Deaths of decision-making among young men

Key takeaways:

• The life expectancy of men in the United States is almost 6 years shorter than for women.
• Gender gaps are particularly stark in adolescence and early adulthood, when young men die at much higher rates from “deaths of decision-making”—homicide, suicide, overdoses, and car crashes.
• These deaths are not inevitable, but result from predictable cognitive errors like impulsivity, catastrophizing, and misreading others.
• These errors can be corrected through training to help teens shift from reflexive (System 1) to deliberate (System 2) decision-making.
• Chicago’s “Becoming a Man” cut violent-crime arrests by approximately 50% in an RCT, and similar CBT approaches reduced recidivism by around 20% when delivered by existing detention staff during downtime.
• Scaling programs at low cost is possible by repurposing “low value-added” time in youth-serving systems—especially public schools and detention facilities—for decision-making training and practice.
• Prioritizing these preventions would yield large “years of potential life saved,” narrowing the gender life-expectancy gap.

Women in America live nearly 6 years longer than men do. This gap in average life expectancy shows no signs of narrowing over time; if anything it is widening. Why is this? Common explanations include biology, the potential reluctance of men to go to the doctor, or the possibility that men are more likely to wind up in stressful or hazardous jobs. Yet none of these explanations fully explain the problem.

The problem comes much earlier in the life cycle: adolescence and early adulthood. Young men are far more likely than young women to die from homicide, suicide, drug overdose, and what the public health field euphemistically calls “unintentional injuries,” most of which are car crashes. There is a common thread that connects these: They are fatal outcomes systematically linked to predictable errors of reflexive decisions—in other words, they are deaths of decision-making.

Are deaths of decision-making simply an unavoidable fact of life? For a long time, everyone thought so. Neuroscientists told us that the sensation-seeking, risk-taking, peer-approval-needing part of the human brain develops in early adolescence, while the executive functioning capacity of the brain usually isn’t fully developed until the mid-20s. As one colleague put it, this is nature’s cruel joke: The brain’s “gas pedal” develops a full decade earlier than the “brake pedal.”

But while that neuroscience is true, it turns out deaths of decision-making are not an unavoidable fact of life. They are preventable, which opens the door to dramatically improving people’s (especially men’s) quality and quantity of life, and reducing life expectancy disparities between men and women.

New research in behavioral economics shows us how. While one reason deaths of despair get less common as people age is changes in brain architecture, there is another reason as well: experience. Through trial and error, people learn to better navigate the world. But trial and error can be a costly way for young men to learn in an American society full of fast cars, drugs, and guns. Behavioral economics shows we can accelerate the maturation curve by giving people more chances to learn through trial and error in simulated decision-making tasks. We can give people more decision-making “reps” in low-stakes settings where errors aren’t so costly.

Best of all, we have the potential to do that at both large scale and low cost, through our many social institutions that work with large numbers of young people and are filled with low value-added time. That time could be repurposed to more productive uses at low financial (and opportunity) cost. Those institutions include the public schools, which creates the potential for what public health calls “primary prevention,” and detention facilities, which creates the potential for “secondary prevention” among those whose detention itself indicates they would benefit from additional support.

The only thing we need to do is do it.

Bad decisions drive the gender gap in early deaths

Across almost every stage of life, men comprise a larger share of deaths than women. 

Figure 1

What causes these disparities? The charts below give us some clues. The first looks at mortality rates for men and women of all ages; we show mortality rates for both groups separately, rank-ordered by leading cause of death. Most of those are causes of death for the elderly—heart disease, cancer, strokes, Alzheimer’s, chronic lower respiratory diseases. Men typically have higher mortality rates than women, but only slightly so in proportional terms, usually not more than 10-20%. For some leading causes men even have lower rates.

Figure 2

The picture looks quite different for deaths to young people (ages 15-20), as shown in figure 3. The leading causes of death for men are not just slightly higher than for women, they are dramatically so. The number one cause of death for young men is homicide, with a mortality rate that is nearly seven times that of women (or, put differently, nearly 700% the rate of women). The second leading cause of death to young men is suicide, with a rate three times that of women. Fatal overdoses and car crashes are more than twice as common among young men.

Figure 3

These large disparities among the young are particularly important from the public health perspective of “years of potential life lost.” When (say) a 75-year-old dies of a heart attack, that person has already experienced the great majority of a normal lifespan, with the opportunity to enjoy decades of life, relationships, and experiences. Family and friends may grieve deeply, wishing for a few more shared years. But it is clearly far, far worse for both public health and family and friends when (say) an 18-year-old dies in a car crash or homicide or suicide. Decades of potential life are now no longer lived.

Put differently, a medical intervention that changes the odds of surviving a heart attack will mostly add a few years of life expectancy to people who have already lived long, meaningful lives. By contrast, anything we can do to reduce mortality to people in their teenage years has an outsized effect on potential years of life lost.

So what do we do with this insight? When we look at the leading causes of death that afflict young men, what connects them?

Consider suicide. Most spells of suicidal ideation last only a very short time—just a matter of hours. But the cognitive error in these acute spells is actually a common one: All of us tend to think the future will be just like the present. That’s why, for example, homes with pools sell for a lot more when the house happens to be shown on a warm, sunny day than on a cold, rainy one. This means a young person who feels like they have nothing to live for is not naturally inclined to realize that feeling will pass—even though the evidence shows that it will.

Consider homicide. While most people think murders (most of which are committed with guns) are premeditated, carried out either by morally bad people unafraid of the criminal justice system or economically desperate people simply trying to survive, most shootings are actually not that (See Ch. 5). They’re garden-variety arguments that could have been avoided or de-escalated but instead go in the other direction and end in tragedy because someone has a gun. Several years ago one of us was talking to a Chicago police detective about the most recent homicide he was working. He described a shooting in a South Side neighborhood nicknamed “Terror Town,” in which one teen got off the train and, walking down the sidewalk, accidentally stepped on another teen’s sneaker. He was asked to apologize, indicated that he wouldn’t, and then someone wound up dead. Here too, a common cognitive error is at work: All of us are imperfect at “mindreading” the intentions of others; but in this case, that ended in tragedy.

Consider drug overdoses, which seem to be increasingly common in a world in which fentanyl—incredibly lethal even in miniscule doses—is increasingly used to lace all sorts of other drugs. Why do young people choose to use drugs, given these risks? Surveys suggest that one reason teens use illicit substances is because they think everyone else does too (but the survey data also suggest that teens are wrong in this assessment—they think their peers are more supportive and involved with drugs than they actually are, victims of a “false consensus”).

Consider car crashes. Research shows that teens are much more likely to drive in risky ways when they have another peer in the car. Why? Every automobile outing is filled with opportunities to be risk averse versus risk seeking (“do I try to pass this truck in time to make my upcoming exit, or do I just stay where I am behind the truck and drive a bit slower than I’d like…”). All of us are prone to all-or-nothing thinking; for a teen behind the wheel this can show up as what’s called “catastrophizing,” or “nothing is worse than looking like a dweeb in front of my friends.” Despite the powerful hold this feeling can impose on someone’s mind, it turns out there is something worse than looking like a dweeb in front of your friends: wrapping your car around a tree or telephone pole.

These are all, in other words, deaths of decision-making. And now that we are finally starting to understand better how young people (and all people) make such decisions, we have new opportunities to do something helpful to prevent these deadly decisions.

The most important insight from behavioral economics and behavioral science over the past several decades is that we are all, essentially, of two minds. The key insight of the “dual self” model or “dual-systems model of cognition” is that the little voice in our heads, the type of thinking of which we are aware and that we normally think of as “thinking,” is just one of the two types of cognition our minds engage in. That conscious thought, what psychologists call “slow thinking” or “System 2” thinking (see for example Daniel Kahneman’s Thinking, Fast and Slow), is what gives us our power to be deliberate and rational and solve complicated problems. That’s what enabled us to double life expectancy around the world since 1900 and invent the internet and put a man on the moon.

But that type of System 2 thinking is also enormously mentally taxing. And because System 2 is so draining, our minds developed to engage in a second type of thinking as well that happens fast and effortlessly, below the level of consciousness—to automatically, almost involuntarily, deal with routine, low-stakes situations that we see over and over again. Psychologists call this “fast thinking” or “System 1” thinking. System 1 tries to form as coherent and complete a picture of the world as possible from information that is available (whether that’s volumes or mere scraps), and respond to familiar patterns without conscious thought. (How many doors have you opened today? Unless you’re reading this over breakfast, the answer is not zero and yet you have no idea yourself.)

But since nothing in life is free, the cost of what we gain in low-effort fast responses is, sometimes, accuracy. System 1 can sometimes make a mistake in its pattern recognition and overgeneralize a normally-useful response into the wrong situation.

System 1’s normally useful tendency to quickly assess a situation from whatever scraps of information are available helps us, for example, to slam on the brakes when a soccer ball bounces into the road while driving: We don’t stop to think “there might be a child chasing the ball”—rather, our brain automatically recognizes the hazard and acts before we even have time to think. But that same world-building feature of System 1 can feed the conclusion that the handful of people a kid has seen doing drugs at a party means everyone is doing it, increasing the odds of their own drug use and, in turn, overdose. Or it can create the presumption that the future will feel just like the present, which can lead to the sort of despair that often fuels suicide.

Another feature of System 1 is its normally-useful ability to quickly form a default “theory of mind” for others—that is, an assumption that what’s in another person’s head is the same as what’s in our own. This helps us immensely in day-to-day conversations: When we refer to a basketball or dog or the new mayor who is leading our city into a ditch, we don’t need to explain every single reference we make in order for the other person to understand. But sometimes we have private knowledge others don’t have, and it is only by engaging System 2 that we can adjust our default theory of mind to account for that fact. But because System 2 is mentally exhausting, we all tend to under-adjust, giving rise to something psychologists call the “curse of knowledge”—incorrectly assuming that others know what’s in our heads. The teen who just stepped on another kid’s sneaker assumes the person must know it was an accident—and so responds with indignation when the other person becomes belligerent. The situation escalates, and someone winds up dead.

Similarly, System 1’s tendency to engage in binary thinking—“is this situation fine or a total disaster?”—has been helpful for mankind over the long arc of evolution when physical danger was common and people needed to quickly distinguish dangerous situations from safe ones. But that same tendency towards binary thinking can lead to the sort of catastrophizing that causes people to perceive negative events even more negative than they are—like thinking nothing is worse than looking like a dweeb in front of your friends.

The good news is that because the sort of mistakes that all of our System 1’s tend to make have some predictable structure to them, we have a chance to do something about it.

Programs that teach better decision-making can save young men’s lives

In Chicago, a nonprofit organization called Youth Guidance created the Becoming a Man (BAM) program, which aims to help young people intervene with themselves before a challenging situation escalates into violence. In the first exercise teens do in BAM, called “the fist,” they’re divided into pairs. One teen is given a ball; the other has thirty seconds to get it. Almost all of them rely on force to try to complete the assignment: they try to pry the other person’s hand open, or wrestle or even pummel the other person. During the debrief that follows, a BAM counselor asks why no one asked for the ball. Most kids respond by saying their partner would have thought they were a punk (or something worse). The counselor then asks the partner what he would have done if asked. The usual answer: “I would have given it, it’s just a stupid ball.”

This exercise gets kids to see how, from the very beginning, their System 1 has made an assumption about what situation they were in—providing a low stakes way to get feedback that their System 1 made a mistake, and that their automatic assumptions are often wrong. This exercise, and the BAM program more broadly, helps teach participants to recognize that when there’s a high-stake situation, it’s worth engaging System 2 to ask “hey, what’s really going on here?”.

A large-scale randomized controlled trial of BAM conducted by our research lab, the Crime Lab at the University of Chicago, found that BAM participation led to a drop in violent crime arrests of nearly 50 percent—an extraordinary effect. BAM-like programs that use this same kind of training to help people recognize the cognitive errors their System 1 might be making (often called “CBT” or “CBI” programs) have also been shown to improve outcomes on other bad or risky decisions, like dropping out of school, or suicide. The evidence is clear: By changing how young people recognize and rely on System 1, we can change all sorts of decisions that can cause all these deaths of decision-making—without having to wait for people to reach full brain maturity at age 25 or 26.

The biggest challenge with any social program is scaling it; and BAM is no exception. While the success of BAM has led to millions of dollars of government investment in the program, it still has a cost. So in thinking about the potential of these System 1 interventions, figuring out low- or no-cost means for implementation is paramount.

A few years ago the Crime Lab worked with the Cook County, Illinois Juvenile Temporary Detention Center, which was training correctional staff to deliver this sort of program during the afternoon downtime when kids would have otherwise just been watching TV. That cost almost nothing—after all, the corrections officers were already there—and reduced recidivism by over 20%. Study after study has shown similarly encouraging results.

To really transform public health for young men and narrow gender gaps in life expectancy, we’ll of course have to reach far more than just young men in detention. But the finding from the detention setting is more broadly encouraging about the potential for scale: we just need to repurpose some of the vast amounts of low-opportunity-cost time we have in so many of our other youth-serving institutions, particularly our schools. After all, if detention staff can successfully deliver this sort of program at basically no extra cost, surely so can teachers. And the public school system is the mother of all social programs, serving over 50 million young people every year at an annual cost of over $900 billion.

Where is that low-opportunity-cost time in the public-school setting? Introspecting on our own public school experiences reminds us that virtually every American high school student is forced to take a high school health class at some point. What did we learn? To eat more broccoli and less candy. To avoid alcohol. To sleep more. To understand the human reproductive system. Maybe this made sense in the 19th century when universal public schooling was developed, but it’s hard to see how that makes sense in a 21st century where malnutrition is rare and internet access is common. By repurposing even a modest amount of that time—as little as 20 hours—the impacts on public health in the United States, and the life expectancy gap between men and women specifically, could be dramatic.

These examples point toward several promising, scalable directions for policy and practice:

• Use schools as a platform for prevention. Most students already take health class and repurposing even 20 hours of that time for practice in recognizing and interrupting reflexive (System 1) decisions could yield large public-health gains.
• Leverage existing staff and downtime. Detention officers, teachers, and youth-program staff can deliver brief cognitive-behavioral exercises with minimal new resources.
• Expand proven behavioral programs. Programs that incorporate cognitive behavioral exercises—including BAM and many others—can cut violence, dropout, and recidivism rates and can be adapted across settings.
• Invest in evaluation and replication. States and school districts should measure long-term outcomes like injury, incarceration, and years of potential life saved to guide scaling.

Conclusion

Deaths of decision-making among young men are an under-appreciated cause of the life expectancy gap between men and women in America. The underlying behaviors that lead to such deaths obviously have negative consequences even in non-fatal situations. The young men who are arrested for assault and battery or even attempted murder can spend years or even decades behind bars. The surviving victim of such interpersonal violence can spend years dealing with the trauma. Drug use can squander someone’s potential. Car crashes can leave people permanently injured.

While medical care is expensive, prevention turns out to be cheap—especially for deaths of decision-making, given that we have so many institutions that work with young people that have so much low-value-added, low-opportunity cost time. In contrast to the trillions of dollars the United States invests in medical care each year, as a society we do almost nothing that is intentionally oriented towards preventing deaths of decision-making.

For example, detention facilities, which disproportionately serve the young men who, through their behavior, are demonstrating that the decisions they make are problematic, all too often spend all afternoon sitting in front of a television instead of changing in ways that might save their lives upon release. The public schools implicitly hope that teaching you the periodic table of elements or the quadratic formula or the plot devices of Hamlet will also as a byproduct magically teach you decision-making itself. Few schools try to do that deliberately in any way. Just ask yourself: Who was your 9th grade decision-making teacher?

It doesn’t have to be this way. It shouldn’t be this way.