heads-up displays with projections onto the windshield, voice control . . . It’s like distraction on steroids.”
There are NHTSA guidelines for automakers for in-car displays, Strayer said, but they’re voluntary and rarely followed. Strayer and his team have been working to develop benchmarks for distraction to evaluate vehicle design. They’ve found that some tasks, like using speech-to-text technology or voice command, take a surprising cognitive load (Cognitive Research: Principles and Implications, Vol. 16, 2016). Users sometimes remained distracted from the road for up to 27 seconds after disengaging with speech-to-text. NHTSA guidelines state that users should not have to look away from the road for more than two seconds at a time while interacting with a car’s technology, and no more than 12 seconds in total, but participants in Strayer’s studies sometimes spent up to two minutes fiddling with things like in-car navigation systems.
Strayer and his team are now consulting with the Australian Automobile Association, which hopes to develop a 5-star rating system for in-vehicle technology. Right now, Strayer said, there is no way for consumers to easily compare the usability of cars’ displays like they might fuel efficiency or crash ratings. “Maybe you get a 2-star rating saying, ‘This car is really nice and fun to drive, but the electronics in the car are extremely challenging,’” Strayer said.
Psychologists are also working to test maximally effective, minimally distracting tech in cars. Benjamin Wolfe, PhD, and Anna Kosovicheva, PhD, psychologists at the University of Toronto Mississauga in Canada, have been testing alert systems that warn drivers of a hazard, such as an animal in the road. They’ve found that simple alerts, such as a flashing red bar across the dash, are just as effective as more complex ones, like a box projected around the object on the roadway (Cognitive Research: Principles and Implications, Vol. 6, No. 80, 2021). Alerts like these can buy people about 60 milliseconds over a no-alert condition, Wolfe said, which is enough time to travel half a car length at highway speeds.
Better tech, from seat belts to air bags to driver-alert systems, has made cars safer. And increasing vehicle automation could potentially reduce the killing power of speed, impairment, distraction, and fatigue.
But automation can also be a crutch, and it is one that psychological scientists worry about. A car that does most of the driving itself, requiring a driver to take over only in times of rare emergencies, would be a very unsafe car indeed given the human inability to remain alert while not engaged. As cars do more of the work, psychologists will need to be there to push back against automation that makes roads more dangerous. (See “Along for the Ride,” January 2015 Monitor).
This is especially true when considering vulnerable road users, such as pedestrians, bikers, and motorcyclists. These users have become an increasing share of motor vehicle deaths for a variety of reasons, including the features of cars that make drivers safer, such as higher, heavier vehicles. There is also an increasing share of older pedestrians on the road because of the aging of the population, said CARRS-Q’s King: “Once you’re older, it’s actually quite easy for a knock that would be just some bruising for a young person to be life-threatening.”
Many cities have committed to Vision Zero, an effort to eliminate all traffic deaths and severe injuries by designing roads and policies to lessen the likelihood and severity of crashes. This focus on safety may put psychologists in more demand than ever. Preliminary research by Wolfe and Kosovicheva has found that when dangerous situations—such as a person walking into traffic—become rare, drivers become less likely to react quickly and decisively. In other words, the safer roads get, the harder it becomes, cognitively speaking, to prevent every accident. “We can do all of this work in infrastructure and tech and potentially in training,” Wolfe said, “but we’ve got this little sliver that we need to figure out how to address separately as a cognitive problem for drivers.”
Information as a source of distraction
Federal Highway Administration Research and Technology, 2015.
