You do not resist change. You resist change that violates your model.
You have tried to build a new habit and failed. You have tried to change someone's mind and hit a wall. You have watched an organization attempt a transformation and collapse back into its old shape within a year. In each case, the standard explanation is "change is hard." But that explanation hides the real mechanism.
Change is not generically hard. It is specifically hard when the method you use to create change is mismatched with the kind of change you are attempting. And the method you choose is not random. It flows directly from your schema about how change works — your implicit theory of what causes things to shift, how long shifts take, what resistance means, and when a change has actually landed.
This is the meta-schema that governs every other schema update in your cognitive infrastructure. When you try to revise a belief, build a new behavior, or reorganize how you work, you are executing a theory of change whether you have articulated one or not. If that theory is wrong — or, more commonly, if it is too narrow to cover the kind of change you are facing — your efforts will fail in ways that feel mysterious but are entirely predictable.
The linear model and its limits
The most widely taught schema about change is linear: identify the current state, define the desired state, execute the transition. Kurt Lewin formalized this in 1947 as the unfreeze-change-refreeze model. In Lewin's framework, change requires three phases. First, you destabilize the existing equilibrium — unfreeze the current pattern by creating awareness that the status quo is insufficient. Then you introduce the new behavior, structure, or belief — the change itself. Finally, you stabilize the new pattern — refreeze — so it becomes the new default.
Lewin's model is not wrong. It captures something real about how stable systems resist perturbation and need deliberate destabilization before they can move. The problem is that most people internalize a degraded version of this model — not Lewin's nuanced original, but a simplified sequence: decide to change, execute the change, be changed. One decision. One action. Done.
This linearized schema produces a specific failure pattern. You decide on Monday that you will wake up at 5:30 AM. You set an alarm. You hit snooze on Tuesday. By Friday, you have abandoned the project and concluded that you lack discipline. But the failure was not discipline. The failure was applying a single-step change model to a behavior that is embedded in a web of other behaviors — sleep timing, evening routines, energy management, identity. The linear schema told you this was a switch to flip. The reality was a system to reconfigure.
Organizational change research tells the same story at scale. John Kotter's eight-step change model, published in Leading Change (1996), emerged precisely because Kotter observed that most corporate transformations fail — not because the new strategy is wrong, but because leaders treat change as a one-time announcement rather than a sustained, multi-phase process. Kotter's steps include creating urgency, building a coalition, generating short-term wins, and anchoring the change in culture. Each step addresses a different reason why linear change fails: people do not feel the need, do not trust the leaders, do not see early evidence it works, or revert to old patterns once pressure eases.
The insight is not that Kotter's model replaces Lewin's. It is that different schemas about change illuminate different failure modes. If your only schema says "decide and execute," you will never build a coalition or generate short-term wins, because your model does not include those as necessary steps.
Change as stages of readiness
A different schema entirely comes from clinical psychology. James Prochaska and Carlo DiClemente's transtheoretical model, developed through the 1980s by studying how people actually quit smoking, describes change not as a sequence of actions but as a sequence of psychological states.
In the transtheoretical model, change moves through five stages: precontemplation (you do not see a problem), contemplation (you see it but are ambivalent), preparation (you intend to act soon), action (you are actively changing), and maintenance (you are sustaining the new pattern). Critically, Prochaska's research showed that most people cycle through these stages multiple times before a change sticks. Relapse is not failure — it is a return to an earlier stage, from which the cycle begins again.
This schema produces fundamentally different behavior than the linear model. If you hold Prochaska's schema, you do not berate yourself for relapsing on a habit change. You assess which stage you have returned to and address the needs of that stage. If you have returned to contemplation, you do not need an action plan — you need to resolve ambivalence. If you are in preparation, you do not need motivation — you need a concrete implementation plan.
The transtheoretical model also changes how you try to influence others. If someone is in precontemplation — they do not see the problem — then providing action steps is not just ineffective but counterproductive. It triggers resistance because you are addressing a stage they have not reached. This is why unsolicited advice about health, finances, or career so often fails. The advice assumes the recipient is in the preparation or action stage. They are usually in precontemplation or contemplation. The mismatch is not about the quality of the advice. It is about the change schema the advisor is running.
Change as system dynamics
Both the linear model and the stages model share an assumption: change happens to an entity. A person changes. An organization changes. A habit changes. Donella Meadows offered a schema that reframes change entirely: what changes is the system, and entities are part of the system.
In Thinking in Systems (2008), Meadows described twelve leverage points — places where a small intervention in a complex system can produce large changes. These leverage points are not equal. At the lowest leverage, you adjust parameters: tweaking numbers, changing quotas, adjusting budgets. These are the easiest interventions and produce the least change. At the highest leverage, you change the paradigm — the fundamental assumptions from which the system's goals, rules, and structures emerge.
Meadows' hierarchy reveals why most change efforts fail. They target low-leverage points. You rearrange your desk to be more productive (parameter change). You adopt a new project management tool (information flow change). You reorganize team structure (system structure change). Each of these addresses a different leverage level, and each can produce change — but the magnitude of change scales with the level. Rearranging your desk does not change how you think about work. Changing how you think about work can rearrange everything.
The most counterintuitive implication of Meadows' schema is that systems resist change at every level. Feedback loops exist specifically to maintain the current state. A thermostat keeps the room at 68 degrees not through inertia but through active correction — it detects deviation and counteracts it. Your cognitive system does the same thing. When you try to change a belief, your existing beliefs generate counterarguments, selective attention, and emotional resistance. These are not bugs. They are the stabilizing feedback loops of your epistemic system doing exactly what they are designed to do.
This means that sustainable change requires either overwhelming the feedback loop (brute force, which exhausts itself) or restructuring the loop itself (which is harder but permanent). If your schema about change does not include feedback dynamics, you will keep trying to overpower systems that are actively correcting against your efforts.
The Stoic schema: change what you can
There is an older schema about change that operates at a different level entirely. Stoic philosophy, particularly as articulated by Epictetus in the Discourses, begins with a partition: some things are within your control and some things are not. Your judgments, intentions, desires, and aversions are within your control. Other people's behavior, external events, your reputation, and your body's eventual decline are not.
This is not passivity. It is a schema about where change effort should be directed. The Stoic schema says that attempting to change what is outside your control is not merely ineffective — it is the primary source of suffering. You do not suffer because your colleague is difficult. You suffer because your schema says your colleague should not be difficult, and you direct change effort at their behavior rather than at your own response.
Marcus Aurelius, writing in Meditations around 170 CE, returned to this schema repeatedly as he governed the Roman Empire through plague, war, and political betrayal. His private journals reveal a mind that was not denying the severity of external problems but was relentlessly redirecting change effort to the one domain where it could actually produce results: his own perception and response.
The Stoic schema about change is a scope limiter. It does not tell you how change works — it tells you where to aim. Combined with a dynamic model of how change works within that scope, it prevents the most common waste: spending energy trying to change things that are structurally immune to your influence.
Adaptive leadership: the technical-adaptive distinction
Ronald Heifetz and Marty Linsky, in Leadership on the Line (2002), introduced a schema about change that explains why organizations keep solving the wrong problems. They distinguish between technical challenges and adaptive challenges. Technical challenges have known solutions — the problem is clearly defined, the expertise exists, and someone with authority can implement the fix. Adaptive challenges require the people with the problem to change their own values, beliefs, or behaviors. No expert can solve it for them because the solution requires internal transformation, not external intervention.
Most change failures happen because people treat adaptive challenges as technical ones. A company's sales are declining. The technical schema says: hire a new sales director, implement a new CRM, restructure territories. These are technical solutions. But if the real problem is that the company's culture discourages honest customer feedback — which would reveal that the product no longer meets market needs — then no amount of technical intervention will help. The people in the system need to change what they are willing to see and say.
Heifetz calls this "giving the work back to the people." The leader's job in adaptive change is not to provide answers but to create the conditions where the people involved can confront the uncomfortable reality that their own patterns are the problem. This is extraordinarily difficult because both leaders and followers prefer technical solutions. Technical solutions do not require anyone to change themselves. Adaptive solutions require exactly that.
Your personal change efforts follow the same pattern. When you treat an adaptive challenge as a technical one — buying a productivity app instead of confronting why you avoid deep work, reading another relationship book instead of examining your own attachment patterns — you get technical solutions to adaptive problems. The schema mismatch ensures failure.
The AI parallel: models that update continuously
Machine learning systems offer a revealing schema about change because they make the mechanics explicit.
A static ML model is trained once on historical data and deployed. It performs well as long as the world resembles the training data. When the world shifts — new user behaviors, market changes, seasonal patterns — the model degrades. This is concept drift: the statistical relationship between inputs and outputs has changed, but the model has not.
The solution is online learning: models that update their parameters continuously as new data arrives. Instead of freezing a model and deploying it indefinitely, online learning systems incorporate each new observation, adjusting weights incrementally. The model never finishes training. It is always in a state of partial revision.
This is a powerful schema about change when applied to your own thinking. Most people operate like static models. They form their core schemas during formative experiences — education, early career, first serious relationships — and then deploy those schemas for decades without systematic retraining. When the world shifts, they experience concept drift: their predictions degrade, but they attribute the failure to the world being unpredictable rather than to their model being stale.
Operating like an online learning system means treating every experience as training data. Not in a frantic way that revises your beliefs after every conversation, but in a structured way that weights new evidence appropriately: surprising observations get more weight, confirmations get less. Your schemas update incrementally, continuously, without requiring a crisis to trigger revision.
The continuous deployment model in software engineering encodes the same principle. Rather than accumulating months of changes and releasing them in a single, high-risk deployment, continuous deployment pushes small changes constantly. Each change is small enough to monitor, test, and roll back if it fails. This reduces the risk of any single change while increasing the overall rate of improvement.
Your cognitive infrastructure can work the same way. Small, frequent schema updates — adjusting a belief slightly after a new observation, refining a mental model after a conversation — compound over time into fundamental transformation. The change is continuous rather than catastrophic. And because each individual update is small, the emotional cost of any single revision is manageable.
Why you hold the change schema you hold
Your schema about change is not one you chose deliberately. It was installed by your environment.
If you grew up in a household where change happened through crisis — everything was stable until it was not, and then everything changed at once — your schema probably says that change requires hitting bottom. You wait for things to get bad enough to justify action. You dismiss gradual deterioration because your model says real change only happens through rupture.
If you grew up in an environment that emphasized discipline and willpower, your schema probably says that change is a matter of effort. You believe that people who fail to change simply did not try hard enough. You apply this to yourself mercilessly, interpreting every relapse as a character failure rather than a signal that your change strategy was mismatched with the challenge.
If you were raised in a context that emphasized external control — change happens when authorities permit it, when conditions align, when you get lucky — your schema says that individual agency in change is limited. You wait for permission, for the right moment, for circumstances to shift in your favor.
None of these schemas are entirely wrong. Change sometimes does require crisis. Effort does matter. External conditions do constrain. The problem is holding any single schema as the complete theory. A schema about change that fits every situation fits none of them precisely.
Calibrating your change schemas
The practice of this lesson is not to find the "correct" theory of change. It is to build a repertoire of change schemas and learn to match the right one to the right situation.
Linear change works when the path is clear and the resistance is low. Adopt it for straightforward skill acquisition, procedural changes, or situations where you have high control and low complexity.
Stage-based change works when the challenge involves motivation and identity. Adopt it for habit change, health behavior, addiction, and any domain where ambivalence is the primary obstacle.
Systems change works when you are embedded in feedback loops that resist your efforts. Adopt it for organizational dynamics, relationship patterns, and any situation where you keep making the same change and reverting.
Stoic scope-limiting works when you are spending energy on things you cannot control. Adopt it as a first pass before any change effort: ask what is within your sphere of influence and direct your effort there.
Adaptive change works when the problem requires you to change yourself, not just your circumstances. Adopt it when technical solutions keep failing and the common denominator is you.
The meta-skill is recognizing which schema to apply. When a change effort fails, the first question should not be "How do I try harder?" It should be "Am I using the right model of change for this kind of change?"
From change to other minds
Your schema about change determines how you approach every transition in your life — from daily habits to career pivots to fundamental belief revision. It is the meta-schema that governs all schema updates. Getting it wrong means every subsequent change effort operates under a flawed theory, and getting it right means matching your method to the actual structure of the challenge.
But change does not happen in isolation. Most of the changes that matter involve other people — their cooperation, their resistance, their beliefs, their behavior. And just as you hold schemas about how change works, you hold schemas about how other people work. Your default assumptions about human nature — whether people are fundamentally trustworthy or self-interested, capable of growth or fixed in their ways, rational or emotional — shape every interaction you have.
In L-0330, you will examine those schemas about other people: where they come from, how they distort your perception, and how to upgrade them so your model of others matches the complexity of actual human beings rather than the simplified caricatures your defaults produce.
Your theory of change is only as good as your theory of the people doing the changing.