Identity
Self-understandings, roles and values a person reports. These need not be internally consistent.
ASK / What feels important? FIELD NOTES / ISSUE 01 / 24 OCT 2026
WHY KNOWING
ISN'T DOING.
You know the email matters. You want to send it. You have the address. And somehow, another day passes.
Sometimes what looks like a failure of will is a collision between uncertainty, resources, incentives, social stakes and a specific moment. The Identity–Action Gap is one proposed way of asking what got in the way. It is not a validated test.
Conceptual / provisionalIndependent inquiry by Sahil Tayal
HYPOTHESIS CHANNEL● SIGNAL INTERRUPTED
01 / THE PREMISEThe gap is not
The gap is not
the whole person.
Start with a concrete behaviour, not a sweeping label about someone's character.
A person can value a goal, describe it eloquently and still fail to take the next step. That mismatch is interesting. It does not, by itself, prove laziness, hypocrisy, a broken identity or a lack of desire.
Consider an applicant staring at a job form. They may want the job, know which button to click, and still be stopped by an ambiguous question, fear of public rejection, an inaccessible website, an exhausting shift or uncertainty about whether the role is legitimate. Different causes can produce the same visible outcome: no application.
The proposed Identity–Action Gap framework asks us to separate what a person endorses, what happens, and what conditions sit between the two. That separation is useful for asking questions. Whether IAG is an original, reliable or predictive scientific model remains unestablished.
Action
An observable, time-bounded behaviour—not an impression that someone is “productive.”
ASK / What happened, by when?Environment
Resources, institutional conditions, opportunities and situational constraints.
ASK / What made action possible?Support
Helpful feedback, social backing, tools, practical assistance or scaffolding.
ASK / What assistance was available?Friction
Task complexity, ambiguity, perceived risk, coordination costs or other obstacles.
ASK / What made this step harder?I, A, E, S and F are organising labels proposed for this project, not independently validated psychological constructs or accepted causal coefficients. Terms overlap; the categories may need revision. “Identity” does not imply a stable or measurable essence.
02 / THE CLAIM UNDER PRESSUREFour propositions.
Four propositions.
Four things to doubt.
Endorsing ≠ enacting
Reported values and intended actions can diverge in observable behaviour.
Challenge: Was the “intention” sincere, stable, specific or measured at the right time?Conditions may matter
Opportunity, resources and task demands may affect whether an intention can become action.
Challenge: Could the observed relationship be explained by skill, health, incentives or changing goals?Support may shift the path
Feedback, assistance or an easier first step may change the experience of initiating action.
Challenge: Which kind of support, compared against what—and for whom?Friction may mediate
Some obstacles could help explain when intentions fail to turn into a particular behaviour.
Challenge: Does “friction” explain anything beyond the established models?These are conceptual propositions for scrutiny, not confirmed study findings. No fitted equation, numerical prediction or treatment effect is asserted.
03 / THE INSTRUMENTChange the conditions.
Change the conditions.
Not the person's worth.
A deterministic thought experiment. The sliders describe a fictional situation, not a measured mind.
CHOOSE A CASE FILEPRESETS ARE FICTIONAL
EDITABLE CONDITIONS / NONE ARE PERSONALITY TRAITS
The unfinished application
An applicant cares about a role but hasn't submitted the form. We do not know why.
LIVE INTERPRETATION / NOT A FORECAST LOCAL
What deserves
closer attention?
Features in this illustration
- Several obstacles may be salient.
- Available support may be limited.
CHANGE THE LENS
Before calling this a motivation problem, inspect what the form and surrounding environment are asking of the applicant.
Before / after (descriptive only)
BEFORE
Earlier conditions
NOW
Current conditions
Adjust a slider or apply a hypothetical change to compare.
SCENARIO LIMITATION
A missing application might reflect a deliberate change of mind, not failed execution.
04 / SIX CASE FILESSame missing action.
Same missing action.
Different possible worlds.
All cases and quoted inner thoughts are invented for teaching; none describe a real client, participant or research subject.
CASE / 01FICTION
The application never sent
“I could imagine doing this job.”
Visible behaviour: No application was submitted before the deadline.
Alternative accounts: The form was confusing, the role's requirements appeared exclusionary, feedback was unavailable, or the applicant decided it was not worth pursuing.
Useful probe: Did simplifying the first page change completion, without changing the applicant's desire?
Open in instrumentCASE / 02FICTION
The course left unfinished
“I really want to understand this.”
Visible behaviour: A learner opens lessons but doesn't complete practice questions.
Alternative accounts: Materials presume missing prerequisites; study time is fragmented; the lessons are irrelevant; or passive review creates false familiarity.
Useful probe: Would a short, aligned practice task with feedback help more than another motivational reminder?
Open in instrumentCASE / 03FICTION
The conversation postponed
“We should talk about what happened.”
Visible behaviour: A needed conversation is repeatedly deferred.
Alternative accounts: The other person may be unavailable, the relationship may not be safe, the words may be unclear, or waiting may be judicious.
Useful probe: Can a written opening reduce ambiguity while respecting boundaries and safety?
Open in instrumentCASE / 04FICTION
The exercise plan abandoned
“Moving more would be good for me.”
Visible behaviour: The planned activity does not occur.
Alternative accounts: Fatigue, unsafe streets, time constraints, pain, unsuitable exercises or changing priorities may all matter.
Useful probe: Would a feasible, accessible activity change participation? Avoid assuming everyone's constraints are negotiable.
Open in instrumentCASE / 05FICTION
The draft never published
“I want my ideas out in the world.”
Visible behaviour: A creator revises the same draft but never posts it.
Alternative accounts: Perfectionism is one possibility; so are privacy concerns, untested claims, audience mismatch or legitimate editorial hesitation.
Useful probe: Does separating a private draft from a clearly labelled provisional publication shift the decision?
Open in instrumentCASE / 06FICTION
The habit that keeps restarting
“This time I'll remember.”
Visible behaviour: A self-chosen routine is attempted intermittently.
Alternative accounts: The routine lacks a dependable cue, competes with obligations, requires unavailable resources or no longer serves the person's aims.
Useful probe: Would a specific situational cue help, or is the goal itself due for reconsideration?
Open in instrumentCOMPARISON / 01—06
These examples share a visible gap. They do not share a demonstrated cause. A good explanation must distinguish rival accounts, specify a measurable action and allow evidence to change its mind.
05 / COMPETING MAPSOther thinkers were
Other thinkers were
here before us.
If IAG cannot add value beyond existing frameworks, that is important—not an inconvenience to hide.
B = MAPESTABLISHED MODEL
Fogg Behavior Model
Behaviour occurs when motivation, ability and a prompt converge at the same moment. This directs attention to an exact target behaviour and practical design conditions. On Fogg's account, a missing prompt alone can explain non-occurrence.
Contrast: IAG foregrounds the tension between stated identity and observed action, but may simply redescribe issues already addressed by motivation, ability and prompts. Fogg is not reducible to “just motivation.”
Read Stanford's accountCOM–BPUBLISHED FRAMEWORK
Capability · Opportunity · Motivation
Michie, van Stralen and West locate behaviour inside a system of capability, opportunity and motivation, the core of the Behaviour Change Wheel. Structural opportunities are not footnotes: they are part of the explanation.
Contrast: IAG's environment, support and friction categories overlap substantially with opportunity and capability. IAG must justify its constructs instead of assuming an original mechanism.
Read the 2011 open-access paperIF → THENRESEARCH SYNTHESIS
Implementation intentions
Rather than only forming a goal, a person specifies a situational cue and a response: “If situation Y occurs, I will do X.” Gollwitzer and Sheeran's 2006 meta-analysis investigated effects on goal achievement across studies.
Contrast: Planning can address some initiation problems, but an if–then statement cannot manufacture safe conditions, free time or institutional permission. The cited effect is not an IAG effect.
Read the 2006 reviewCOMMON QUESTION
Which explanation is doing the real work?
If a simpler model anticipates the same observations with fewer assumptions, it deserves preference until the more elaborate account demonstrates added clarity or predictive usefulness.
06 / HOW TO DISPROVE ITA model that cannot fail
A model that cannot fail
cannot teach much.
This is a proposed research roadmap, not a completed protocol or claimed dataset.
Define one observable act
Specify a target behaviour such as “submitted a completed application within 14 days,” including eligibility, timing and what counts as completion. A broad identity label such as “ambitious” is not an outcome.
Measure constructs independently
Pre-register how identity endorsements, intentions, task barriers, external opportunity, supports and behaviour will be measured. Establish item clarity, construct validity and reliability before claiming a new scale. Distinguish self-reports from verified actions.
Identify the rivals
Measure relevant confounders and compare IAG with parsimonious alternatives, including COM-B, Fogg-style design variables, prior behaviour, skills and changing intentions. Social desirability, selection effects, measurement error and reverse causality can mislead.
Manipulate conditions ethically
Where feasible, randomise an intervention—for example, a clearer application step or optional feedback access—rather than inferring causation from correlations. Check fairness and accessibility; do not withhold essential support.
Pre-register and replicate
Pre-register hypotheses, outcomes, exclusions, planned contrasts, and safeguards. Estimate uncertainty; evaluate out-of-sample performance and subgroup differences. Seek independent replication, including cases where the model fails.
FALSIFIABILITY / QUESTIONS
Questions that should make the author uncomfortable
What if “friction” predicts nothing after COM-B variables are included?
Then the proposed construct may be redundant, badly measured or context-bound. A claim of incremental explanatory value would need to be withdrawn or narrowed.
What if a changed intention, not friction, explains the missing act?
That would be an alternative causal story. Intention should be measured over time; not acting can be a reasonable choice, not a psychological failure.
What if easier tasks do not improve action rates?
Check whether the manipulation actually changed the intended obstacle. If it did, revise the predicted mechanism instead of relabelling every result as “hidden friction.”
Could the framework harm people?
Yes, if employers or educators use an unvalidated construct to rate personality, assign blame or justify pressure. Any tool should avoid diagnosis, coercive use and individualising structural disadvantage.
Research status as of 24 October 2026: no participant study, validation statistics, preregistration, peer-reviewed IAG findings or predictive performance are claimed by this website. The simulated slider labels are qualitative teaching aids, not measures.
07 / TAKE SOMETHING USEFULAn experiment of
An experiment of
your own design.
Write down one situation. The notebook remains in this browser tab, disappears on refresh, and sends nothing anywhere.
WORKSHEET / 01
Separate the story from the observation.
Start with evidence you could show another person. Then list at least two explanations—one of them uncomfortable for your favourite theory.
- Name an action you can actually observe.
- Distinguish your intention from someone else's assumption.
- Describe a condition you could ethically vary.
- Decide what result would change your explanation.
08 / EVIDENCE LEDGERKeep the labels
Keep the labels
on the jars.
A claim ledger and source trail, so a reader can disagree without having to guess what is evidence and what is invention.
Prior behaviour-change literature
Fogg's model, the Behaviour Change Wheel and research on implementation intentions predate this project. Their publication and descriptions can be checked at the linked primary sources.
“Don't infer character from a missed action”
A cautious inference principle in this essay. Helpful for avoiding overconfidence, but not presented here as a tested intervention.
The Identity–Action Gap framework
The I / A / E / S / F vocabulary and four propositions are Sahil's working conceptual synthesis. Construct boundaries, originality, causal pathways and added predictive value require examination.
Six case files and five sliders
The characters, presets, recommendations and responsive messages are invented. The interface uses fixed editorial rules and never calculates outcomes, likelihoods or clinical risk.
Research notes & primary sources
- Fogg, B. J. (2009). A Behavior Model for Persuasive Design; contemporary presentation from the Stanford Behavior Design Lab. Stanford source ↗ The current form uses “prompt”; earlier literature used “trigger.”
- Michie, S., van Stralen, M. M., & West, R. (2011). “The behaviour change wheel: A new method for characterising and designing behaviour change interventions.” Implementation Science, 6, 42. DOI ↗
- Gollwitzer, P. M., & Sheeran, P. (2006). “Implementation Intentions and Goal Achievement: A Meta-analysis of Effects and Processes.” Advances in Experimental Social Psychology, 38, 69–119. DOI ↗
Primary pages were checked on 24 October 2026. They support the descriptions of established models, not validation of IAG. Original articles may be subject to publisher access restrictions.
GLOSSARY / QUICK REFERENCESWords that can
Words that can
hide assumptions.
Intention
A person's stated commitment or plan to perform an action, ideally specifying what and when. It can change.
Behaviour
An observable act defined at a particular level of specificity and time window. “Be disciplined” is not a discrete behaviour.
Friction
This site's provisional umbrella for obstacles such as ambiguity, task load and perceived stakes. Not a validated latent variable.
Support
Available assistance, information, feedback or enabling tools. Support can have unintended costs and is not always welcome.
Confounder
A factor associated with an explanatory variable and an outcome that can distort the apparent relationship.
Falsifiability
The capacity of a clear empirical claim to be contradicted by possible observations; not a guarantee that the claim is false.
What would change
your explanation?
Not every gap needs closing. Some require resources. Others require a better question—or the freedom to choose a different action entirely.