A structured heuristic evaluation for surfacing unconscious assumptions in learning design — before they become friction for the learner.
Screen reader and keyboard-optimised version. This is a fully linear, plain-structure version of this tool with no animations, JavaScript interactions, or visual-only content. All prompts and response fields are presented in reading order. An interactive version with a rotating 3D object and dynamic severity ratings is also available. This page is derived from the case study When Infrastructure Fails, Access Shouldn't.
This tool works best when you treat it as a thinking instrument, not a scoring exercise. There are no correct answers. There is only greater or lesser honesty about the assumptions you brought into this design. The prompts will ask you to be specific. Vague answers produce vague insight. If you find yourself writing in generalities, that is a signal to go one level deeper.
About this tool
How to use this tool
Work through the three movements in order. Be specific — vague answers produce vague insight. If you find yourself writing in generalities, go one level deeper.
Use the severity ratings in Movement 3 to prioritise what to address before your next review. At the end, complete your three outputs: a positionality statement, an examined learner model, and a friction log.
Using this tool with a team
Each evaluator completes Movement 1 independently before the session. Share positionality statements first — before discussing the design. Then complete Movements 2 and 3 independently. Compare friction logs and treat disagreements as data: different ratings reveal different standpoints, and that difference is itself a finding.
The goal of this movement is to make your standpoint visible before it operates invisibly in the design. Every designer brings a perspective. A perspective that has not been examined does not disappear — it shapes every decision you make about what is obvious, what is necessary, and what the learner already knows.
Positionality inventory
Prompts 1.1 – 1.4
1.1
Describe your relationship to this content. How long have you known it? How did you come to know it? What was genuinely difficult about learning it that no longer registers as difficult?
The last question is the most important. What you have forgotten it was hard to learn is often exactly what you have failed to teach.
1.2
What is your relationship to the context this learner works in? Have you done this job, held this role, or navigated this system? If not, what is your understanding of their context based on — and who provided that understanding?
1.3
How similar are you to this learner in terms of educational background, professional experience, language, and familiarity with this subject area? Where the differences are largest, note them explicitly.
1.4
What assumptions did you arrive with today about who this learner is — before you opened this tool? Write them down exactly as they came, without editing or defending them.
Pause here. Read back what you wrote. These are your starting conditions. They will shape what you see and what you miss for the rest of this evaluation.
The goal of this movement is to build an explicit, examined model of your learner — and to be honest about where that model comes from. A learner model built on projection will produce a design that works for the designer. A learner model built on observation will produce a design that works for the learner.
Learner model
Prompts 2.1 – 2.5
2.1
Describe your learner as specifically as you can. Not a demographic profile — a person in a moment. What are they doing immediately before they encounter this design? What do they want from it? What are they worried about? How much time do they have? What else is competing for their attention?
2.2
For each thing you wrote in 2.1 — mark it. Is this something you observed directly or were told by the learner? Something you inferred from data? Something you assumed without evidence? Something you projected from your own experience? Be specific about the source of each element of your model.
This step is uncomfortable. It is meant to be. The discomfort is useful information.
2.3
What does this learner already know that is directly relevant to what you are designing? Where does that knowledge come from — their training, their experience, their role? How confident are you in that assessment, and what is that confidence based on?
2.4
What does this learner need to be able to do when this design is finished — not what they need to know, but what they need to be able to do, in their actual context, under their actual conditions?
2.5
What is the gap between 2.3 and 2.4? That gap is what you are designing to close. State it plainly. Is your current design closing it, or is it producing something adjacent to it?
Pause here. Compare what you wrote in Movement 2 with what you wrote in Movement 1. Where did your own experience, context, or expertise shape the learner model you just built? Mark those places before you move on. They are the locations of your highest-risk assumptions.
Where does the design encode your assumptions rather than serve the learner's goal?
Work through each of the seven assumption domains below. For each domain: identify the assumption your design makes, rate its likelihood and impact, and determine whether it is necessary or incidental. A necessary assumption is one the design genuinely cannot function without. An incidental assumption is a residue of who you pictured when you built it. Most assumptions that feel necessary are incidental.
How to use the severity ratings
Rate each assumption on two dimensions:
Likelihood: How probable is it that this assumption fails for a significant portion of your actual learner population? (1 = low, 2 = medium, 3 = high)
Impact: If it fails, how severely does that impede the learner's real-world goal? (1 = low, 2 = medium, 3 = high)
Priority score = Likelihood × Impact. Scores of 6–9 are high priority and should be addressed before SME review. In the response fields below, note your ratings and your reasoning.
Domain 1Prior knowledge
Curse of knowledge — Camerer, Loewenstein and Weber (1989); schema theory — Bartlett (1932)
3.1a
Walk through your design from the beginning. At each point where the learner is asked to do, understand, or decide something — what do they need to already know in order to do it? List every piece of prior knowledge this design assumes.
Be exhaustive. Include what feels obvious.
3.1b
For each assumption you listed: what is the evidence that this learner has that knowledge? Is it documented, observed, inferred, or assumed?
3.1c — Severity rating
Rate the prior knowledge assumptions you identified. Note your Likelihood rating (1–3), your Impact rating (1–3), your Priority score (L × I), and whether each assumption is necessary or incidental.
3.1d
For each assumption rated 4 or above: is it necessary or incidental? If incidental, what is the lowest-friction path to removing it?
Describe a specific change, not a principle.
Domain 2Motivation and agency
Self-determination theory — Deci and Ryan (1985)
3.2a
What does this design assume about why the learner is here? Does it assume they chose to engage, that they believe the content is relevant to them, or that they are motivated to complete it? State the assumption explicitly.
3.2b
What does this design do — specifically — to earn the learner's engagement rather than assuming it? Where does it acknowledge that the learner's time and attention are a gift, not a given?
3.2c — Severity rating
Rate the motivation and agency assumptions you identified. Note your Likelihood rating (1–3), Impact rating (1–3), Priority score, and whether necessary or incidental.
Domain 3Context of use
Situated cognition — Lave and Wenger (1991)
3.3a
Describe the context you designed against. What device, environment, connection quality, time of day, and level of available attention did you picture? Be specific.
3.3b
How was that context determined? Did you observe it, research it, or assume it? How different might the actual context of use be from the one you designed against?
3.3c
What happens to this design in a context different from the one you pictured — less time, lower connection, more interruption, different device? Where does it break, and for whom?
3.3d — Severity rating
Rate the context of use assumptions you identified. Note Likelihood, Impact, Priority score, and whether necessary or incidental.
Domain 4Language and register
Plain language research; Bernstein (1960)
3.4a
Read the language in your design as if encountering it for the first time, without your current knowledge of the subject. What vocabulary does it assume? What sentence structures? What domain-specific shorthand?
3.4b
Who does this language work for, and who does it exclude? Consider: non-native speakers, people with lower familiarity with the domain, people whose professional register differs from the one used in the design.
3.4c — Severity rating
Rate the language and register assumptions you identified. Note Likelihood, Impact, Priority score, and whether necessary or incidental.
Domain 5Cognitive load
Cognitive load theory — Sweller (1988)
3.5a
What cognitive demands does this design place on the learner at each stage? How many new concepts, decisions, or pieces of information are introduced before any consolidation is offered?
3.5b
What does this design assume about the learner's available working memory? What else might be competing for that capacity — emotional, situational, or task-based demands outside the design?
3.5c
Where does this design ask the learner to hold more in mind simultaneously than is necessary for the learning goal? What is present because it made sense to include, rather than because the learner needs it at this moment?
3.5d — Severity rating
Rate the cognitive load assumptions you identified. Note Likelihood, Impact, Priority score, and whether necessary or incidental.
Domain 6Transfer
Perkins and Salomon (1992); Bransford, Brown and Cocking (2000)
3.6a
Describe the context in which this learning is expected to be applied. How similar is that context to the context in which the learning takes place? What bridges the two?
3.6b
What does this design do — specifically — to support transfer? Where does it connect the learning context to the performance context? Where does it ask the learner to practise in conditions that resemble those they will actually face?
3.6c — Severity rating
If transfer is not deliberately designed for, it is not reliably achieved. Rate the transfer assumptions you identified. Note Likelihood, Impact, Priority score, and whether necessary or incidental.
Domain 7Feedback and self-assessment
Kruger and Dunning (1999); Sadler (1989)
3.7a
What feedback does this design provide to the learner about whether they have understood something correctly? Is that feedback specific enough to be actionable, or general enough to be ignored?
3.7b
What does this design assume about the learner's ability to accurately assess their own understanding? Where does it ask the learner to judge their own competence without giving them the criteria to do so?
3.7c
What does this design assume the learner will do when they have not understood something — will they seek help, re-engage, or move on? What does the design do to support the most useful of those options?
3.7d — Severity rating
Rate the feedback and self-assessment assumptions you identified. Note Likelihood, Impact, Priority score, and whether necessary or incidental.
Return to the three assumptions with the highest priority scores.
For each high-priority assumption, work through the following before accepting your determination of necessary or incidental. The second pass exists to test the feeling of necessity — most assumptions that feel necessary are incidental.
Second pass prompts
SP.1 – SP.3
SP.1
State the assumption and your current determination — necessary or incidental. Now make the case for the opposite determination. What would have to be true for you to be wrong?
SP.2
What evidence would you need to be certain? Do you have it? If not, what is the cost of acting on the assumption without it?
SP.3
If this assumption is incidental and you removed it, what would the design look like? Describe the specific change — not a principle, a change.
Use the fields below to draft your three outputs. These are not deliverables for a client. They are instruments for you — evidence that you did the harder work of designing for someone other than yourself. Bring the positionality statement and learner model to your next sprint. Update them as you learn more. The friction log gets shorter as the design gets better.
Positionality statement
Draw from your responses to 1.1–1.4. State your relationship to the content, to the learner's context, your similarity to or distance from the learner, and the assumptions you arrived with.
Examined learner model
Draw from your responses to 2.1–2.5. Describe the learner in a moment, the sources of each element of the model, their prior knowledge, what they need to be able to do, and the gap your design is closing.
Friction log — ranked by priority
List the assumptions identified in Movement 3, ranked by priority score (highest first). For each, note the domain, the score, and whether the assumption is necessary or incidental. Incidental assumptions rated 4 or above should be addressed before SME review.
Camerer, C., Loewenstein, G., & Weber, M. (1989). The curse of knowledge in economic settings: An experimental analysis. Journal of Political Economy, 97(5), 1232–1254. Foundation for understanding how expertise obscures the learner's experience.
Bartlett, F. C. (1932). Remembering: A Study in Experimental and Social Psychology. Cambridge University Press. Origin of schema theory and its implications for prior knowledge assumptions in design.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic Motivation and Self-Determination in Human Behavior. Plenum. Grounding for the motivation and agency domain.
Lave, J., & Wenger, E. (1991). Situated Learning: Legitimate Peripheral Participation. Cambridge University Press. Foundation for the context of use domain.
Sweller, J. (1988). Cognitive load during problem solving: Effects on learning. Cognitive Science, 12(2), 257–285. Origin of cognitive load theory.
Perkins, D. N., & Salomon, G. (1992). Transfer of learning. International Encyclopedia of Education (2nd ed.). Grounding for the transfer domain.
Bransford, J. D., Brown, A. L., & Cocking, R. R. (Eds.). (2000). How People Learn: Brain, Mind, Experience, and School. National Academy Press.
Kruger, J., & Dunning, D. (1999). Unskilled and unaware of it: How difficulties in recognizing one's own incompetence lead to inflated self-assessments. Journal of Personality and Social Psychology, 77(6), 1121–1134. Grounding for the feedback and self-assessment domain.
Sadler, D. R. (1989). Formative assessment and the design of instructional systems. Instructional Science, 18(2), 119–144.
Norman, D. A. (1988). The Design of Everyday Things. Basic Books. Human factors and usability principles underlying accessible design.
Kahneman, D. (2011). Thinking, Fast and Slow. Farrar, Straus and Giroux. Cognitive bias and its relevance to assumption-making in design.
OCAD University Inclusive Design Research Centre. (Ongoing). Inclusive Design Resources. IDRC. Framework for designing for the full range of human diversity.