Behavior Change Models

The classic stage, process, and planning models that structure how researchers and practitioners think about change over time: Lewin's unfreeze–change–refreeze, Prochaska's transtheoretical stages of change, Gollwitzer's implementation intentions, Fogg's three-step model, and the Norcross studies of New Year's resolvers. Each model answers a different question — when change is possible, what people are doing at each phase, how to translate intention into action, and why resolutions actually work more often than the folklore claims. A subpage of habits-and-behavior-change.

Lewin 1958: Unfreeze – Change – Refreeze

Kurt Lewin's "Group Decision and Social Change" is the foundational paper for modern change management. Lewin argued that behavior in any stable system is held in place by a balance of forces — some pushing for change, some resisting. Durable change requires not just adding a new force but deliberately altering the equilibrium.

The three-phase model:

1. Unfreeze — Destabilize the existing equilibrium. Create awareness that the current state is insufficient; reduce the forces supporting the status quo. Without unfreezing, change efforts bounce off the existing structure.
2. Change — Introduce the new behavior, new norms, new systems. This is the active transition phase, where people feel uncertain and incompetent relative to the old state.
3. Refreeze — Stabilize the new equilibrium. Reinforce the new behaviors with supporting norms, structures, and cues so they become the new default.

The second key Lewin insight: group decision is a superior lever to individual persuasion. In wartime studies asking homemakers to cook with unfamiliar organ meats, group discussions that ended with public commitments produced roughly 10x the behavior change of equally informative lectures. The mechanism was social — commitment within a group created both a sense of obligation and a shifted perception of the norm. This is why accountability partners, habit groups, and cohort-based programs outperform individual study. See the social element in the main habits-and-behavior-change hub.

Prochaska's Transtheoretical Stages of Change

Prochaska, DiClemente & Norcross ("In Search of How People Change") synthesized studies of addiction recovery, smoking cessation, and behavior therapy into the transtheoretical model. The claim: people move through identifiable stages on the way to sustained change, and different interventions help at different stages.

• Precontemplation — Not considering change; often doesn't see the behavior as a problem. Interventions focused on action bounce off. What helps: information, pros/cons reflection, consciousness raising.
• Contemplation — Ambivalent. Aware there is a problem, weighing the costs and benefits of change. Can stall here for years. What helps: motivational interviewing, cost–benefit clarification, small experiments.
• Preparation — Intent to act in the near future (weeks, not months). Concrete planning begins. What helps: implementation intentions, environment design, precommitment.
• Action — Actively executing the new behavior. High risk of relapse. What helps: counter-conditioning, stimulus control, reinforcement, social support.
• Maintenance — Sustained new behavior, typically after six months. Vigilance against relapse, rebuilding identity. What helps: relapse prevention, identity consolidation.
• (Termination / Relapse) — Either the behavior is so integrated it no longer requires effort, or a slip returns the person to an earlier stage.

The model is spiral, not linear. Relapse is treated as a normal part of the cycle, not a failure. Most people cycle through the stages several times before achieving sustained change — smokers quit on average 3–4 times before lasting cessation. The practical implication: matching the intervention to the stage matters. Action-oriented advice given to someone in contemplation wastes everyone's time.

Norcross: New Year's Resolvers Actually Succeed

The folklore says New Year's resolutions fail. The data says otherwise. Norcross, Mrykalo & Blagys 2002 ("Auld Lang Syne") ran a six-month prospective study comparing resolvers (people who made a New Year's resolution) with non-resolvers who had the same change goal but no resolution. At six months, 46% of resolvers were still maintaining the change — against 4% of non-resolvers. The commitment ritual worked.

Norcross, Ratzin & Payne 1989 ("Ringing in the New Year") identified coping processes that predicted success across resolvers: stimulus control, reinforcement management, and self-liberation (the belief that one can change). Willpower mattered less than concrete procedural change — altering cues, building in rewards, and operating from a belief in one's own agency.

This is an important corrective to cynicism. The resolution format — a dated commitment with concrete behavioral targets — is structurally sound. Where it fails is the execution: people don't translate the resolution into implementation intentions, don't redesign their environment, and interpret the first slip as a total failure (the "what-the-hell" effect and false-hope syndrome). The format is not the problem.

Gollwitzer: Implementation Intentions

Gollwitzer 1999 ("Implementation Intentions: Strong Effects of Simple Plans") is one of the most practically important findings in behavior change research. A goal intention specifies a desired outcome ("I want to exercise more"). An implementation intention specifies the when, where, and how in if–then format ("If it is Monday at 7am, then I will run for 30 minutes").

The effect sizes are substantial. Across dozens of studies, implementation intentions roughly double the rate of goal completion for everything from cervical cancer screening to exam preparation to recycling. The mechanism: pre-specifying the cue and response transfers behavioral control from effortful prefrontal deliberation to a simpler cue-detection loop. The cue automatically triggers the response without requiring re-deliberation at the moment of action.

Orbell & Sheeran's field study on post-surgery activity resumption showed the effect in a real clinical context. Gollwitzer also showed with Henderson & de Liver 2008 ("Effects of an Implemental Mind-Set on Attitude Strength") that adopting an implemental mindset — focused on execution rather than deliberation — makes attitudes toward the action more extreme and more predictive of behavior. Once you have crossed from deliberating whether to planning how, the psychology of the entire endeavor changes.

Duckworth, Grant, Loew, Oettingen & Gollwitzer 2011 applied this to adolescents, combining implementation intentions with mental contrasting (vividly imagining the desired outcome alongside the obstacles). Students who practiced the combined technique improved their self-disciplined behaviors (homework completion, study time) more than controls. The pairing is important: mental contrasting generates commitment; implementation intentions translate commitment into action.

Bargh, Gollwitzer et al.'s work on the automated will (see habit-formation-and-neuroscience) shares the mechanism — once a cue is strongly linked to a response or goal, pursuit runs without conscious control. Implementation intentions create this linkage deliberately.

Fogg 2010: Three Steps to New Habits

BJ Fogg's "3 Steps to New Habits" short framework complements his Tiny Habits work (see hub page). The three steps:

1. Get specific. Replace vague intentions ("exercise more") with a specific behavior tied to a specific context ("after my morning coffee, I will do two pushups").
2. Make it easy. Shrink the behavior until it can be done even on the worst day. The goal is consistency first, magnitude later.
3. Trigger the behavior. Attach it to an existing cue that already fires reliably. The anchor behavior provides the trigger; the new behavior gets installed immediately after.

Fogg's contribution is emphasizing that motivation is unreliable and ability is the operative constraint. If a behavior is easy enough, you don't need high motivation to execute it — and the repetition, not the magnitude, is what builds the habit. Large behaviors fail not because people lack discipline but because the motivation required to execute them reliably is unsustainable.

Why Change Fails: False Hope and the What-the-Hell Effect

Polivy & Herman ("If at First You Don't Succeed: False Hopes of Self-Change") describe false hope syndrome: the cycle in which people make unrealistic resolutions, feel briefly empowered, fail, and then make new unrealistic resolutions. Each cycle reinforces the belief that change is possible while selecting against the strategies that actually work. The brief motivational high of declaring change is itself rewarding, which perversely rewards the cycle of declaration-and-failure.

The related what-the-hell effect (documented in McGonigal's Willpower Instinct and elsewhere) describes the cascade after a minor lapse. A dieter who eats one cookie tells themselves "I've blown it," then eats the whole box. The single cookie mattered behaviorally almost nothing; the interpretation of the cookie produced the binge. The intervention is pre-written relapse plans and self-compassion — research consistently finds that people who forgive a lapse recover faster than those who catastrophize it.

This is why the CARRT framework (see hub) pre-writes the relapse plan during the design phase, and why Norcross's data on resolvers shows that coping ability at the first slip is a stronger predictor of six-month success than initial commitment strength.

Synthesis: A Stage-Matched Toolkit

The practical integration:

• In precontemplation and contemplation, use Lewin's unfreezing logic and motivational interviewing. Don't push action.
• In preparation, deploy implementation intentions, mental contrasting, environmental design, and precommitment.
• In action, use habit cues, reward design, social support, and stimulus control.
• In maintenance, anticipate relapse, build identity and narrative around the new behavior, and plan for slips with the what-the-hell antidote: one slip is not the failure; the response to the slip is.

Each of these models is partial. Together they cover the arc of change from "I might need to do something about this" to "this is who I am now."

Related Topics

• habits-and-behavior-change — Hub page and practical translation
• habit-formation-and-neuroscience — Context cues, striatum, dopamine, model-based/model-free
• self-control-and-willpower — The parallel literature on willpower
• cognitive-biases-and-psychology — Dual-process cognition and belief effects
• coaching-philosophy — How these principles inform coaching practice