The theory behind it

Theory and practice in the same document.

Where each mechanism comes from, how we grade the evidence, and what we do when the literature is still arguing. Nothing here is a trade secret: the method only works if it can be checked.

The skeleton

COM-B: three conditions, none of them optional

For a behaviour to happen, it takes capability (the person can), opportunity (the context allows it) and motivation (it is worth doing). If one fails, the behaviour does not occur — and the right intervention depends on which one failed.

That is why training someone who already knows changes nothing, and why recognition does not fix system friction. The diagnosis exists so you do not get the condition wrong.

Capability — they cannot

They know they should, but lack knowledge, criteria or mental space. Six mechanisms: A1 to A6.

typical intervention: instruct, simplify, put criteria in writing

Opportunity — the context will not let them

The tool, the routine or the group gets in the way. Two mechanisms: B7 and B8.

typical intervention: remove steps, change the channel, make the new practice visible

Motivation — the current way pays better

They can and could, but choose the usual way. Seven mechanisms: C9 to C15 — the largest sector, and that is true of the problem, not an accident of the design.

typical intervention: redesign incentive, authorship, reciprocity

The 15 mechanisms of the Barrier Mandala

each with the signal that gives it away and what tells it apart from its nearest neighbour

codemechanismgradethe signal that gives it awayhow to tell it apart
A1 Operational unfamiliarity A Repeated questions about where and how to publish; nobody can state the path. from A6: here it was never known; in A6 it was known and habit won.
A2 Ambiguous sharing rule B “I am not sure I can upload this” — and policy does not answer. from A4: A2 is about permission, A4 is about quality.
A3 Cost of translating the artefact B The material would have to become something else to fit the required format. from C10: if assisted conversion solves it, it was A3, not ownership.
A4 Uncertain minimum quality threshold B “It is not ready yet” with nobody able to define ready. from C13: A4 is a missing criterion, C13 is fear of judgement.
A5 Overload and competition for attention A The new task competes with the urgent one and loses every week. from B7: A5 is a mental resource, B7 is system friction.
A6 Consolidated habit of the old workflow A The hand goes to email before the reasoning starts. from C11: A6 is automatic, C11 is a stated preference.
B7 Operational friction of the platform A Too many steps, slowness, requirements the old channel does not have. from A1: if teaching solves it, it was A1; if it stays painful once known, it is B7.
B8 Social norm and current practice of the group A Nobody in the group publishes; whoever does becomes a talked-about exception. from C9: B8 is what the group does, C9 is what the organisation rewards.
C9 Structural incentive for the current behaviour A Deadlines and volume are chased; visibility of the process is not. from C15: C9 is a formal incentive, C15 is a sense of fairness in the exchange.
C10 Psychological ownership and territoriality B Language of possession: “my spreadsheet”, “my way”. from C11: C10 is ownership, C11 is familiarity.
C11 Preference for the known state B The current one is defended for being current, with no verifiable advantage. from A6: C11 is a justified choice, A6 is autopilot.
C12 Overvaluing your own artefact C “Mine is better” with no comparison anyone could check. from C10: C12 is about perceived quality, C10 is about authorship.
C13 Evaluation apprehension and psychological safety A Publishing exposes you to public correction; a recent mistake weighed. from C14: C13 is protection from judgement, C14 is deliberate advantage.
C14 Informational power and deliberate asymmetry B Information is held back because knowing first is an advantage. from legitimate withholding: here there is no justifiable secrecy, only position.
C15 Perceived fairness and reciprocity B “We published and nobody reciprocated” — effort recalibrates. from C9: C15 depends on comparison with others, C9 does not.
The grades

How we grade the evidence

The grade does not measure how much we like the model: it measures the state of the literature. A D stays in the base because naming the phenomenon helps the conversation — as long as you know the ground.

AReplicated in meta-analysis

Effect sustained by multiple independent studies and quantitative synthesis. Direction is reliable; magnitude varies with context.

BConsistent evidence, partial replication

Solid base in the field or the lab, without the density of replication of an A. Use it, and observe the result.

CEffect disputed, context weighs heavily

It exists, but its size and conditions are contested. Treat it as a hypothesis to test, never as the sole basis for a decision.

DFragile or under review

Popular, yet with a thin or methodologically contested empirical base. It serves as vocabulary, not as justification.

The replication crisis affects us — and we say where

Much of the behavioural science popularised in the 2000s did not replicate well. Rather than removing those models from the base and pretending they never existed, we mark them C or D and explain what is in dispute.

A grade can be downgraded

When a new meta-analysis weakens an effect, the grade drops and the change enters the changelog with date and reason. You can see what the tool used to recommend and what changed.

The grade is not the recommendation

A mechanism graded A can have low support in your case, and a C can be shouting in the signals. The grade is about the literature; support is about your session. The two rulers appear together, never summed.

Changelog of the base

What changed, when and why

Human curation: the base does not rewrite itself from usage. Every change to the taxonomy goes through review by whoever answers for the method — and stays public.

datetargetchange
12 Aug 2026 Dunning-Kruger Grade lowered from C to D — methodological criticism about a statistical artefact. Kept as warning vocabulary; removed from intervention justifications.
30 Jul 2026 C10 · C11 Added an explicit differentiation test to the card: “does the resistance disappear if the platform reproduces the artefact under someone else's name?” Reduces confusion between ownership and familiarity.
18 Jul 2026 nudge / architecture Grade kept at C after review: aggregate effects smaller than those reported in the early literature. Intervention text rewritten so it does not promise magnitude.
02 Jul 2026 the “do not intervene” outcome Legitimate withholding becomes a first-class outcome, with its own colour and a re-evaluation record — it previously appeared as a note at the foot of the result.

No author quote enters the tool without human verification of the source. Popular but disputed attributions are published without attribution or simply not used.