Main Content

Beyond Willpower: Cognitive Strategies for Self-Control

by Angela Duckworth, Katherine Milkman and David Laibson

People with greater self-control fare better in terms of important dimensions of human flourishing. Here’s how to join their ranks.


SELF-CONTROL FAILURES CONTRIBUTE to a range of outcomes in our society, from educational achievement and retirement savings to the obesity epidemic. Not surprisingly, people with greater self-control fare better in terms of health, wealth and many other dimensions of human flourishing. Unfortunately, temptations — rewards that provide short-term gratification but impede people from long-term goals — are ever more abundant today, thanks to convenience stores, one-click shopping, social media, 24/7 streaming, and other newfound vices.

In a recent paper, we synthesized the research on approaches to reducing failures of self-control, organizing them as approaches that are either cognitive or situational in nature. In this article we will summarize some of the key cognitive strategies for improving self-control. We will categorize these strategies under two headings: interventions that are self-deployed and interventions that are other-deployed — i.e. by policymakers or organizations.

Self-Deployed Cognitive Interventions

The following suite of cognitive interventions enable people to change the way they think, making long-term choices more appealing or actionable and short-term temptations less so.

GOAL SETTING. Goals are mental representations of what we hope to accomplish. Across dozens of studies, setting specific, difficult goals has been shown to help people achieve higher performance than exhortations to ‘do your best’. Goal setting is effective in part because goals direct our attention and energy towards a desired behaviour. In addition, failing to achieve the reference point set by the goal feels like a loss, and as Daniel Kahneman and Amos Tversky taught us long ago, losses loom larger than gains in our minds, creating enhanced motivation for us to persist.

The research indicates that it can be helpful to break far-off goals into smaller, more proximate sub-goals. Accomplishing these sub-goals leads to small wins, which fosters a sense of progress and boosts self-efficacy. Setting intermediate deadlines with tangible financial consequences may also reduce procrastination.

To yield significant benefits, self-controlled behaviours must be enacted consistently over time.Tweet this

In one illustrative study, elementary school children were instructed to complete one math module during each of seven math training sessions (the ‘sub-goal condition’); others were instructed to complete seven math modules by the end of seven sessions (the ‘distal goal condition’). On a post-test the former group outperformed the latter group by about 40 per cent.

Directing attention to sub-goals is not always more beneficial than directing attention to super-ordinate goals. In one study, dieters were randomly assigned by researchers to adopt either distal goals (weekly limits of food consumption) or proximal goals (limits for each of four time periods in a day). The majority of participants in the distal condition spontaneously adopted proximal goals, and those who did not subsequently lost no weight. In contrast, dieters who were assigned to set proximal goals reduced their food intake and lost more than a pound per week.

More recent research has focused on the benefits of a goal-commitment strategy called ‘mental contrasting’ (MC). This technique entails imagining — in vivid detail — a positive outcome of attaining a goal and then bringing to mind a negative obstacle that presently stands in the way. At least two mechanisms explain the benefits of MC on goal attainment. First, the juxtaposition of a desired future with an obstacle that obstructs its realization energizes the individual to take action. Second, this juxtaposition can provoke spontaneous planning.

PLANNING. Goal commitment is necessary but not sufficient for accomplishing goals. After committing to a goal, the attainment of that goal has been shown to be a function of the specificity of plans (i.e. when, where and how a person will take action). A certain type of plan, called an ‘implementation intention’, links an anticipated cue with a desired action. Such ‘if-then’ implementation intentions have been shown to create mental associations that are automatically enacted and psychologically costly to break. As a result, they reduce procrastination and forgetfulness and support perseverance when carrying out difficult actions.

Making plans has been shown to increase self-controlled choices, including exercising, completing assignments, getting a flu shot and getting a colonoscopy. An analysis of 94 independent tests found that making if-then plans increased goal attainment. For example, such prompts boosted flu vaccination rates by four percentage points when included in reminder mailings and increased voter turnout by one percentage point when included in get-out-the-vote phone scripts.

MENTAL CONTRASTING + IMPLEMENTATION INTENTIONS. In recent studies, MC has been paired with implementation intentions (II). After articulating a goal, an individual mentally contrasts the imagined positive outcome (‘What would be the best result of accomplishing this wish?’) with the obstacle that stands in the way (‘What might prevent me from accomplishing this wish?’). Finally, the individual makes a plan (‘What is an effective way to tackle this obstacle?’).

This combination (‘MCII’) has been shown to support self-control better than either creating implementation intentions or engaging in mental contrasting alone. Mental contrasting increases readiness to make if-then plans and, in addition, prompts the identification of obstacles that can then be addressed with if-then plans. For example, MCII has been shown to increase self-reported physical activity at 16 weeks after intervention and three months after intervention. It has also been used to improve academic outcomes. Compared with classmates randomly assigned to a placebo control condition, fifth graders trained in MCII earned higher report card grades, came to school on time more reliably, and were rated by their teachers as superior in classroom conduct. Note, however, that after one marking period, these improvements in academic outcomes diminished to non-significance, suggesting the need for additional support and reinforcement.

SELF-MONITORING. To yield significant benefits, self-controlled behaviours must be enacted consistently over time. For example, resisting dessert, studying, going for a run or saving a few dollars for retirement all pay dividends for long-term well-being only if repeated again and again. This presents a challenge, because attention to goals can lapse. New Year’s resolutions, for example, are at the top of one’s mind in early January but quickly lose their urgency. Moreover, people may not be fully aware of how much their snacking, web surfing, couch sitting, impulse shopping, and other bad habits undermine their long-term goals.

Self-monitoring is the intentional and consistent observation of one’s own behaviour. The potential benefits of this are especially well documented in the domain of weight loss. One early study found that dieters who consistently monitored their food intake lost more weight than those who did not. A more recent systematic review confirmed a consistent relationship between self-monitoring and weight loss, although it noted that more studies using objective outcome measures are needed. With that caveat in mind, we note that self-monitoring has been shown to help alcoholics drink less and to help students improve academically.

PSYCHOLOGICAL DISTANCING. People are most attracted to temptations when they are available in the here and now. Accordingly, each of the four dimensions of ‘psychological distance’ — spatial, temporal, social and hypothetical — provides a mental separation from otherwise alluring temptations. In other words, the more a temptation is ‘not here’, ‘not now’, or ‘not for me’, the weaker its allure.

Conversely, psychological distance tends to promote the pursuit of more valued goals, the benefits of which are often more abstract than the gratifications of the present moment. Research has demonstrated the capacity of both children and adults to construe situations in terms of psychological distance. In one famous study, children were rewarded if they could resist the temptation to eat a marshmallow sitting in front of them. Those who were encouraged to think of marshmallows as ‘round and puffy like clouds’ were able to wait more than twice as long as children encouraged to think of their ‘chewy sweet taste’.

Likewise, both children and adults who process emotionally upsetting events in the third person (using their name or a third-person pronoun) rather than the first person (‘I’) demonstrate superior emotion regulation. Psychological distancing has also been posited as an active ingredient of both mindfulness and cognitive therapy.

MINDFULNESS. Mindfulness refers to non-judgmental awareness of one’s present experience. Although rooted in a 2,500-year-old Buddhist tradition of scholarship and practice, it has become a popular modern practice. At least three mechanisms of mindfulness may explain its benefits for self-control: reduced craving, reduced mind wandering, and the decoupling of cravings from behaviour.

Studies also indicate that meditation training can reduce stress-related mental-health conditions, including anxiety and depression. Evidence from a smaller number of random-assignment field studies suggests that mindfulness may ameliorate outcomes associated with self-control, including addiction, blood glucose stabilization, and emotion regulation and aggression. A recent meta-analysis of field trials found that mindfulness training reduces impulsive eating and increases physical activity.

It is socially uncomfortable to deviate from the crowd.Tweet this

COGNITIVE THERAPY. More than 50 years ago, Aaron Beck and Albert Ellis pioneered cognitive therapy. These psychotherapists identified ‘distorted thinking’ as the underlying pathology in an array of psychological problems. In contrast to behaviour therapy, cognitive therapy is particularly concerned with mental states, including subjective beliefs about the self and the world, that are presumed to give rise to behaviour.

Typically, cognitive therapy comprises weekly sessions with a therapist and, in addition, therapist-assigned ‘homework’ completed between sessions. With therapist serving as teacher and coach, clients are trained first to notice negative automatic thoughts (e.g. ‘I can’t stand this craving for a cigarette! I must smoke!’), then to interrogate thoughts for accuracy and realism (‘Is it really true that I can’t endure this craving? What if my child’s life depended on my not smoking now? Is it more accurate to say this craving is very painful?’), and finally to engage in the development of more realistic and accurate thoughts (‘I crave a cigarette, and it will be difficult, but not impossible, to resist smoking now’).

Cognitive Interventions Deployed By Others

We now turn to cognitive interventions that are typically initiated by outsiders, such policymakers or employers.

DESCRIPTIVE SOCIAL NORMS. When people learn that the majority of their peers are engaging in a particular behaviour, they are motivated to shift their behaviour in the direction of that ‘norm’ for at least two reasons. First, they assume that the information is conveyed by the crowd: Maybe their peers know something they do not? And second, it is socially uncomfortable — and even ostracizing — to deviate from the crowd.

Research has identified a medium-to-large association between descriptive social norms and an individual’s behavioural intentions, particularly for health-risk behaviours such as smoking and drinking. Field experiments have shown that carefully designed cues can change perceptions of the prevalence of peer behaviour.

In one study, hotel patrons were eight percentage points more willing to reuse their towels when they were told that 75 per cent of other patrons had done so. And in a massive field study run by the company OPower, being informed about the energy consumption of one’s neighbours in similar homes was shown to promote energy conservation, leading to reliable two per cent year-over-year reductions in energy usage. Moreover, these reductions persisted over multiple years. 

Marketing campaigns based on descriptive social norms are common. In one study, nearly half of surveyed universities had adopted some form of this strategy for reducing binge drinking. However, descriptive social norms can backfire if ‘upward social comparisons’ discourage, rather than heighten, motivation to change. For example, in one experiment, information about the high rates at which peer employees were contributing to their 401(k) retirement plans actually decreased individual savings rates by creating discouragement among low earners who perhaps felt unable to ‘keep up with the Joneses’.


Social identities are categorical affiliations with a social group (‘people like me’). These identities come with beliefs (‘what people like me believe’) and behavioural norms (‘how people like me behave’). Any one individual is in possession of multiple identities. One can identify, for example, as a woman, as an Asian American, as a leader, as a liberal, as an exerciser, as a mother, and so on. However, only a subset of these identities may be active at any given moment or in a particular context.

For instance, at home on the weekend, an individual may identify as a mother but at work during the week may identify as a leader. Scripted identity-congruent behaviours tend to be carried out without complete consideration of expected costs and benefits, which by definition favour more impulsive rather than self-controlled choices. Thus, social labelling — prompting a particular social identity — can encourage self-control by circumventing deliberate cost-benefit calculations and instead encouraging identity-congruent behaviour.

In one early field study, children who were told that they were ‘neat and tidy people’ were less likely to litter and more likely to clean up than children who were told that they ‘should be neat and tidy’. In a more recent field study conducted in college dormitories, posted flyers promoted responsible drinking. In the social-labelling condition, flyers linked alcohol consumption to a distinct social group — graduate students — whereas in the control condition, flyers focused solely on the negative health effects of alcohol. Two weeks later, freshmen exposed to the social labelling flyers reported consuming 50 per cent less alcohol than freshmen in the control condition.

Social labeling may be particularly effective when decision makers are mentally distracted: In one study, only when under cognitive load (i.e. asked to keep a six-digit number in memory) did adults labeled as ecologically conscious consumers make more environmentally friendly but more expensive shopping choices than adults simply urged to be more ecologically conscious or adults in a no-treatment control group.

MAKING THE FUTURE SELF MORE RELATABLE. One key source of self-control failure is an inability to empathize with our future self. Interventions that make the future self more relatable have shown some promise as a means of increasing self-control. For instance, in a series of laboratory studies of savings decisions, exposing participants to age-progressed renderings of their future selves led participants to allocate up to twice as much money to hypothetical retirement savings accounts.

In other studies, to make participants’ future selves feel closer, researchers asked them to judge how easily they could generate two (rather than 10) reasons why their identity would remain stable over 12 months. Generating two reasons was easy, giving participants in this condition the impression that they were relatively close to their future selves (compared with participants in the 10-reason arm of the study). This led to more patient purchasing decisions and lower measured discount rates. Likewise, participants in another study who wrote letters to their future selves (an intervention designed to increase connectedness) were less inclined towards hypothetical delinquent behaviours (e.g. buying desirable stolen goods). Viewing age-progressed renderings of their future selves (instead of seeing a rendering of their current selves) also reduced the rate at which participants in a laboratory study cheated on a quiz to earn additional money.

Research suggests that there are predictable moments when people are especially motivated to make self-controlled decisions, and one class of such moments arises at the beginning of new cycles (e.g. the start of a new week or year, following a birthday or holiday). These so-called fresh-start moments facilitate the attainment of long-term goals because they help people feel disconnected from their past failures, which elevates their current self-image and confidence.

Evidence shows that self-controlled acts (e.g. searches for the term ‘diet’ on Google, gym visits, and the creation of goals on one popular goal-setting website) increase naturally on fresh-start dates. Interventions can be designed to explicitly leverage fresh starts as a means of promoting self-controlled decisions. For instance, noting that an otherwise unremarkable date corresponds to the start of a new cycle (e.g. the first day of spring, the start of university’s summer break) increases the rate at which study participants choose that date to receive a reminder about pursuing their goals.

Likewise, participants in a field experiment who were invited to start saving for retirement in the future were more interested in signing up to save when the future savings opportunity was labelled as following their birthday, saving 30 per cent more than others over a nine-month follow-up period as a result. By emphasizing fresh starts on the calendar as opportunities to make self-controlled decisions, a small but growing body of evidence suggests it may be possible to encourage more self-controlled choices.

In closing

As indicated herein, there is no shortage of self-control research for organizations to benefit from. And this is just the tip of the iceberg: Interested readers can access our complete paper online to review the full suite of self-control strategies — including situational interventions that are self-deployed and other-deployed.

Given the wide-ranging and growing temptations people face in the modern world, we believe there is an urgent need for an applied Science of Self-Control — one that incorporates insights from both psychological science and economics. It is our hope that our review is a positive step in this direction.  

Angela Duckworth is the founder and CEO of Character Lab, a non-profit that uses psychological science to help children thrive. She is also the Christopher H. Browne Distinguished Professor of Psychology at the University of Pennsylvania, faculty co-director of the Penn- Wharton Behaviour Change for Good Initiative, and faculty co-director of Wharton People Analytics. Katherine Milkman is a Professor at the Wharton School, where she holds the Thompson Endowed Term Chair for Excellence in Teaching. She has a secondary appointment at Penn’s Perelman School of Medicine. David Laibson is the Robert Goldman Professor of Economics and a Faculty Dean of Lowell House. He leads Harvard University’s Foundations of Human Behaviour Initiative. This article has been adapted from their paper, “Beyond Willpower: Strategies for Reducing Failures of Self-Control,” which was published in Psychological Science in the Public Interest (Sage Publications). The complete paper is available online.

This article appeared in the Spring 2020 issue. Published by the University of Toronto’s Rotman School of Management, Rotman Management explores themes of interest to leaders, innovators and entrepreneurs.

Share this article:

Read More Follow Us on twitter Email List Subscribe Today