By MARIA KONNIKOVA
WHAT do you do if, when you get to a subway platform, you see that it is already packed with people? Do you join the throngs to wait for the train, or do you shake your head and seek an alternative way to get where you’re going?
If you go the first route, you probably think that the crowd means there must not have been a train for some time and that one is imminent. If you choose the second, you’ve come to the opposite conclusion: It’s crowded, a train hasn’t come in a while, so it’s likely there’s some sort of problem — and who knows how long you’ll end up waiting. Better cut your losses and split.
When we think of self-control, we don’t normally see it in these terms — a reasoned decision to wait or not. In fact, the ability to delay gratification has traditionally been seen in large part as an issue of willpower: Do you have what it takes to wait it out, to choose a later — and, presumably, better — reward over an immediate, though not quite as good one? Can you forgo a brownie in service of the larger reward of losing weight, give up ready cash in favor of a later investment payoff? The immediate option is hot; you can taste it, smell it, feel it. The long-term choice is far cooler; it’s hard to picture it with quite as much color or power.
In psychological terms, the difference is typically seen as a dual-system trade-off: On one hand, you have the deliberative, reflective, cool system; on the other, the intuitive, reflexive, hot system. The less self-control you have, the further off and cooler the future becomes and the hotter the immediate present grows. Brownie? Yum.
But what if the reality is a little different? What if the ability to delay gratification is actually more like the commuter faced with a crowded train platform than like a dieter faced with a freshly baked treat? A failure of self-control, suggest the University of Pennsylvania neuroscientists Joseph W. Kable and Joseph T. McGuire, may not be a failure so much as a reasoned response to the uncertainty of time: If we’re not quite sure when the train will get there, why invest precious time in continuing to wait?
Mr. Kable, who has been working on the psychology and neuroscience of decision making for more than a decade, argues that the truth is that in real life, as opposed to the lab, we aren’t nearly as sure we’ll get our promised reward, or if we do, of when it will come.
“The timing of real-world events is not always so predictable,” he and Mr. McGuire write. “Decision makers routinely wait for buses, job offers, weight loss and other outcomes characterized by significant temporal uncertainty.” Sometimes everything comes just when we expect it to, but sometimes even a usually punctual bus breaks down or an all-but-certain job offer falls through.
When we set a self-control goal for ourselves, we often have specific time frames in mind: I’ll lose a pound a week; a month from now, I’ll no longer get cravings for that cigarette; the bus or train will come in 10 minutes (and I’ve committed to taking public transportation as part of lessening my carbon footprint, thank you very much).
But what happens if our initial estimate is off? The more time passes without the expected reward — it’s been 20 minutes and still nothing; I’ve been dieting for a week and a half now and still weigh the same — the more uncertain the end becomes. Will I ever get my reward? Ever lose weight? Ever get on that stupid train?
In this situation, giving up can be a natural — indeed, a rational — response to a time frame that wasn’t properly framed to begin with, according to a series of new studies conducted by Mr. Kable’s decision neuroscience lab at the University of Pennsylvania and published in Cognition and Psychological Review.
“There are lots of situations, probably the majority of situations, in the real world,” Mr. Kable told me, “where waiting longer is actually a valid cue that the reward is getting further and further away.”
Mr. Kable and Mr. McGuire tested this logic on a group of shoppers in a mall in New Jersey. As people went about their usual routine, some of them were asked to take part in a 10-minute study during which they could make between $5 and $10. Study participants would see a yellow light appear on a computer screen and could choose to do one of two things: Keep their mouse cursor over a box marked “wait for 15 cents” or move the cursor to a second box marked “take one cent.” What they didn’t know was how long they would have to wait if they opted for the promise of more money. In some cases, the larger rewards were given at relatively regular intervals. In others, however, the timing was more uncertain: the longer you waited, the larger the chance you’d have to keep right on waiting.
The researchers found that while the shoppers seeing the regular intervals looked like the very model of persistence and self-control, those seeing the erratic intervals grew increasingly less persistent over time — even if they had initially been quite patient. The uncertainty of the reward timing was itself enough to push them toward behavior that looked increasingly impulsive.
They also resorted more frequently to skipping trials altogether. They immediately chose to get the cent, instead of waiting a bit to see if a larger payoff was imminent. They weren’t simply impatient, Mr. McGuire and Mr. Kable concluded. They were reacting appropriately to the unpredictability of the future.
Our environment trains us about the value of persistence. Sometimes, it makes sense to wait. At other times, the adage about the bird in hand begins to make sense.
“When you add future uncertainty to the mix,” Mr. Kable pointed out, “it completely changes the problem. Now it’s not just about your ability to wait. With uncertainty, you realize that everybody’s deep intuition, that when you’re waiting, you’re getting closer, is off.” The future may change on you, so what are you waiting for?
In fact, that’s exactly what Mr. McGuire and Mr. Kable found happens in a laboratory setting: You don’t think you get closer as you wait longer. Quite the opposite. In a study published earlier this year, they began by asking participants to estimate how much longer they had to wait for a more desirable future reward — a chocolate chip cookie or a candy bar, depending on their preference. Over and over, they found the same thing: The longer the wait time — anywhere from 2 to 130 minutes — the longer they thought they’d have to keep on waiting.
“The basic idea,” Mr. McGuire said, “is that while a decision maker is waiting, he is constantly re-evaluating the thing he’s waiting for. You’re waiting for the same reward, but your assessment of it changes as a function of the passage of time.”
In a second test, the Penn researchers looked at whether that perception of time changed when you were dealing with everyday behavior: weight loss, improving your time running a mile, improving your scores on a standardized test or improving your piano skills. Once again, they found that the more time passed without reaching the target, the more time people thought remained until they’d get there.
That reaction is the exact opposite of the rational conclusion. Hard logic — or at the very least, logical intuition — would suggest that the more time you invest in something, the closer you are to achieving it. If I practice the piano, I will improve. If I run every day, my time will get faster. But somehow, when we’re in the middle of it all, our minds don’t see it that way. The more time passed, the further the study participants felt from the prize.
Once we realize how our sense of time works and of how long some things will actually take, some famous experiments start to look a little different. Consider what is perhaps the best-known example of self-control: Walter Mischel’s infinitely cited work from the 1960s, which measured how long 4-year-olds could wait for another treat before grabbing a marshmallow that was right in front of them — a study that is the sine qua non of any discussion, academic or popular, of delayed gratification.
Could it be that the kid who stopped waiting had simply miscalculated the amount of time he would have to go treatless? That if he had been given a more concrete estimate — in the original study, the children weren’t told exactly how long they’d be waiting — he would have been able to hold off? That logic would make sense — and is in fact what inspired Mr. McGuire and Mr. Kable to begin their own research.
Mr. Kable works in the same department as the psychologist Angela Duckworth, who has undertaken self-control research of her own. During one lunchtime conversation, Mr. Kable told me, she pointed out that in the marshmallow paradigm, you don’t know when the experimenter is coming back. “You don’t know when you’re getting the second marshmallow,” Kable said. “Well, now you have a completely different situation” — one where temporal uncertainty enters into the picture.
In fact, that time-based logic was suggested by Mr. Mischel himself early on, when he noted that the ability to wait was not tied to length of the wait as such, but to the “accuracy of time conceptions,” as he would later put it.
Similar reasoning led Celeste Kidd, a cognitive psychologist at the University of Rochester, to ask whether uncertainty itself wasn’t to blame. Ms. Kidd created two types of environments: one in which a reliable researcher provided children with a promised reward — a set of art supplies instead of used crayons — and one in which the researcher proved unreliable — he would return and apologize for not having the better reward he’d promised.
The children then took part in the traditional marshmallow study, where they could wait for two marshmallows or eat one now. The prior reliability of the experimenter was decisive: those in the unreliable condition waited an average of three minutes, while those who had interacted with a reliable researcher waited 12. Children, Ms. Kidd concludes, are far more rational than we give them credit for.
Of course, none of this means that actual self-control stops mattering — especially in so-called hot situations, where the thing that’s most tempting to you personally is what’s at stake. I, for instance, would be just fine in the old marshmallow paradigm. I’ve never much cared for the white fluffy things. But place a freshly baked oatmeal raisin cookie from my favorite bakery (Levain, for the record) in front of me, and no abstract promise of future svelteness is likely to make a difference.
While we’re only starting to understand the underlying relationship between time uncertainty and delay ability, it would make sense that I would find it easier to resist the cookie if I had a sense of the exact effect it would have on my weight — and when, exactly, that effect would come. “Part of our argument,” Mr. Kable said, “is that there’s an underlying similarity between the harder problems, the smoking and the diet problems, and the problems that seem unrelated, like waiting for the bus or waiting for the subway. In both cases, we need to find a way to resolve the uncertainty of time.” It might in fact be harder to make excuses — one cookie won’t make a difference — in light of hard, time-based evidence that soon enough it actually will.
So what does this mean, practically speaking? “I spent 10 years of my life quitting smoking,” Mr. Kable said. “I spent the next 10 years trying to diet. This work is of highly personal interest to me.” And does he think it will bring him closer to his goals? “I’m hopeful that it will be helpful,” he responded. “I’ll fully admit that a solution to harder self-control problems is trickier, but I’m hopeful.”
For those of us battling with goals we just can’t seem to reach, the knowledge that our perception of time — and not some inherent shortcoming — is partly to blame may enable us to be more successful in the future. Instead of beating ourselves up for a failure of willpower, we can instead focus on learning to better calibrate our time expectations from the get-go, setting realistic, concretely framed time goals that capture the reality of the task we’ve set for ourselves.
That simple reframing could have very real repercussions for behavior. When Washington, D.C., and New York City introduced signs on their subway platforms that signaled just how long you had to wait for the next train, Mr. Kable pointed out, the decision uncertainty disappeared. “You no longer have to decide if you have time to wait or will be late for your meeting and should just grab a cab,” he said. “When you have that kind of cue, when you can resolve the uncertainty, when it’s a matter of pure knowledge, the decision becomes far easier.”
And what about situations where that kind of cue is more difficult? “I’m hopeful that the same principle will be important,” he said. If you understand exactly how long it will take you to lose weight and incorporate the uncertainty into your thinking — if you realize that it may be a two-to-four-months rather than a two-weeks-or-bust situation — you would be far more capable of resisting that brownie in the present moment. It won’t be as simple as seeing the train wait time tick down in front of you, but it will be better than having no sign on the platform at all. At least you’ll understand that waiting longer doesn’t always mean waiting indefinitely. Investing upfront in realistic time frames — and learning to adjust those time frames as new information becomes available — may help us resist the pull of rewards that come too soon. Controlling our sense of the future, in other words, may help us control our behavior in the present.
Maria Konnikova is the author of “Mastermind: How to Think Like Sherlock Holmes.”