Identity-Based Goals: Why "I Am a Runner" Beats "I Want to Run More"

Two people are trying to stop smoking. One of them, when offered a cigarette, says, no thanks, I'm trying to quit. The other says, no thanks, I'm not a smoker. The first sentence is a goal. The second is a claim about who the person is. They sound almost the same, and they predict very different outcomes. The first person is describing an action they are working against. The second person is describing a category they belong to, and the cigarette is simply not the kind of thing a person in that category accepts. There is nothing to debate. There is no inner negotiation. The refusal is automatic because the identity is.

The reason this matters is that most of the goals people set for themselves are written in the first voice. I want to read more. I want to run three times a week. I want to drink less. Each of those is a verb dressed in a wish. The verb has to be re-summoned every time the moment for it arrives, and the wish has to outargue whatever else is competing for that moment. A growing body of social-psychology research, going back fifty years and concentrated in the last twenty, argues that this is the slow, fragile way to change a behavior. The fast and durable way is to change the noun.

The two framings

The cleanest way to see the distinction is to put the two framings next to each other. Goal-based framing names a behavior you intend to perform: I want to vote. I want to exercise. I want to write every day. Identity-based framing names a person you intend to be: I am a voter. I am someone who exercises. I am a writer. The behavior in question is the same. The internal posture toward it is not.

Goal-based framing keeps the behavior at arm's length. Each instance of the behavior is a separate decision, and each decision can be won or lost. Identity-based framing collapses the decision. The behavior is what someone like me does, so the question of whether to do it does not really come up. You do not deliberate about brushing your teeth in the morning, because brushing your teeth is not a goal you have. It is something a person of your kind simply does. The aspiration of identity-based work is to move other behaviors into that same uncontested category.

Oyserman's three pillars

The most rigorous theoretical account of this is Daphna Oyserman's work at the University of Southern California, where she has run a research program on Identity-Based Motivation since roughly 2007. Her 2015 book Pathways to Success Through Identity-Based Motivation, published by Oxford University Press, lays out the framework in full. The architecture rests on three pillars, and each one matters for how you might design a cue that supports the work.

The first pillar is dynamic construction. Oyserman's claim, which descends from Henri Tajfel's social identity theory in the 1970s and from a half-century of work on the self-concept, is that identity is not a fixed trait you carry around. It is assembled in the moment, from cues in the environment. Different settings make different identities salient. At a family dinner you are a son or a daughter, and that identity shapes what you do without anyone asking you to perform it. At work you are a colleague. On a trail you are a runner, if the trail and your shoes and the morning light cooperate to make that identity available. The implication is consequential: identity is not something you decide to have once and then own. It is something that has to be re-triggered.

The second pillar is action readiness. When an identity is currently salient, the actions consistent with that identity feel easier and more available. The runner identity, once triggered, makes putting on the shoes feel like the next obvious step rather than a decision. The student identity, once triggered, makes opening the book feel ordinary rather than effortful. Action readiness is the difference between behavior you have to drag out of yourself and behavior that flows from who you understand yourself to be in that moment.

The third pillar is procedural readiness, and it may be the most underappreciated of the three. When an identity is salient, difficulty gets reinterpreted. The runner who finds the hill hard does not conclude that they are not a runner. They conclude that this is what running is, that the hill is part of the thing, that someone like them handles it. The student who finds the chapter hard does not conclude that school is not for them. They conclude that this is what learning feels like. Without identity in place, the same difficulty gets read as evidence of incompatibility: I'm not the kind of person who can do this. With identity in place, the same difficulty gets read as procedure: this is what someone like me deals with.

The school-identity study

The empirical foundation for the theory includes a well-known intervention study Oyserman ran with Daphna Bybee and Deborah Terry, published in the Journal of Personality and Social Psychology in 2006. The work, conducted in low-income middle schools, used a structured curriculum to help students articulate a plausible future self in academic terms and to connect everyday school behavior to that self. The result, measured a year later, was meaningful: students in the intervention showed improvements in attendance, engagement, and grades relative to controls. The mechanism the paper proposed was identity-based motivation. A future self had been made available, and the daily friction of school had been reinterpreted as the procedure of getting there.

The voter study, with the asterisk it deserves

The most cited demonstration of the noun-versus-verb framing in the popular literature is the voter study by Christopher Bryan, Gregory Walton, Todd Rogers, and Carol Dweck, published in PNAS in 2011. In the experiments, participants who were asked about being a voter turned out at higher rates than those asked about voting. The framing shifted a noun for a verb, and the noun won. The result was framed as identity-relevant by the authors and is the source of the popular shorthand that noun framings outperform verb framings.

Honesty requires an asterisk. Later replication attempts have not consistently reproduced the original effect at the original magnitude. Some have found smaller effects, some have found no effect, and the literature continues to argue about the conditions under which the framing matters. The fair reading is that the Bryan finding is illustrative of a mechanism the broader literature supports, and that the magnitude of the framing effect in any single experiment should be held loosely. The deeper point, that identity framing changes behavior, has plenty of other empirical support; the convenient single-sentence version of it does not.

A related and better-replicated case comes from Jonah Berger and Lindsay Rand's 2008 work in the Journal of Consumer Research, where social-norm messages framed around identity (don't be a litterer) shifted behavior in ways that verb-framed messages did not. The pattern is consistent. The exact size of the pattern, across studies and populations, is what scientists are still arguing about.

Every action you take is a vote for the kind of person you wish to become.

Why static identity reminders fade

Once you take dynamic construction seriously, an awkward consequence follows for the ordinary tools of self-improvement. A vision board pinned to a wall is, in theory, an identity cue. In practice, it stops being one almost immediately. The wall and the board become part of the room, and the room becomes part of the unconscious background of life. You walk past the board on the way to make coffee and you do not see it. The cue is still there. The triggering is gone.

The same fate meets the affirmation taped to the bathroom mirror, the sticky note on the monitor, the bracelet on the wrist that was supposed to remind you of something. All of them are static, and static cues stop firing. The brain is extraordinarily efficient at filtering predictable, unchanging stimuli out of conscious awareness. This is a feature. You would go mad if every visual stayed in the foreground. It is also why identity reminders, in the form most people build them, expire on a timescale of days.

Adjacent to this is Carol Dweck's mindset work, which has shaped popular ideas about growth-versus-fixed framings of ability. The mindset literature has also had a hard replication decade, and the careful reading is that the effect is real but smaller and more context-dependent than the popular version implies. The honest thing to say is that the family of ideas around identity, mindset, and self-construal is real, useful, and worth designing for, while the specific effect sizes you may have read about should be held with some humility.

Designing cues that re-trigger identity

If identity has to be re-fired, then the design question is not how to remind someone of a sentence. It is how to put them, briefly, back in contact with the self that already knows what to do. The cue does not need to argue. It does not need to instruct. It needs to make the right identity available in the moment, so that action readiness and procedural readiness do their quiet work.

A cue that re-triggers identity has a few properties. It changes, because a cue that never changes stops being a cue. It lives where it can be glanced at, because effortful attention is the wrong kind of attention for this. It speaks in a voice that is close enough to the person's own thinking that the triggering feels like recognition rather than instruction. It does not nag, because nagging activates the wrong identity entirely (the one being scolded). And it shows up across the day rather than once, because identity has to be reassembled at many different points.

James Clear's Atomic Habits, published in 2018, popularized this design principle for a general audience under the phrase identity-based habits. His framing, that every action you take is a vote for the kind of person you wish to become, is in the same family as Oyserman's work, even if it travels without the citations. The practical upshot, in both the academic and popular literatures, is the same: behavior change is downstream of identity, identity is built and rebuilt by cues, and the cues that build it are the ones you can live with for years.

Further reading

For the academic backbone, start with Oyserman, Bybee, and Terry, Possible Selves and Academic Outcomes, in the Journal of Personality and Social Psychology, 2006; then Oyserman's 2015 book Pathways to Success Through Identity-Based Motivation, Oxford University Press. For the often-cited noun-versus-verb result, Bryan, Walton, Rogers, and Dweck, Motivating Voter Turnout by Invoking the Self, in PNAS, 2011, alongside the replication discussions that followed. For the consumer-behavior parallel, Berger and Rand, Shifting Signals to Help Health, in the Journal of Consumer Research, 2008. For the historical ancestor, Henri Tajfel's work on social identity in the 1970s. For the popular treatment, James Clear's Atomic Habits, 2018, particularly the chapters on identity-based habits.