Part 2. Habit Foundation
Habit Evolution: From Repetition to Representation
The map is not the territory. But it turns out the map matters rather a lot.
Kristine Kutuzova
Founder, Habit Design Hub
Abstract
By the mid-twentieth century, behavioural science had constructed a reasonably tidy account of habit formation: stimulus, response, reward. Observable, measurable, replicable in laboratory conditions. The difficulty was that human beings kept declining to behave accordingly. Cognitive science entered with an intention to complicate behaviourism by introducing memory, attention, and expectation as variables that shaped whether any habit actually held. Habits were reframed as internal representations: stored, retrieved, revised. The field moved inward. So did the questions. If habits are mental structures rather than conditioned sequences, why are they still so difficult to sustain in practice? This article traces the conceptual turn from observable repetition to internal representation, and considers what it resolved, and what it quietly left open.
Behaviourism had been, by any measure, a productive framework. It produced results. In controlled environments, with predictable consequences and stable conditions, it could explain, and to a meaningful degree predict, human behaviour. That is not a small achievement. The problem was not that it was wrong. The problem was that it was increasingly, inconveniently, incomplete.
The post-war intellectual climate was not especially patient with incompleteness. The 1950s brought computing, information theory, and a new vocabulary for thinking about the mind, but not as a black box to be set aside, but as a system to be modelled. If a machine could process information, store it, retrieve it, and produce outputs shaped by what it had previously encoded, then perhaps the human mind was doing something structurally similar. Cognition, in other words, was not a philosophical indulgence. It was an engineering problem. And engineering problems, unlike philosophical ones, invite solutions.
Memory became the new unit of analysis. Where behaviourism had asked what does the organism do, cognitive science began asking what does the organism know, and how does what it knows shape what it subsequently does. The shift sounds subtle. Its implications were not.
An internal representation, in the cognitive sense, is not simply a memory of an event. It is a compressed, contextualised structure, a schema, a script, a pattern, that the mind uses to interpret incoming information and anticipate what comes next. Habits, understood through this lens, were no longer sequences stamped in by repetition. They were encoded architectures.
The distinction matters because it changes the question entirely: from how often did you do it to what did your mind make of what you did.
The crack in behaviourism had been visible for some time before the field formally acknowledged it. The most elegant early demonstration came not from a human laboratory but from a rat maze. Edward Tolman, working in the 1940s, showed that rats navigating familiar environments were doing something considerably more than following a conditioned route. When the route was blocked, they found alternatives. When the environment was rearranged, they updated their behaviour in ways that suggested they had been building an internal map - a cognitive representation of space, rather than simply executing a reinforced sequence.
Tolman called these cognitive maps. The behaviourists found them inconvenient. If rats were encoding the structure of an environment rather than merely responding to stimuli within it, then the stimulus-response framework was not just incomplete - it was missing the mechanism. The behaviour was a product of internal representation. The representation was doing the work that conditioning had been credited with.
The human evidence was no tidier. The same environment produced different behaviours in different people. The same person produced different behaviours on different days in the same environment. Intentions shifted outcomes. Expectations altered responses. Two people could receive identical cues, identical reinforcement schedules, and arrive at entirely different habits. Behaviourism could describe this as noise. Cognitive science began to treat it as signal.
The question the field had to reckon with was not simply what is being stored. It was how stored structure shapes what is retrieved, and how retrieval, in turn, shapes what is done.
Note. This diagram traces key paradigm shifts in habit theory, showing how the field moved from memory and observable behaviour to cognition, neural systems, and context. Each shift repositions where habits are understood to reside and what governs their formation. Together, they reflect not a replacement of models, but an expanding framework developed across the historical and modern perspectives presented in this library.
The cognitive reframing of habit is, at its core, a reframing of where behaviour comes from. Behaviourism located the answer outside — in the environment, the stimulus, the consequence. Cognitive science located it inside — in the structures the mind builds from experience, and the expectations those structures generate.
Frederic Bartlett's work on memory, developed in the 1930s and gaining wider influence through the cognitive revolution, introduced the concept of the schema: a mental framework built from prior experience that actively shapes how new information is perceived and stored. Memory, Bartlett demonstrated, is not a recording. It is a reconstruction. What you recall is shaped by what you already knew when you encountered it. The implications for habit were considerable: if behaviour is retrieved from stored structure rather than stamped in by repetition, then the quality of what gets encoded matters as much as the frequency.
Ulric Neisser extended this into a broader theory of cognitive processing — the mind as an active constructor of experience rather than a passive receiver of stimuli. Attention, on this account, is not simply a spotlight that illuminates input. It is a selective process that determines what gets encoded in the first place. A habit, then, is not just what you repeated. It is what your attentional system deemed worth retaining.
And then there is expectation. The cognitive turn introduced the idea that behaviour is not simply a response to what is present, but an anticipation of what comes next. The mind, operating on prior experience, generates predictions. When a prediction is confirmed, the pattern reinforces. When it is violated, the system updates. Habit, understood this way, is the accumulated residue of confirmed expectation. Which is a more unsettling account than it first appears — because it means that what a habit is made of is not action, but anticipation. What your system expects to do is already shaping what it does before the cue arrives.
This is the paradigm shift. Behaviour is no longer the output of external conditioning alone. It is the product of internal architecture shaped by experience, interpretation, and expectation. The field had moved inside — and it had taken the explanation of habit with it.
The elegance of behaviourism was always partly its accessibility. Stimulus. Response. Reward. The chain was short, the logic was clear, and the intervention points were visible. Cognitive models offered greater accuracy at the cost of greater complexity, and complexity, in practice, creates its own kind of resistance.
The tension between the two traditions did not resolve cleanly. It produced, instead, a series of attempted syntheses.
The most influential was dual-process theory, developed most accessibly by Daniel Kahneman:
System 1 as fast, automatic, contextually triggered processing;
System 2 as slow, deliberate, effortful reasoning.
Habits, on this account, belonged to System 1. Behaviour change - the attempt to install new habits or disrupt old ones, required System 2. The framework was genuinely useful. It was also, in practice, a source of considerable confusion, because it implied that the solution to habit problems was more deliberate thinking. The cognitive turn had already demonstrated why that implication was incomplete.
Ann Graybiel's neuroscientific research provided a different kind of synthesis - more biological. Her work on the basal ganglia showed that repeated behaviours become encoded as neural chunks: compressed sequences that fire as units once initiated, operating largely outside conscious awareness. The cognitive account had argued that habits were mental structures. Graybiel showed where those structures lived, and demonstrated that once encoded, they ran with minimal cortical involvement. Cognition was necessary for initiation. It was not, once the pattern was established, particularly involved in execution. What this produced was a layered picture.
Behaviourism
was correct that repetition shapes behaviour.
Cognitive science
was correct that internal structure mediates it.
Neuroscience
confirmed that both processes operate simultaneously, in different systems, at different timescales.
Each tradition had located a real mechanism. None had located all of them. The synthesis, such as it was, explained more, and made habit design simultaneously more tractable and more humbling.
Wendy Wood and Dennis Rünger's landmark review, published in 2016, drew the contemporary consensus into focus. Habits, they argued, are learned associations between context and behaviour - formed through repetition in stable conditions, and governed, once established, by automatic activation rather than conscious intention. The cognitive turn had been necessary. It had also, Wood suggested, overcorrected. The late twentieth century's emphasis on mindset, identity, and intention had produced a generation of habit advice that privileged internal states at the expense of structural ones. Cognition initiates change. Context sustains it. The two are not interchangeable, and treating them as though they were is precisely why most habit interventions dissolve on contact with real conditions.
The practical consequence of the cognitive turn, properly understood, is not that thinking harder produces better habits. It is that the internal architecture of expectation, attention, and interpretation must be accounted for in how habit systems are designed, alongside, and with equal weight to, the environmental conditions that trigger and maintain them. A tool built solely on cues and rewards is working with half the model. A tool built solely on mindset and identity is working with the other half. Neither half, alone, is sufficient.
This is where the Habit Design Hub platform begins. The field spent a century locating the mechanism of habit - in memory, in neural pathways, in reinforcement, in schema, in context. What it produced, across that century, was not a single answer but a set of overlapping partial ones. Each framework contributed a layer. Each layer added accuracy and complexity in equal measure. The contemporary practitioner inherits all of them, usually without the lineage, which means they also inherit the confusion.
Habit Literacy™ is the capacity to read the layered contextual architecture with enough precision to work with it. It means understanding not just what a habit is, but which system is running its and what that system implies for how that habit can be changed.
Change the context, or the context will outlive your intentions.
References:
Bartlett, F. C. (1932). Remembering: A study in experimental and social psychology. Cambridge University Press. Graybiel, A. M. (2008). Habits, rituals, and the evaluative brain. Annual Review of Neuroscience, 31, 359–387. https://doi.org/10.1146/annurev.neuro.29.051605.112851 Kahneman, D. (2011). Thinking, fast and slow. Farrar, Straus and Giroux. Leroy, S. (2009). Why is it so hard to do my work? The challenge of attention residue when switching between work tasks. Organizational Behavior and Human Decision Processes, 109(2), 168–181. https://doi.org/10.1016/j.obhdp.2009.04.002 Neisser, U. (1967). Cognitive psychology. Appleton-Century-Crofts. Tolman, E. C. (1948). Cognitive maps in rats and men. Psychological Review, 55(4), 189–208. https://doi.org/10.1037/h0061626 Wood, W., & Rünger, D. (2016). Psychology of habit. Annual Review of Psychology, 67, 289–314. https://doi.org/10.1146/annurev-psych-122414-033417
Few questions this article leaves open
If habits are context-dependent mental structures, what happens to them when the context changes completely?
The cognitive model explains persistence well. It is considerably less clear on what happens when the activating conditions are removed entirely — relocation, life transition, structural disruption. The representation doesn't disappear. Without its context, it cannot fire. What does habit design look like in the absence of a stable environment to encode against?
Further Reading: Destination-Based Habit Design™
Attention is a variable in cognitive habit models. What happens to habit formation under conditions of sustained attentional fragmentation?
Cognitive science introduced attention as a mediating factor in whether behaviour becomes habitual. The assumption was broadly that attention could be directed. That assumption was formed before the attentional environment of contemporary life. Leroy's research on attentional residue suggests that attention is not simply available to direct. What does this mean for any habit that requires cognitive engagement to initiate?
Further Reading: Habit Hygiene™
Dual-process theory distinguishes fast automatic processing from slow deliberate reasoning. Which system does habit design actually target?
Duhigg's cue-routine-reward loop distilled Skinnerian architecture into a three-part model that is genuinely useful and genuinely incomplete. Understanding what it captured, and what it compressed out of view, clarifies both its value and its limits as a design tool.
Further Reading: Habit Literacy™
Wood and Rünger argue that habits are not goal-driven but context-driven. What does this mean for the role of intention in habit design?
Intention may start a behaviour, but only structure determines whether it continues. The design question shifts entirely. What, then, is the correct role of motivation in a well-designed habit system?
Further Reading: Individual Habit Design™
The cognitive model reframed habits as internal representations: schemas, scripts, stored sequences.
Where does identity sit in that model?
A habit can encode without an identity claim attached to it. The question of whether identity strengthens, weakens, or simply sits alongside habitual encoding is not settled. Habit advice that collapses identity and behaviour into a single mechanism may be offering clarity at the cost of accuracy.
Further Reading: Habit Literacy™
The content of this article is intended as an informational and analytical resource. It reflects research-based perspectives on habit science and does not constitute therapeutic, psychological, or clinical advice. If you are working through behavioural challenges that affect your wellbeing, please consult a qualified professional.
© Habit Design Hub™. All rights reserved.
The content of this article, in full or in part, may not be reproduced, redistributed, adapted, or used for commercial purposes without prior written permission. Sharing a link to the original publication is always welcome.