Most of us start out with the best of intentions. Then we walk right past the fruit bowl in search of the devil’s food cake. Or drink one glass of wine too many. Or, after yet another glass, kiss that co-worker at the holiday party. Unfortunately, life constantly presents us with situations that pit our well-reasoned resolutions against the promise of immediate pleasure. As screen legend Mae West once purred, “I generally avoid temptation unless I can’t resist it.” Withstanding temptation takes self-discipline—no easy trick when immediate gratification plumps our sense of well-being. But it is well worth the effort. Self-control saves us and other people from embarrassing or, worse, damaging consequences.

So why do we so often succumb to the siren song and act against our own self-interests? Scientists have tried for decades to understand this all too human conundrum. Sigmund Freud, the founder of psychoanalysis, viewed all behavior as fallout from conflicts among the id, the ego and the superego. In 1986 psychologist Icek Ajzen of the University of Massachusetts Amherst and economist Thomas J. Madden of the University of South Carolina developed a well-known explanation—the theory of planned behavior—in which all our actions derive from our intentions alone. More recently, though, researchers have turned to models that explain self-control—or a lack thereof—as the outcome of a battle between two emotional systems: our impulses and our powers of reflection.

These dual-system models, particularly one developed in 2004 by psychologists Fritz Strack and Roland Deutsch of the University of Wrzburg in Germany, are fairly straightforward: our impulsive self makes fast associations—vending machine equals chocolate. It scans the environment for potentially pleasurable stimuli and sets habitual actions in motion. The strength of these urges varies from one individual to another and from one situation to the next. Personality (are you a risk taker?), current needs (are you hungry?) and prior experiences (did your parents give you chocolate as a reward?) all influence the strength of the impulse. Reflective thought, on the other hand, draws on reasoning and planning; it comes into play whenever someone sets a long-term goal, such as losing weight. Compared with impulses, reflection is resource-intensive, demanding time and memory, but it affords us a good measure of control over our actions.

Because our impulses and our reflections engage different information-processing pathways, dual-system models neatly explain why we are very much of two minds when it comes to temptation. The classic image of an angel on one shoulder and a devil on the other fits well with what researchers have learned: the two systems compete for control over our response to some want; the winner is whichever one experiences greater activation under the circumstances. It is easier to ignore a weak impulse than a strong one (try food shopping on an empty stomach). At the same time, it is easier to engage the reflective system if it clearly recognizes undesirable behavior. Its potency depends on how strongly a person identifies with his or her long-term goals and how firmly those goals are held in working memory, among other factors.

Just Say No
Indeed, a range of influences can help or hinder self-control. In the 1970s psychologist Walter Mischel and his co-workers at Stanford University investigated under what circumstances elementary school pupils were able to resist a small but immediate reward—a piece of candy—in exchange for a larger one later. Among other things, they found that the children were better able to delay gratification—that is, put off the smaller reward and wait for something bigger—when the candy was hidden. Concealing the candy was enough to dampen the children’s impulses. But self-control is not always so easy as out of sight, out of mind. More recent studies have demonstrated that mental strain, stress and the influence of alcohol can impair an adult’s ability to bypass temptation.

Roy F. Baumeister and his colleagues at Florida State University did groundbreaking research in this area in the 1990s. They tested the idea that mental challenges sap the energy required to maintain self-control, much as physical challenges deplete muscle strength. They reasoned that any activity requiring a certain level of mastery would weaken an individual’s self-control in subsequent tasks. Imagine, for example, that you have a job interview at 11 a.m. Naturally, you want to convey a positive image of yourself—an exercise that demands a fair amount of composure. According to Baumeister’s theory, after the interview, you would be less able to resist the urge to have french fries at lunch; after a morning at home, though, you might easily forgo the fries and opt for a healthy salad.

In 1998 Baumeister and his associates performed an intriguing experiment in which they presented subjects with freshly baked chocolate-chip cookies—ostensibly as part of a taste test. They allowed only some participants to try the cookies; others were given radishes. In a later session they asked the subjects to try to solve what were actually insoluble problems. It turned out that individuals who had been forced to withstand the cookie temptation gave up on the problems more readily—on average after only eight minutes. In contrast, those permitted to stuff their face with cookies held out for almost 19 minutes. A control group, made up of subjects who received neither cookies nor radishes, worked for more than 20 minutes before quitting the problems.

Researchers have dubbed this phenomenon, in which external circumstances alter our capacity for self-control, short-term ego depletion. Following the lead of Baumeister and others, we hypothesize that impulses hold greater sway over our behavior when our powers of reflection have fewer resources to draw on. Using a variety of scenarios, we have explored how our ability to act in accordance with long-term goals depends on whether we possess the mental reserves needed to meet them. For instance, we have found that people are often unsuccessful at turning down chocolate—even if they are trying to diet or believe sweets are unhealthy—when they are under the influence of alcohol.

In this experiment, half the participants drank 0.3 liter of vodka and orange juice some 15 minutes before the test; the other half received unadulterated orange juice. We gave all the subjects questionnaires to learn about their consciously held attitudes toward sweets. We also administered the Implicit Association Test (IAT), developed by Anthony Greenwald and his co-workers at the University of Washington, to gauge the extent to which individual subjects associated chocolate with something pleasant—and so how strongly they might be tempted to eat it. We found that it was easy to predict how much a person would eat based solely on their attitudes, provided they had not consumed alcohol. No matter how much the chocolate tempted them, the sober subjects were typically able to stick to their convictions. Among tipsy subjects, however, the more they liked chocolate, the more they ate.

Of course, short-term ego depletion is not the whole story. Drunk or sober, some people seem remarkably disciplined, whereas others have never met a temptation they didn’t like. Various cognitive functions probably account for such differences. Working memory—which seems to govern whether we are able to focus our attention on some aim—most likely plays a role. Far more important may be inhibitory control—that is, the mental brakes we possess to rein in our most pressing desires. After all other control mechanisms have failed—the potato chip hovers near our lips or the cigarette is already lit—inhibitory control can offer us a last-second reprieve. Numerous studies have shown that people who have strong cognitive control—a robust working memory and good inhibitory brakes—are better equipped to keep their resolutions. Poorly developed cognitive control, on the other hand, often correlates with impulsive behavior.

Taking Charge
The dual-system models offer valuable insight into self-control and impulsive behavior: people are generally able to work toward long-term goals—losing weight, quitting smoking, finding a new job—so long as they are not mentally or emotionally taxed, in which case cravings, and the old habits that go with them, grab the wheel. As the studies we have discussed show, self-control can run aground for a variety of reasons, among them a lack of awareness or the presence of overpowering urges. Alternatively, a person might face tremendous strain, which chips away at otherwise intact inhibitory brakes; other individuals might simply lack the cognitive control needed to stop them from acting on impulse. In the final analysis, self-control always depends on the interplay of all these factors and possibly others as well.

Fortunately, there are effective methods to bolster self-control. Traditional approaches generally attempt to strengthen a person’s resolve by equipping them with knowledge, which stands to reason if they fail to see the consequences of their behavior: “You must not smoke, because it will harm you.” Such tactics do not help, though, if the person understands the risks and is nonetheless not motivated—or has no plan—to act otherwise. In these cases, it is often useful to have them formulate small intermediate steps toward their long-term objective, thereby building up so-called implementation intentions. These mini milestones are concrete “if then” resolutions that link critical situations to some desired behavior: “If I am offered a cigarette, I will politely say no.” Many studies have demonstrated the efficacy of implementation intentions, which have been developed by Peter Gollwitzer of the University of Constance in Germany.

Yet another approach aims to train the impulsive system so that it no longer handicaps our pursuit of long-term objectives and may even help. Practitioners repeat neutral or good habits until they eventually replace more deleterious ones—for example, ordering nonalcoholic beer at a restaurant instead of spirits. Such training programs can cause real stress at first, but consistent repetition usually leads to a tipping point, after which the impulsive system automatically triggers the desired response. Dutch psychologist Reinout Wiers and his colleagues at the University of Amsterdam have found that even simple exercises can serve to retrain our impulses. The researchers asked alcohol-dependent patients to repeatedly move a joystick in a certain direction to signal rejection whenever they saw a photograph of alcohol on a computer screen. When they tested the program at a substance abuse clinic, the results were promising: one year after discharge the recidivism rate among patients drilled on the computer was lower than that among those who received only standard treatment.

Still other strategies target working memory in an attempt to fortify weak cognitive control. Torkel Klingenberg and his team at the Karolinska Institute in Stockholm have tested this idea in children and the elderly, but such a program might also aid in adults lacking self-discipline. The surest ploy may simply be avoiding temptation wherever possible. You are unlikely to wrestle with self-control if you steer clear of potentially compromising circumstances in the first place.

But of course, that is not always possible. So the next time you face the choice between short-term gratification and a long-term goal, think about the battle going on in your brain between impulses and reflective thoughts. It might just help you to muzzle your cravings.