Explanation Design Principles Framework

DocID: EXD within FPF Library.

Help a recipient recover and use a warranted connection by choosing the question, account, expression and support that the situation needs.

Anatoly Levenchuk, with AI-assisted development and review.

Release date: 20 September 2026.

License: CC BY 4.0 for original framework content; third-party material retains its own terms.

Table of Contents

Use a public unit for orientation or source access. Search the complete pattern index by the working difficulty or useful result. EXD numbers are stable addresses; they prescribe no order of explaining Work.

Public units

Patterns by working question

Readme

Practical entries

Explanation Design connects a recipient’s question with a warranted account and an expression the recipient can use. An example can expose the needed relation; the recipient’s next contribution can reveal a missing step, change the question or show that the explanation is already sufficient. These returns make the methods a language for explaining rather than a fixed production sequence.

The two situations below show different ways to use that language. The Table of Contents gives direct access to the six methods. The shared section Obtain and Carry the Subject Explanation supplies their common return to the mathematical, causal, technical or advisory grounds. A polished expression cannot supply a missing derivation or justify a recommendation whose grounds remain unknown.

You can ask an assisting agent to explain or apply these methods in the language of your work, without framework jargon, while preserving the distinctions needed for the question.

EXD-CHANGE — Distinguish a changed question from a mistaken inference

Situation: A recipient rejects an answer or offers another result. It is unclear whether they have changed the question, missed a relation or disputed a premise.

Question: What should the next explanation answer, and which part of the existing account can it reuse?

First useful result or blocker: A revised answer to the changed question, a repaired relation under the same question, or the subject input needed to resolve a disputed premise.

Start with: EXD.4 - Co-Construct and Repair an Explanation in Dialogue, using the recipient’s actual contribution. Return to EXD.1 - Establish and Revise What a Recipient Needs Explained if it changes what the answer must explain.

Stop or return: Close when the recipient can use the needed relation. Understanding a recommendation does not require agreeing with it; return a different preference to the decision rather than extending the explanation.

1. Use the current question to recover the relevant subject relation. In the worked exchange, one group has two values with mean 3 and another has six values with mean 8. For the mean across all individuals, reconstruct the sums: 2 × 3 + 6 × 8 = 54; divide by eight to obtain 6.75. Equal weighting of the two group means gives 5.5 and answers a different question.
2. Make that correspondence usable. EXD.2 - Build and Bound an Explanatory Example connects each count and mean to its group sum. EXD.3 - Coordinate the Expression of an Explanation can align those same relations in a table with the weighting condition beside it. Use one sufficient expression; adding a missing mathematical relation is a content change, not merely a new layout.
3. Let the reply determine the return. “Each branch should have equal influence regardless of size” changes the target. EXD.1 establishes that question and the answer becomes 5.5; the 6.75 result remains valid for individual weighting. If the target is still individual weighting but the recipient divides by two because there are two averages, EXD.4 identifies a wrong relation under the unchanged question. Repair what the denominator counts. If the dispute concerns which weighting policy should be used, return that choice to the person responsible for the report.
4. Use a learner’s own explanation when learning is the purpose. EXD.5 - Guide a Recipient’s Own Explanation preserves the initial product, gives feedback about groups versus individuals, and asks for the affected use again. Three values with mean 4 and nine with mean 8 give (12 + 72)/12 = 7; equal group influence instead gives 6. Retain the help used in reaching either answer. Immediate correction does not establish independent performance after a delay.

The same return matters in advising. “Why does this assignment fit my present strengths?” and “Why would it develop my missing capability?” require different grounds. A change from the first question to the second calls for the development rationale. If that rationale is understood but the person prefers another social setting, the remaining question belongs to their choice.

EXD-REPAIR — Repair what the recipient cannot use without rewriting everything

Situation: An explanation fails in use, and plausible repairs include more subject content, another form or additional help. These changes would contribute different things.

Question: Which repair addresses the actual difficulty, and is it worth its burden compared with the available explanation?

First useful result or blocker: A targeted repair, a sufficient explanation to retain, or the missing subject or reader evidence that prevents choosing between the alternatives.

Start with: EXD.6 - Compare Explanations for the Recipient’s Use and Decide Whether a Repair Is Worthwhile, fixing what the recipient needs to recover and the criterion for using it.

Stop or return: Keep a sufficient explanation when no worthwhile contribution is missing. Repair an explicit false claim from its subject basis; obtain reader evidence when the unresolved question concerns recoverability and the observation can change the choice.

1. Identify the relation that the explanation must supply. Consider a constructed explanation of active effort for one person’s task. Its stated accounting rule counts time actually spent on the task. The draft instead says to subtract start time from handoff time. On an uninterrupted task both rules agree, so that example cannot reveal the error.
2. Use an example that separates the rules. EXD.2 can use work from 09:00 to 10:00 and 11:00 to 12:00, with no task work during the intervening hour. Active effort is two hours; elapsed turnaround is three. The subject definition justifies changing the draft’s rule. EXD.3 can then label separate effort and turnaround columns and place the interruption beside the calculation. Relabelling columns without correcting the rule would leave the content defect in place.
3. Respond to the actual remaining difficulty. If a recipient still counts the interruption, EXD.4 uses that contribution to locate the missing distinction. For a learning purpose, EXD.5 asks the recipient to explain what each interval contributes, provides the needed help, and retries the relation on a relevant changed case. Preserve the first response: a prepared reader may have corrected the false instruction from prior knowledge, and a later correct answer may depend on the supplied help.
4. Choose the conclusion the evidence supports. EXD.6 can justify the rule repair through subject inspection. It cannot attribute an unobserved reading-speed or learning advantage to the new table. If an existing prose account already supplies the correct distinction and serves the receiving task, retain it unless the table offers a worthwhile additional use, such as repeatedly locating corresponding intervals. A comparison of forms keeps the selected claims and help comparable.

If the changed task now concerns several people working concurrently, obtain the population accounting rule before extending the one-person explanation. Layout alone cannot decide how person-hours combine. Evaluation of an instructional material as a whole uses HCD.18/.19; a claim about later human performance, transfer or retention uses the relevant HCD method. The explanatory repair can finish without requiring those stronger conclusions.

Using the repertoire

Begin where the needed input is already available. A clear question with a sufficient account may need only an expression adjustment; a missing subject premise returns to the subject practice. Use dialogue when a recipient’s contribution is available and can guide the next move, not as a compulsory questionnaire. For a one-way explanation, state the consequential conditions and provide a usable return to the fuller account.

The Architectural Reference develops the distinctions used above. The use profiles adapt them to instruction, technical and advisory work, and human–AI interaction.

Explanation Design Principles Framework Architectural Reference

EXD.Reference:1 - The practice and the problem it solves

Use Explanation Design when someone needs to recover a connection: why an outcome occurred, how a calculation follows, how something works, why one recommendation follows from its grounds, or how a selected arrangement answers its working problem. The governed practice includes preparing an explanation and performing an exchange in which the recipient contributes. Its first useful result is an account whose relevant connection and conditions the recipient can use, or a question identifying the subject contribution still needed.

A formally adequate account can remain difficult for a particular recipient to reconstruct. A fluent, engaging account can expose a relation clearly while leaving that relation unsupported. The practical problem is to connect subject adequacy with recipient-relative recoverability without treating either as proof of the other.

Explanation Design brings together the relevant contributions of formal explanation and empirical craft around one explaining practice. It asks both what warrants a connection and what enables this recipient to recover it for this purpose. Questions about available attention, preparation, access, assistance and later learning determine different design and assurance moves.

A recipient who only needs an already understood fact can use it directly. Creating a scientific explanation, selecting a management policy, securing agreement and establishing a human learning effect may supply or receive explanation work, but each retains its own question and evidence. A satisfying exchange alone settles none of them.

EXD.Reference:2 - The distinctions that change the work

The subject is what the explanation concerns. The question selects the connection sought; the purpose explains why that question matters here. The account supplies the selected claims, relations and conditions. Its expression is the prose, formula, table, diagram, demonstration or combination through which it is available. The recipient interprets and may contribute under particular preparation, access, assistance and effort conditions.

These distinctions support concrete diagnoses:

FPF C.2.8 characterizes extractable structural information for an expressed account, its expressing form and an observer under selected conditions. For an explanation, state which relations the recipient should recover, what counts as a correct reconstruction, and which preparation, operations, help, access and budget matter. The question is “How much of the selected structure can this recipient recover at these conditions?” A useful local answer can identify correctly reconstructed relations and consequential omissions. A numeric comparison needs a justified scale and comparable conditions.

This structural characterization is relational. The same text may make a connection available to an expert and leave a novice unable to reconstruct it at the same effort. Repetition may assist access or retention while introducing no new relation into the account. A familiar relation still belongs to the recoverable structure; recoverability is not a novelty score. For human recipients, attention and memory inform the use conditions and selected outcome.

Formal computational epiplexity has its own model and computational bound. Moving from it to a claim about human understanding requires a justified correspondence. The practical use of C.2.8 here does not require estimating a person’s understanding in bits. Subject warrant, extraction effort, retention, transfer, usefulness and agreement remain separately answerable.

EXD.Reference:3 - Obtain and Carry the Subject Explanation

Recover the account that supports the connection the recipient needs. It may be a mathematical definition and derivation, a causal finding, a Method account or a recommendation’s rationale. Select its relevant claims while preserving answer-changing conditions, alternatives and limits. The minimum useful account can be very short.

For the mean of a finite list, the subject basis is its sum divided by its count. If A has two values, 2 and 4, its sum is 6 and mean is 3. If B has six values of 8, its sum is 48 and mean is 8. Across individuals, the sum is 54 and the count is eight, giving 6.75. Giving the two group means equal weight instead gives 5.5. Both results have explicit rules; the recipient’s task selects which question to answer.

When a required connection lacks a warrant, state what is missing and which subject work could supply it. An unsupported causal claim needs subject inquiry. An unexplained recommendation needs its decisive premises, alternatives and conditions. A missing mathematical step needs its definition or derivation. An explanation can preserve a useful conditional answer while that input is recovered.

Select a supplying practice by the actual subject. Core’s Method-development patterns supply Method content and the rationale for a pattern language. Decision Design supplies recommendation grounds; Decision Design supplies the advising engagement’s conditions. Human Capability Development supplies representative instructional tasks, feedback tied to their criteria and task-specific assistance. Structure-to-Narrative Rendering supplies a connective account when that form is useful. Research Method Practice supplies the research return when a scientific explanation still needs construction or criticism.

The research return can itself produce useful progress. For example, a research exercise can move from a puzzling observation to a question, two or three rival hypotheses, criticism before experiment and an initial report. Select further empirical work when it can change the answer. The RMP methods qualify a research question, choose a design or sufficient-evidence stop, and revise the affected claims. Communicating an available account does not replace these subject contributions.

This entry carries supplied subject results into explanation work. The six constructive Methods address the remaining explanatory difficulties.

EXD.Reference:4 - A worked exchange and its consequential returns

An analyst asks why the average of two branch means, 5.5, was rejected. The branches contain the eight observations just described. The report initially requires the mean across individual observations. Establishing that question with EXD.1 makes the required rule explicit.

The analyst can add and divide but asks why the branch counts matter. EXD.2 constructs the instance: count multiplied by mean recovers each branch sum; add those sums and divide by the total count. EXD.3 can place count, mean and sum together in a table or retain a sufficient prose calculation. The warranted answer is 54/8 = 6.75, with the connection between aggregation and individual weighting exposed.

Now the analyst says, “I want each branch to have equal influence regardless of its size.” EXD.4 treats this as a changed question. EXD.1 revises the selected weighting, and the explanation returns “Mean of branch means, each branch weighted equally: 5.5.” It retains 6.75 for the original question. Whether equal branch influence is the appropriate management policy remains a decision question. The exchange can close with these qualified answers.

A different contribution needs a different move. If the analyst still intends equal weight per individual but divides the sum of the two branch means by two, the question has stayed fixed and the aggregation relation needs repair. If instructional use warrants a learner contribution, EXD.5 asks what the denominator counts, preserves the initial answer, supplies targeted feedback and uses a changed calculation as a retry. If dialogue is unavailable, the written explanation can state both weighting rules and the condition selecting each. An explicit choice need not be delayed by another clarification.

EXD.6 can compare a proposed repair with the current explanation. Adding the missing sum relation changes the account; rearranging those same claims changes form; supplying a hint changes assistance. If the current account already lets the analyst use the distinction, retaining it is a useful result. The case is a worked construction with explicit arithmetic criteria. It supplies no observation about human attention, learning or a population effect.

EXD.Reference:5 - Why these Methods form a useful language

Question formation and dialogue have different first useful results. EXD.1 establishes or revises what needs explaining, including during preparation without a live partner. EXD.4 uses an actual contribution to decide the next explanatory move or a sufficient close. Keeping both direct entries supports prepared text and responsive interaction.

Examples, expression and a recipient’s own explanation also answer different difficulties. EXD.2 makes a claim–instance correspondence and its boundary available. EXD.3 coordinates the contributions of selected forms. EXD.5 asks the recipient to explain a relation, checks that explanation against the task criterion, and guides correction and retry when needed. EXD.6 connects these possible repairs with the value and burden of the receiving use.

Their useful relations do not force a pipeline. An example can expose an ambiguity that returns to EXD.1. A diagram can reveal a missing subject premise. A learner’s response can supply the contribution for EXD.4. Comparison can select a small expression change or end with the incumbent. Several entries can cooperate in one exchange without becoming compulsory steps or establishing that every explanation has six parts.

The conceptual synthesis learns from Narrativization and Narrative Studies: connect source-faithful construction, cognition and production craft through their actual contributions. It does not import a narrative genre requirement. A derivation can explain without becoming a story, and a story’s memorability does not establish the warrant of its explanation.

EXD.Reference:6 - Assurance, limits and practical consequences

Recognize a promising explanation when its question, needed connection and use conditions are explicit enough to choose a constructive move. Assure the particular result with the relevant criterion: verify the mathematical derivation; check whether the recipient reconstructed the selected relation; inspect a recommendation’s actual premises; or examine a later unaided task when that is the claimed outcome.

A smooth paraphrase, positive reaction or correct number can motivate further inquiry without proving the underlying relation was reconstructed. Conversely, a recipient can understand a recommendation and decline it. Preserve the actual response and the assistance used; distinguish demonstrated results from stipulated examples and expected gains.

The principal trade-offs are useful detail against effort, guidance against the independent contribution sought, responsiveness against interruption, and additional comparison against the value of a sufficient incumbent. Prefer the smallest change that addresses a consequential difficulty. Endless elaboration, repeated elicitation after an explicit answer and unsupported certainty all consume the resource the explanation needs.

The examples favor explicit relations, short exchanges and inspectable outcomes. Tacit, affective, embodied and culturally situated understanding may require other forms and criteria. A recipient’s unfamiliar terminology or access needs are conditions for design, rather than evidence of general incapacity. The human–AI profile further limits what can be inferred from an AI’s output.

In practice the language changes where repair is directed and when work stops. It supplies a warranted connection, a worked correspondence, a coordinated expression, an adapted exchange, a supported retry or a justified comparison. It returns a missing subject basis, a changed decision or a stronger assurance question to the work that can answer it.

EXD.Reference:7 - Framework and subject suppliers

FPF remains the external foundational framework. This edition uses the September 2026 current-pattern text available on 9 September 2026, specifically the named units below; it does not redefine FPF’s kinds or import those bodies as Explanation Design members. The public repository is an access and discovery route. A later file date alone does not replace a selected supplying claim.

The source choices above are use-specific. Reopen a receiving claim when its premise or evidence no longer fits, a necessary source becomes unavailable, or the proposed use asks for a stronger effect. Preserve independently supported accounts and explicit conditional answers.

For source qualifications, citation and edition information, see the Sources and Edition Notes.

EXD.Reference:End

Patterns

EXD.1 - Establish and Revise What a Recipient Needs Explained

Status: Stable

EXD.1:1 - Problem frame

Use this pattern when someone needs an explanation and the needed connection, contrast or use is still unclear, or when a contribution changes what the explanation should answer. The reader may be preparing a lesson, explaining a technical account, discussing advice or assisting a human–AI exchange.

Begin with the recipient’s request and the situation in which its answer matters. Ask the smallest question that distinguishes the consequential alternatives. A usable first result is a question the participants can answer under named conditions, together with the intended use and the subject input still needed.

When a request already identifies a sufficient question, use it. “Show how the total of these eight values gives their mean” can lead directly to the available calculation. Additional elicitation is useful when it can change the next explanation.

EXD.1:2 - Problem

“Why is this wrong?” can hide several questions. The recipient may have made a calculation error, used a different aggregation rule, questioned a premise or asked whether the result serves a decision. Answering the first interpretation that comes to mind can produce a valid account whose decisive relation is irrelevant to the recipient.

Preparation is also easy to misread. A course title, professional role or fluent acknowledgement does not reveal whether the person can use the particular relation needed now. Question work must recover enough of that local understanding to choose the next contribution without turning every request into an assessment session.

EXD.1:3 - Forces

EXD.1:4 - Solution

EXD.1:4.1 - Recover the requested connection and its use

Start with the recipient’s own words and the task at hand. Identify what the answer would let them infer, distinguish, predict, calculate or do. Keep the question and purpose distinguishable: “What is the average across these people?” asks for a statistic; “I need a fair comparison of branches” states a purpose whose rule of fairness remains to be settled.

When alternatives would lead to different answers, offer a brief discriminating clarification. For an average, ask whether each individual or each group should receive equal influence. For a recommendation, ask whether the recipient needs its existing grounds or evidence that following it will develop a capability. Choose the contrast from the actual task and available account.

Accept a clear answer, a corrected paraphrase or a still-unresolved choice. If the recipient cannot choose, state the conditional questions. “We can calculate either summary; we still need to decide which unit should count equally” is a useful result.

EXD.1:4.2 - Locate usable preparation at the needed grain

Use what the recipient has already supplied: a question, attempted calculation, annotation, example, objection or description of intended use. If the next contribution depends on a particular prerequisite and that prerequisite is uncertain, ask for a small relevant inference. “What total does a group of six with mean eight contribute?” tests the needed relation more directly than asking whether the person knows statistics.

Treat the response as evidence about this task under its conditions. If you supply the multiplication step before the person answers, their later answer is assisted. Preserve that distinction when it matters to choosing help or judging independent use. A recipient asking for advice may need only a premise explained; an instructional diagnosis is then unnecessary.

When interaction is unavailable, prepare from the known audience and task. State assumptions that change the answer, expose the relevant alternatives in the text and provide a usable return. A reader can select an explicit branch without first completing a questionnaire.

EXD.1:4.3 - Form the question that the subject can answer

State the target relation, relevant conditions and the needed answer. A short sentence usually suffices: “For these eight values, why does equal weight per individual require weighting the group means by their counts?” Obtain its subject basis through Obtain and Carry the Subject Explanation.

If the requested connection lacks a warranted account, return that question to the relevant subject work. A causal claim may need an investigation; a recommendation may need another premise; a mathematical answer may need a definition or proof. Keep the question available while that work proceeds. Polished wording cannot close a missing warrant.

EXD.1:4.4 - Revise when a contribution changes the target

Compare a new contribution with the current question. Does it reveal a different intended use, change a relevant condition, dispute the subject basis or expose an error under the same question? Make the consequence explicit in the next explanation.

For a changed target, state the new question and what remains valid from the earlier answer. For an error under the same target, retain the target and repair the affected inference. For a disputed premise, recover or revise the subject account. In an exchange, use EXD.4 to choose and interpret the next contribution.

Stop when the question is clear enough to select the required account and a useful next explanation, or when a specific unresolved input has an actionable return. A formal question record is useful only when the work needs one.

EXD.1:5 - Archetypal Grounding

A learner asks why averaging two group means gave the wrong result. Group A has the values 2 and 4; group B has six values of 8. The initial task explicitly asks for the mean across all eight individuals. The learner can add and divide but has used (3 + 8) / 2 = 5.5.

A useful question statement is: “Why do the group sizes matter when we give every individual equal weight?” The account reconstructs the group sums, 2 × 3 = 6 and 6 × 8 = 48, and divides 54 by eight to obtain 6.75. If the missing connection is that count multiplied by mean reconstructs sum, the explanation must expose it.

The recipient then says, “For my report, I want each branch to count equally, regardless of how many people it has.” This explicitly selects a different aggregation rule, so the explainer can answer the revised question directly.

The revised question is the equally weighted mean of the branch means. Its answer is 5.5. Close with the label and condition: “Mean of branch means, each branch weighted equally: 5.5.” The earlier 6.75 remains correct for the average across individuals. The management justification for choosing one summary still belongs to the report’s decision.

This is a constructed case. Its arithmetic fixes the local answer criteria; the conversation illustrates the proposed question work. It supplies no measured learning result.

In a different use, an adviser has explained why a bounded coordination assignment fits Mira’s demonstrated strengths and available protected time. Mira asks why doing it would teach a missing capability. Recover the new learning question and obtain its learning rationale. Repeating the evidence of existing strengths does not supply that additional connection. If Mira instead understands the recommendation and prefers a course’s social setting, the new contribution concerns preference and advising.

EXD.1:6 - Bias-Annotation

The explainer’s expertise can make one interpretation seem inevitable. Begin with the actual requested use and keep a serious alternative visible when it would change the answer. A question bank can reveal possibilities, but its categories may constrain what recipients ask; allow them to supply their own question and reason.

The instructional case makes a correct answer easy to compute. In technical or advisory work, disagreement about the basis can remain unresolved even after the question is clear.

EXD.1:7 - Conformance Checklist

For recognition, locate a consequential ambiguity, missing prerequisite or changed question. If the supplied question already supports a useful answer, direct use is enough.

For assurance of this question result, check that:

• the question and the purpose can be distinguished;
• the selected relation and action-changing conditions are explicit enough to choose a subject account;
• any preparation claim has an actual task basis or is stated as an assumption;
• a new contribution changes the appropriate question, premise, help or interpretation;
• an unresolved input has a specific return, or the current question has a usable next answer.

These checks concern the question and its use. Retention and transfer need their corresponding evidence.

EXD.1:8 - Common Anti-Patterns and How to Avoid Them

EXD.1:9 - Consequences

The next explanation can target a relation the recipient needs and the subject can support. A changed purpose becomes a reason to revise the target rather than a reason to repeat or lengthen the same account.

This work costs attention and can expose a decision or source gap that the explainer cannot settle. The useful stopping point is then a bounded question and return, with any independently supported answer retained.

EXD.1:10 - Architectural Rationale

A question is useful before a text is prepared and within a live exchange. Its first result therefore deserves direct access. The dialogue pattern uses that result and can return a changed question; it also governs contribution size, feedback and repair that question formation alone does not provide.

Separating purpose from question prevents a mathematical definition from silently deciding an aggregation policy. Separating preparation from the answer prevents added help from being mistaken for independent knowledge. Both distinctions change the smallest useful next move.

EXD.1:11 - SoTA-Echoing

Liao et al., 2021 connect question elicitation in concrete tasks with the purposes and explanatory solutions that can answer them. Adapt that move to the recipient’s intended inference and actual subject. A generic catalogue is a useful prompt; one discriminating task question can be sufficient.

Schmude et al., 2025 used free questions, a question bank and clarification in two algorithmic-decision scenarios with 24 participants. Their exploratory work supports allowing recipients to state why information matters and to go beyond prepared categories. It establishes neither completeness of the bank nor a general comprehension effect.

Galbraith and Khalifa, 2026, §§2 and 5 propose an interrogative account of understanding while leaving important epistemic conditions open. Use the question discrimination; subject warrant still comes from the relevant account. The procedure here is a bounded explanation-practice adaptation.

At the same practical grain, “Why is 5.5 wrong?” and “Which unit receives equal weight?” can both be brief. The latter distinguishes the live alternatives in the worked case. That conceptual advantage does not establish a measured reduction in conversation time. Reopen the adaptation when a recurring use cannot express its needed distinction through the proposed question work.

EXD.1:12 - Relations

EXD.4 uses and revises the question during an exchange. The supplied-account entry obtains the answer’s subject relations and conditions.

FPF FPF A.2.9 supplies communicative-use distinctions where that analysis is needed. FPF C.2.8 characterizes what a particular observer can recover from an account and form under selected conditions; it does not choose the question’s purpose.

EXD.1:End

EXD.2 - Build and Bound an Explanatory Example

Status: Stable

EXD.2:1 - Problem frame

Use this pattern when a recipient needs to connect a principle, definition or claim with a particular instance. A worked calculation, comparison, analogy or counterexample can expose the relation that a statement alone leaves hard to use.

Start by naming the relation the instance must show. Choose values or circumstances for which the intended rule and a plausible alternative lead to a consequential difference. The first useful result is a worked instance, its correspondence to the claim and the condition that bounds its use.

A recipient who already has a sufficient direct account may need no further example. When the claim itself lacks a warrant, obtain that subject basis before choosing an illustrative instance.

EXD.2:2 - Problem

An attractive example can make a statement memorable while hiding why it applies. A recipient may copy the numbers or surface features, infer a rule from one convenient success, or carry an analogy beyond its justified correspondence.

The example must therefore do two jobs together: make the needed connection recoverable and discriminate its valid use from a consequential alternative. A correct final answer alone does not establish either.

EXD.2:3 - Forces

EXD.2:4 - Solution

EXD.2:4.1 - Select the relation and a discriminating instance

Recover the current question through EXD.1 when necessary. Obtain the definition, derivation, causal account or rationale through the supplied-account entry.

State what the recipient must connect. In a pooled mean, each group mean multiplied by its count reconstructs the group sum; the group sums and counts combine separately. Select small values that preserve this relation while exposing a likely competing rule. Unequal counts and unequal group means distinguish pooling individuals from giving groups equal influence.

Use a real case when its evidence and complexity fit. A constructed case is useful when its assumptions and answer criterion can be made explicit. Identify which it is.

EXD.2:4.2 - Work the correspondence, not only the answer

Show the intermediate quantities, relations and conditions needed to obtain the result. Connect each significant step to the claim: which quantity instantiates which part, why this operation is legitimate, and what condition makes the inference usable.

For a worked calculation, name the units and what each numerator or denominator counts. For a causal example, distinguish observations from an inferred mechanism. For an analogy, identify the relation that corresponds and a feature that does not transfer. A picture of two similar objects cannot supply that correspondence by itself.

If a recipient contribution would help locate the difficulty, first invite a small unassisted mapping: “Which quantity here is the total represented by the group mean?” Retain the response. Then provide targeted component guidance, such as identifying the count and asking how to recover the sum. A recipient with little relevant preparation may instead need the worked relation first.

EXD.2:4.3 - Vary the condition that matters

Choose a changed case that distinguishes the intended relation from the plausible alternative. Ask what should stay the same, what should change and why. Keep irrelevant complexity low enough for the relation to remain visible.

Inspect the rule as well as the number. Equal group counts allow an unweighted average of the group means to agree with the pooled mean. Equal group means can also make the two numerical answers coincide despite unequal counts. The governing target still determines which weighting rule the explanation should express.

Use EXD.5 when a recipient’s own account and supported correction serve a learning purpose. Use EXD.4 when the response changes the question, premise or next explanation.

EXD.2:4.4 - Bound the example and stop at its useful result

State the admissible correspondence and the nearest consequential limit. A branch can be a unit in one aggregation question and contain several equally weighted people in another. The same data then support different summaries.

Retain the smallest example or pair that supplies the needed distinction. Add another case only when it can reveal a remaining wrong generalization or support the intended next use. An example is evidence of the particular relation it demonstrates; claims about learning from it need the recipient evidence appropriate to that claim.

EXD.2:5 - Archetypal Grounding

A recipient can average a list but combines group means without considering counts. The question is the mean across all individuals.

The mapping is explicit: count × mean reconstructs each group sum; sums add to 54; counts add to eight; mean is total sum divided by total count. Averaging 3 and 8 instead gives equal weight to the two groups and yields 5.5.

Now reduce B to two values of 8. Both groups contain two values. Their combined sum is 22, count four and mean 5.5, matching (3 + 8) / 2. Ask why the shortcut agrees: equal group counts make equal group influence correspond to equal individual influence.

Finally, keep unequal counts but give both groups mean 3. Both summaries are 3. Ask whether that numerical coincidence establishes the same weighting rule. It does not: the means coincide, so changing the weights leaves that number unchanged. This case blocks the overgeneralization that unequal counts must always produce different numerical answers.

The useful outcome is a justified choice of rule, including when a convenient shortcut produces the same result. All three cases are constructed mathematical instances. They demonstrate the stated relations, not an observed learning advantage.

For a constructed technical example, a manual states that a batch of twelve parts costs 30 currency units. A guide promises to explain the cost per part without requiring that manual. It correctly repeats the batch price but omits the batch size. The quoted claim agrees with the source, yet the reader cannot recover the promised calculation: 30 / 12 = 2.50 per part. Supplying the batch size repairs that omission. In a second guide the calculation is fully shown as 36 / 12 = 3, but the price 36 is attributed to the same manual. The calculation can be followed; its price claim contradicts the source. These cases distinguish agreement with a source from recovery of a promised contribution. Both questions matter when both defects occur.

EXD.2:6 - Bias-Annotation

Expert familiarity can hide intermediate relations a recipient cannot yet supply. Inspect the needed mapping at the actual task grain. A familiar story may also smuggle in a causal or normative assumption absent from the subject account; retain only the correspondence the explanation can justify.

EXD.2:7 - Conformance Checklist

Recognition asks whether a particular instance or comparison can make the needed connection usable.

For assurance, inspect whether the example has a warranted basis, an explicit correspondence, a correct worked result and an action-changing boundary. Check that any recipient response retains the help available for it. A matching answer with a wrong rule remains a reason to inspect the explanation.

EXD.2:8 - Common Anti-Patterns and How to Avoid Them

EXD.2:9 - Consequences

The recipient can inspect how the claim applies and where a plausible alternative diverges. The example costs preparation and attention; selective variation controls that burden. A good correspondence may still need different support for another recipient or a more complex subject.

EXD.2:10 - Architectural Rationale

Example construction has a reusable result of its own: an instance and its bounded correspondence. Representation work can express that result; instructional practice can ask the recipient to use it. Keeping those contributions distinct helps the author identify whether a failure lies in the example’s relation, its expression or the available help.

EXD.2:11 - SoTA-Echoing

The 2023 relational-mapping study compared principle–example and example–example mapping in probability topics, including free mapping followed by guided component mapping. Adapt the explicit correspondence and targeted-guidance procedure. Its findings do not make one mapping format universally superior for conceptual or structural learning.

Human Capability Development’s representative-task construction supplies worked reasoning and consequential variation. Fiorella, 2023 further distinguishes the contributions and guidance needs of generated explanations, visualizations and enactments.

At comparable calculation complexity, the unequal-count case distinguishes rules that the equal-count case conceals. Retain both when that distinction matters; reject decorative extra instances that add no needed relation. Reopen the construction when a recipient’s use reveals a different plausible rule, missing component or unsupported analogy.

EXD.2:12 - Relations

EXD.1 selects the target question. EXD.3 coordinates the expression of the worked correspondence. EXD.5 guides a recipient’s product and correction; EXD.6 selects a worthwhile example change or retains a sufficient one.

The instructional profile preserves the HCD task, feedback and assistance returns. The supplied-account entry retains the subject warrant.

EXD.2:End

EXD.3 - Coordinate the Expression of an Explanation

Status: Stable

EXD.3:1 - Problem frame

Use this pattern when a warranted account is available but the recipient has difficulty following its relations through the chosen words, symbols, diagram, demonstration or combination. It also helps when preparing an explanation whose several representations must work together.

Begin by selecting the relation the recipient must recover and the contribution each representation should make. Construct the correspondence at the point where the recipient needs it. The first useful result is an explanation in one sufficient form or a coordinated combination, with its conditions and losses visible.

A missing premise or derivation is a content problem. Obtain that contribution before treating the difficulty as a presentation choice.

EXD.3:2 - Problem

More representations can add search and translation work. A formula, table and diagram may use different labels, omit different conditions or require the reader to reconstruct their correspondence unaided. Simplification may also remove the very condition that makes the answer correct.

A useful expression exposes the selected relation while keeping its subject meaning intact. It does not earn its place by being shorter, more visual or more interactive alone.

EXD.3:3 - Forces

EXD.3:4 - Solution

EXD.3:4.1 - Select each contribution

Name the current question and the claims its answer requires. FPF C.37 supplies the general representation-selection and co-use question. Apply it here to the explanation: what must the recipient locate, compare, connect or enact?

Choose one sufficient representation when it serves that use. Add a representation only for a contribution that matters, such as seeing count/mean/sum correspondence while using a formula. Identify the claims and conditions each form carries and what it omits.

Separate construction choices. Adding the fact that count multiplied by mean reconstructs a sum changes explanatory content. Laying out that already stated relation in a table changes expression. A prompt or worked step supplied during use changes assistance. Preserve these differences when judging the result.

EXD.3:4.2 - Make the correspondence available where it is used

Use consistent names, units and referents. Put a label beside the quantity or feature it names, or give an unambiguous cross-reference when the medium prevents proximity. Make a line, arrow or spatial arrangement’s meaning explicit when it could denote different relations.

In human instructional use, select signaling, proximity, meaningful segmentation or preparation of a needed concept for the actual difficulty. Signal the next relation rather than highlighting everything. Keep a meaningful inference together within a segment. Let a recipient revisit an earlier step when the medium allows it.

Coordinate simultaneous channels so their contributions can be used together. Words may explain what a diagram’s relation means; duplicating a long text while narration and animation compete for attention may add burden. Select the arrangement under the actual task, recipient and medium conditions, retaining text when it provides needed access or reference.

EXD.3:4.3 - Preserve a usable reading path

Make the current question, selected relation and answer reachable in the chosen form. Expose the condition that changes the answer at the point it becomes consequential. A short answer may point to a fuller derivation; the recipient must know when that return is needed.

Provide usable labels and correspondences through the actual available channel. If color carries a distinction, include another cue that lets a recipient recover it without that color. If a static diagram cannot show a temporal dependence clearly, choose a warranted sequence or accompanying account. Where another modality is unavailable, retain the contribution in a form the recipient can use.

EXD.3:4.4 - Inspect the specific gain and loss

Try the intended inference or comparison under the stated conditions. EXD.6 helps choose a bounded comparison when its result can change which expression is retained.

Keep content, expression and help changes identifiable. A promising arrangement can be retained provisionally on a stated design reason; an observed advantage requires the corresponding evidence. Restore an omitted condition or return to the source when the representation invites a different claim. Stop adding forms when one arrangement is sufficient.

EXD.3:5 - Archetypal Grounding

The account explains the mean across all eight individuals in two groups. Its prose form states:

“Group A has two values with mean three, so its sum is six. Group B has six values with mean eight, so its sum is forty-eight. The group sums add to fifty-four and the counts to eight. Dividing total sum by total count gives 6.75. Averaging the two means instead gives equal influence to the groups and yields 5.5.”

A table candidate arranges those same claims:

Add the same boundary sentence beside it: “Averaging the two group means instead gives each group equal influence and yields 5.5.” The column labels expose the correspondence and preserve the distinction between group and individual weighting.

The table is a form candidate for this selected claim set. Removing the boundary sentence would lose a relevant claim. Beginning with only a formula and adding the complete verbal derivation would be a content change as well as a presentation change.

For a recipient who cannot yet use the mean-to-sum relation, point to one group’s count and mean and work that relation. Preserve the initial response separately from the response after help. A successful assisted calculation is not evidence that the original table independently supplied everything the recipient needed.

In a technical account, a box grouping three checks can be mistaken for a statement that they are parts of one Method. Label the box’s actual meaning, such as “available checks,” and state the question each check answers. If the grouping adds no useful relation, a direct list may suffice.

EXD.3:6 - Bias-Annotation

An author may favor a visual or terse form because it fits their own expertise and tools. Check the actual recipient’s access and usable operations. Human multimedia results constrain a human instructional use; shared words such as attention do not establish the same effect in an AI recipient.

EXD.3:7 - Conformance Checklist

Recognition concerns a recoverability or coordination difficulty in expressing an available account.

For assurance, check that each representation has a selected contribution; labels and correspondences support the required inference; decisive conditions remain present; added content or help is identified; and any observed gain is attributed only to the comparison that supports it. A sufficient single form remains a legitimate outcome.

EXD.3:8 - Common Anti-Patterns and How to Avoid Them

EXD.3:9 - Consequences

The recipient can locate and connect the needed relations with less avoidable translation work in the designed arrangement. Whether effort actually decreases remains an empirical question for that use. Coordinated forms require maintenance: a changed value or condition must remain consistent wherever it is expressed.

EXD.3:10 - Architectural Rationale

Expression construction specializes general representation selection with explanation-specific correspondence, signaling and pacing choices. The account’s subject warrant remains external to those choices. Examples determine what instance and boundary to show; this pattern determines how their existing claims become usable in the chosen medium.

EXD.3:11 - SoTA-Echoing

Mayer, 2024 supplies human multimedia-design operations including proximity of corresponding words and graphics, signaling, segmentation and preparation of needed concepts. Adapt the operation that addresses the actual instructional difficulty. The principles’ learner, task and presentation conditions remain relevant.

Cromley and Chen, 2025 synthesize 181 studies from 92 articles in Mayer’s 1990–2022 research corpus and examine moderators. Use that evidence to keep choices conditional. It does not establish a universal winner among prose, table, diagram or combined media.

FPF C.37 and C.37.CF preserve the general selection and content/form comparison. In the worked case, prose and table carry the same selected relations. The table makes a correspondence spatially explicit; its practical superiority is still a comparison question. Reopen when medium, recipient preparation, access or the relied-on content changes.

EXD.3:12 - Relations

EXD.2 supplies a worked correspondence and boundary. EXD.4 can expose the local expression difficulty during use. EXD.6 compares a candidate with a sufficient incumbent.

FPF C.37 governs representation selection and co-use. FPF C.2.8 characterizes selected structural recovery under explicit conditions. The instructional profile qualifies human learning and assistance uses.

EXD.3:End

EXD.4 - Co-Construct and Repair an Explanation in Dialogue

Status: Stable

EXD.4:1 - Problem frame

Use this pattern while explaining with another participant when their contribution can change the question, the account, the help needed or the next useful inference. A teacher, specialist, adviser or assisting agent may offer an explanation; either participant may supply a relevant relation or correction.

Begin with one contribution that answers the current question at a manageable grain. Leave room for the recipient’s next contribution and use it to choose a continuation, repair, source return or close. The first useful result may be a recovered relation, a revised question, a located disagreement or a specific unresolved subject input.

A clear lookup answered sufficiently by a short message can end there. Select a diagnostic exchange when its different possible answers would change the next explanation.

EXD.4:2 - Problem

A sequence of alternating messages can leave the explanation unchanged. The explainer may treat every objection as a lack of knowledge, respond to every acknowledgement as proof of understanding, or repeat a valid account after the recipient changes the question.

Explanatory dialogue needs a constructive link between the recipient’s contribution and the next move. It also needs a stopping point. Keeping someone talking, agreeing or completing exercises is not the purpose of every explanation.

EXD.4:3 - Forces

EXD.4:4 - Solution

EXD.4:4.1 - Offer a contribution tied to the current question

Use the available question and qualified subject account. If the target is uncertain, EXD.1 helps identify the consequential distinction. Obtain and Carry the Subject Explanation supplies the subject return.

Offer a claim with the ground, condition or correspondence that makes the next inference possible. Choose the grain from the task and the recipient’s usable preparation. A short derivation can be one contribution; a demonstration may need one meaningful intermediate point. A fixed number of sentences or turns cannot determine that grain.

Make the expression accessible through the participants’ actual channel. A pause, an editable example, a text field or a choice to inspect the underlying account can let the recipient contribute. If the channel cannot support an exchange, prepare a usable explanation with explicit conditions and a return instead.

EXD.4:4.2 - Interpret the recipient’s contribution in context

Attend to what the contribution changes. “I could not hear the number,” “I do not see why you multiplied,” “I want a different average,” and “I understand, but I reject that policy” invite different responses.

Acknowledge the concrete point before extending the account. Use feedback about contact or perception to restore the channel or expression. Use a displayed relation or attempted inference to inspect understanding. Treat a premise challenge as a subject question and a changed intended use as a possible question revision. Agreement or willingness to act has its own meaning in the situation.

If the recipient says that the explanation misrepresents their concern, recover the particular condition or claim they are asking you to address. Use a clear correction directly. When its meaning remains consequentially uncertain, restate what you have understood and invite correction before arguing against it. Acknowledge a point you actually accept and state where the disagreement remains. Qualify genuine uncertainty; adding reassuring words or weakening a warranted claim does not supply this repair. If the exchange itself has become the obstacle, a different time, channel or a pause may be the useful next move. Continuing to explain requires a current explanatory purpose.

Several interpretations can remain plausible. In that case, ask one small diagnostic whose alternatives lead to different repairs. “Which quantity should receive equal influence here?” distinguishes a changed weighting target. “What sum does this group mean represent?” distinguishes a missing mean-to-sum relation. The diagnosis remains provisional until the contribution supports it.

EXD.4:4.3 - Change the part that prevents the next useful move

Choose the repair from the located difficulty.

When assistance is added, preserve what was available before and after it if the later judgement depends on that distinction. Supplying a step and hearing it repeated can establish a supported response in that exchange; it does not establish later independent use.

Either participant may provide a useful repair. Check a correction against the subject basis and incorporate it when warranted. Explainer and recipient are positions in this exchange, not fixed allocations of knowledge.

EXD.4:4.4 - Close with the result the exchange supports

Name what can now be used: a distinction, a correctly qualified result, a source return or an unresolved conditional choice. Use one relevant inference or next action as a check when it can still change that conclusion.

A sufficient explanation can close without agreement or recall practice. If the recipient’s goal is learning, the instructional use may additionally require supported practice or a later independent case. Select that work from its purpose and available evidence.

When the exchange must stop with a gap, keep the supported part and state the missing input. “Both calculations are available; the report still needs a decision about which unit receives equal weight” is an actionable ending.

EXD.4:5 - Archetypal Grounding

The following constructed exchange uses two groups. A contains 2 and 4, with mean 3; B contains six values of 8, with mean 8. The initial task asks for the mean across all eight individuals.

The repair changes the target and its answer. A longer version of the pooled-mean proof would still answer the initial question.

Compare a different recipient contribution: “I still need the average across individuals, but I do not see why multiplying the count by the mean gives the total.” The weighting target is stable. Show that six values with mean eight have total 48, using the definition of a mean as sum divided by count. If a supported retry is useful, ask for the total of four values with mean five. The answer 20 demonstrates that response under the available help; subsequent independent use requires its own observation.

In advisory use, a recipient may say, “I understand why the assignment fits my present strengths. I prefer the course because I value its social setting.” Close the explanatory distinction and return the preference to the advising decision. Asking the recipient to reproduce the recommendation would add no useful explanatory result.

EXD.4:5.1 - Recover the concern before adding another explanation

An engineer explains that a proposed service reduces response time from eight minutes to five. A colleague replies: “You keep treating my objection as resistance to speed. I need somebody available throughout the night.” The comparison of response times does not answer that staffing concern.

Use the colleague’s stated correction: “You need continuous coverage; the five-minute figure applies only while an operator is present.” The explanation now exposes its condition. If coverage remains the question, return it to staffing and resource work; another account of the speed calculation will not supply an operator. If the colleague instead asks why the time falls to five minutes, provide that derivation under its stated conditions. The two continuations answer different contributions. Neither establishes agreement to deploy the service.

EXD.4:6 - Bias-Annotation

A helpful tone or frequent acknowledgements can make an exchange feel successful while a consequential relation remains unrecovered. Conversely, an objection can reveal understanding and a legitimate alternative. Interpret each signal with its content and task.

An explainer may also protect their prepared account from revision. Make room for a correction from either participant, while preserving the subject check that distinguishes a warranted correction from a fluent assertion.

EXD.4:7 - Conformance Checklist

Recognition concerns an exchange in which the next contribution can change the explanation’s usefulness. A message count or a conversational interface alone does not establish that situation.

For assurance of the performed explanatory result, inspect the contribution and response that matter:

• the offered account answers an identifiable current question;
• feedback is interpreted at the level its content supports;
• a consequential ambiguity receives a discriminating question or remains explicitly unresolved;
• the next move addresses the located difficulty or revised question;
• added help and independent response remain distinguishable when later reliance uses that difference;
• closure preserves the warranted result, its conditions and any actionable return.

These criteria support a bounded reading of the exchange. A learning or transfer claim requires evidence beyond a fluent closing message.

EXD.4:8 - Common Anti-Patterns and How to Avoid Them

EXD.4:9 - Consequences

The explanation can change while the difficulty is still local to a contribution. The participants can revise the target, preserve a useful earlier account and stop with a result whose limits they can name.

Dialogue demands attention and access to the relevant subject basis. It can expose a disagreement without resolving it. Those outcomes are useful when they identify the next needed work, but they do not support an unconditional claim of improved understanding.

EXD.4:10 - Architectural Rationale

The governing relation is between a contribution, the recipient’s response and the next useful move. A prescribed lecture-then-questions sequence cannot represent every consequential exchange. Prepared accounts and examples remain useful contributions within that flexible organization.

Question formation remains separately accessible because it is also useful before an exchange. Subject inquiry remains separately responsible for warrants. Keeping those returns explicit lets dialogue change the right question, claim or assistance without taking over the entire instructional or advising practice.

EXD.4:11 - SoTA-Echoing

The current Social Explainable AI chapters on Practices, §5.3, Incremental Communication, §§12.3.1–12.3.3 and Adaptation, §§13.2–13.3 connect explaining to situated, jointly adjustable contributions. Adapt that line into the contribution–diagnosis–repair decisions above, keeping learning claims at their studied scope.

Fisher et al., 2022 analyzed eleven natural consultations in which phases alternated and both sides contributed to organization. Use that as a bounded counterexample to a necessary fixed phase order. It does not select one universal conversational schedule.

Axelsson, Buschmeier and Skantze, 2022, §2 distinguish feedback concerning contact, perception, understanding and attitude. Adopt that distinction to avoid choosing a repair from an acknowledgement alone; a signal’s interpretation remains contextual.

Fichtel et al., 2025, §§5–6 found that enhanced prompting elicited more co-constructive behavior in a bounded Llama 3.1 70B study without a significant mean objective-understanding advantage. Keep that limit when designing human–AI explanatory interaction. More interaction is not an effectiveness measure.

For repairing a misrepresented concern, adapt the explicit acknowledgement and engagement examined by Yeomans et al., 2020, Study 4 and §6.2. Their written-message study concerns perceived receptiveness, persuasiveness and intentions to collaborate; it does not establish subject correctness, understanding or actual later cooperation. The contribution used here is making the addressed concern inspectable before adding a rebuttal. Against applying a linguistic recipe, keep actual agreement and uncertainty truthful and retain the subject criterion. A polite response can still answer the wrong question. Reopen this repair when the response shows that the concern remains misrepresented or the exchange no longer serves an explanatory purpose.

At the same local opportunity to respond, repeating the pooled-mean derivation answers the old question. Using the recipient’s explicit change of weighting rule supplies the requested branch-mean answer and preserves the earlier result under its original condition. This comparison explains the proposed move in the worked case. Reopen it when actual recurring interactions expose a different diagnostic or repair need.

EXD.4:12 - Relations

EXD.1 supplies and revises the question. The supplied-account entry provides the subject basis and its return.

FPF FPF C.2.8 qualifies structural recovery at an identifiable point in the exchange by the account, expression, observer and conditions. Changes after help receive their changed conditions. Human capability development supplies task-specific assistance, feedback tied to the task criterion and retry when the receiving purpose is instructional.

EXD.4:End

EXD.5 - Guide a Recipient’s Own Explanation

Status: Stable

EXD.5:1 - Problem frame

Use this pattern when an instructional purpose requires the recipient to construct and use an explanation: connect a worked step to a principle, explain an error, map a diagram or show how an operation follows. The learner’s own verbal, visual or enacted product can reveal a relation that passive acknowledgement leaves unknown.

Begin with the relation the learner should make explicit and the product that would display it. Give a focused prompt with assistance appropriate to the task. The first useful result is the learner’s product, any supported correction and the assistance still needed.

A recipient who only needs an available fact or the grounds for advice may need no learning exercise. Select generation because its result can serve the receiving purpose.

EXD.5:2 - Problem

“Explain it in your own words” can invite paraphrase without the needed relation. A learner may generate a plausible but wrong account, copy a worked answer or fail because a prerequisite was never available. Supplying a correction and asking for repetition can then conceal how much help produced the result.

Useful guided generation specifies what the product should expose, preserves the first response and connects feedback to an affected retry.

EXD.5:3 - Forces

EXD.5:4 - Solution

EXD.5:4.1 - Select the target and the product

Recover the learning purpose, current question and warranted subject relation. Choose a product that makes that relation inspectable: a short explanation of one step, labels and arrows with stated meanings, a worked correspondence or an enactment with an explanation of what each action represents.

State a criterion for the relation, not a preferred phrase. For the pooled mean, the product should show what the denominator counts and how each group’s sum is recovered. Fluency, length or agreement with the teacher cannot replace that criterion.

Use EXD.2 when the instance and contrast need construction. Keep legitimate tools available when the intended performance allows them.

EXD.5:4.2 - Give a focused prompt and preserve the first response

Select assistance from the learner’s usable knowledge of this task. A worked model, partial representation, cue or small open prompt can each be appropriate. If an initial unassisted response can change the diagnosis or help, obtain and retain it before giving the solution.

Ask for the relation that matters: “What does your denominator count, and what should it count for this question?” is more diagnostic than “Explain the answer.” Let the learner express a correct relation in their own suitable form.

Where the learner lacks a necessary component, supply that component and make the assistance visible. An unproductive unsupported attempt is not required before every explanation. Human Capability Development’s task and assistance methods supply the broader decision about support and progression.

EXD.5:4.3 - Locate the error and support its correction

Compare the product with the selected criterion. Distinguish a wrong relation, an absent prerequisite, an expression difficulty and a changed question. A learner using the mean of branch means may be correct after the question changes to equal branch influence; EXD.1 and EXD.4 handle that return.

For an error under a fixed question, guide the learner to identify the wrong relation, explain why it fails here, construct the corrected relation and describe its use. Provide the missing ground or component when needed. Tie the feedback to the task criterion: “Your denominator counts two groups, while this question counts eight individuals” identifies the discrepancy.

Preserve the initial product and what was supplied in correction. A supported explanation is a result under those conditions.

EXD.5:4.4 - Ask for the affected use again

Choose a retry that can reveal whether the corrected relation is usable at the intended grain. Change values or a consequential condition while avoiding irrelevant difficulty. Specify what help remains available. The next response can support a more limited or a stronger conclusion depending on those conditions.

If the error persists, revise the diagnosis or assistance rather than simply repeating the same prompt. If the learner now uses the relation adequately, stop the immediate correction. Select later retrieval, practice, fading or transfer work only when the learning purpose requires it, using the relevant HCD methods and evidence.

EXD.5:5 - Archetypal Grounding

The fixed task asks for the mean across all individuals. Group A has two values with mean 3; group B has six values with mean 8. The learner’s first product is “(3 + 8) / 2 = 5.5, because there are two averages.”

Preserve that response. Ask, “Which things receive equal weight in this calculation?” The learner says, “The two groups.” Ask, “Which things should have equal weight in the question?” If the learner identifies the individuals, guide the mismatch: the denominator two counts groups, while the target contains eight people.

The corrected explanation should state: “Each mean summarizes a different number of people. The group sums are 2 × 3 = 6 and 6 × 8 = 48. Divide 54 by the total count eight to get 6.75.” If the learner cannot reconstruct a group sum, supply that relation explicitly and retain it as help.

A retry uses three values with mean 4 and nine values with mean 8. Ask for the pooled mean and why its denominator has that value. The correct result is (3 × 4 + 9 × 8) / 12 = 7. A response that uses the total count and reconstructs both sums meets the stated criterion under the available assistance. It does not yet establish independent performance after a delay.

For a contrasting target, ask for equal influence of the two group means. The result is then 6. The learner should preserve the target-dependent rule, not replace one memorized shortcut with a claim that group means must always receive count weights.

These are constructed prompts and answer criteria. The actual learner’s responses determine which help and further work are warranted.

EXD.5:6 - Bias-Annotation

Verbal fluency can conceal a wrong relation, while a brief diagram or enacted explanation may show the relation correctly. Judge the selected product by its subject correspondence and use. A teacher’s preferred terminology is necessary only where it carries a consequential distinction.

EXD.5:7 - Conformance Checklist

Recognition concerns a learning use for which the recipient’s product can expose or develop a needed relation.

For assurance, check that the target relation and answer criterion are explicit; the prompt fits the available preparation; the first response and help are distinguishable; correction addresses the actual error or missing component; and the retry tests the affected use. State whether its result is assisted, immediately independent or supported by a later observation.

EXD.5:8 - Common Anti-Patterns and How to Avoid Them

EXD.5:9 - Consequences

The learner’s product becomes usable evidence for targeted support and correction. Producing and inspecting it costs effort, so a focused relation and a meaningful stop matter. Successful supported correction can be valuable even while retention and broader transfer remain open.

EXD.5:10 - Architectural Rationale

This pattern owns explanation-specific product and prompt construction. Human Capability Development supplies broader task design, feedback, assistance and progression. The learner’s product and its discrepancy connect that broader instructional work to the explanatory contribution.

EXD.5:11 - SoTA-Echoing

Fiorella, 2023 distinguishes explaining, visualizing and enacting as sense-making activities with different guidance needs. Adapt focused prompts, partial support and feedback to the relation to be generated. Production alone is not a guarantee of learning.

Zhang and Fiorella, 2024 supply error-specific self-explanation guidance: identify the wrong relation, explain the error and explain the corrected relation’s use. The second experiment’s analysis concerned 73 students who had made errors; correction and near-transfer benefits did not establish a far-transfer advantage. Carry the scaffold with that evidence boundary.

Tetzlaff et al., 2025 qualify assistance by prior knowledge and conditions across their meta-analysis. Keep support task-specific instead of prescribing universal fading. In the worked case, a denominator prompt discriminates the known error more directly than another general request to explain. Reopen if the actual product exposes a different missing relation or assistance need.

EXD.5:12 - Relations

EXD.2 supplies the instance and boundary. EXD.3 coordinates a visual or symbolic product. EXD.4 interprets a contribution that changes the target or repair. EXD.6 selects a worthwhile prompt or explanation change.

The instructional profile names HCD’s task, feedback, assistance and material-evaluation returns and the evidence needed for later learning claims.

EXD.5:End

EXD.6 - Compare Explanations for the Recipient’s Use and Decide Whether a Repair Is Worthwhile

Status: Stable

EXD.6:1 - Problem frame

Use this pattern when a proposed explanation change competes with an available explanation and the comparison can change what you retain, repair or stop. The issue may be a missing relation, costly reconstruction, inappropriate help or a change after an already sufficient answer.

Begin with the recipient’s required inference or action and the usable incumbent. Select the smallest consequential difference. The first useful result is a justified repair, a bounded provisional choice, retention of the incumbent or a specific return to missing subject work.

An explanation already sufficient for the use needs no additional comparison merely because another format or technique is available.

EXD.6:2 - Problem

A longer explanation may add information the recipient does not need. A shorter one may omit a decisive condition. A clearer-looking diagram may contain new content, and improvement after a hint may be credited to the original text.

Comparison must identify what changed and what outcome the evidence can support. Correctness, structural recovery, effort, retention, transfer and usefulness answer different questions.

EXD.6:3 - Forces

EXD.6:4 - Solution

EXD.6:4.1 - Fix the receiving question and its criterion

State what the explanation should let this recipient recover or do. Identify the subject criterion and an available explanation that can serve as the incumbent. A mathematical claim needs its correct derivation; a recommendation needs its actual grounds and conditions.

For a structural comparison, use FPF C.2.8: identify the account, its expression and observer, qualified by the selected relations, correctness criterion, preparation, available operations, access, assistance and budget. Choose a qualitative description of recovered and missing relations when it is sufficient. A numeric amount needs its own declared scale and units or denominator.

Distinguish recovering what an account asserts from establishing that assertion’s subject warrant. If warranted subject structure is the selected target, an unsupported relation cannot count as a correctly supplied ground.

EXD.6:4.2 - Identify the smallest material change

Use the construction that addresses the live difficulty: clarify the question, add a missing ground, build a discriminating example, coordinate an expression or guide a recipient product. Name which of content, expression, preparation, help or target changed.

When comparing forms, keep the selected claims and receiving conditions comparable. When coordinated changes are needed, compare the whole proposed repair but limit attribution accordingly. A claim inventory can help identify what both candidates express; it is not a measure of recovered structure.

FPF C.11.CRC supplies configuration-relative contribution comparison and C.11 the marginal-choice question. For Method descriptions, use C.37.CF’s actual comparison. For instructional material, use HCD.19’s representative-reader work and diagnosis. Their supplied results can answer the comparison.

EXD.6:4.3 - Obtain only the evidence that can change the choice

First inspect subject correctness, missing claims and explicit contradictions. These can justify a repair before a reader trial. Use a reading or application probe when the live alternatives concern recoverability or receiving use and its possible results can change the choice.

Preserve the initial response before assistance. Identify what the recipient had available at each point and what the subsequent help contributed. Prior knowledge can enable reconstruction; independent correction of an absent or false instruction must be identified rather than credited to the expression.

If the same person sees both explanations, prior exposure or practice can affect the second response. Design the comparison for the conclusion sought or retain that attribution limit. A small local observation may suffice for a local repair; a general effect claim needs a design and evidence appropriate to that stronger question.

If useful observation is unavailable, retain a sufficient incumbent or make a reversible design choice with its intended gain stated as an expectation. The absence of a measured human effect does not prevent a qualified explanation from being used.

EXD.6:4.4 - Choose the result and its stopping condition

Relate the demonstrated or reasonably expected contribution to the burden that matters in this use. Retain the incumbent when the candidate adds no worthwhile result. Prefer a targeted repair when it supplies a missing relation or fixes a consequential difficulty. Return unsupported subject claims to their supplier.

Keep result types distinct. More recovered structure within one budget differs from less effort to recover the same structure. Immediate recovery differs from retention after a delay or transfer to another task. Agreement and willingness to act differ from understanding a recommendation.

State what is retained, why, under which conditions and what fact would reopen the choice. A larger explanation is useful only to the extent that its contribution serves the selected purpose.

EXD.6:5 - Archetypal Grounding

Compare a complete prose account of the pooled mean with the table expression in EXD.3. Both carry five selected relations:

• count multiplied by group mean reconstructs group sum;
• group sums combine into total sum;
• group counts combine into the total number of individuals;
• total sum divided by total individual count gives the pooled mean;
• averaging group means equally answers a different weighting question.

The subject values and criteria are available: individual weighting gives 6.75; equal group weighting gives 5.5. The proposed table aligns count, mean and sum. Its intended contribution is correspondence under the recipient’s available preparation and budget, not an additional mathematical ground.

A representative response could reveal that the recipient cannot use the mean-to-sum relation. If the explainer then works that step and obtains a correct response, keep the original response and the supplied help separate. That sequence supports a targeted prerequisite or assistance repair. It does not show that the unassisted table outperformed the prose.

Alternatively, the recipient may already recover all five relations from the prose and use the required statistic. Without another worthwhile contribution, retain the prose. If the table is selected for a known reference task, such as locating counts and sums repeatedly, record that use and the design reason; do not report an unobserved reading-time improvement.

Now suppose the recipient changes the report to equal branch influence. Re-establish the target before comparing the explanations: the selected answer becomes 5.5. Failure to give 6.75 under that new question is not an extraction defect. The source account still supports both conditional answers.

These are constructed comparison branches. Their actual observations, if obtained, must be reported at their own conditions.

For advising, a person may recover the recommendation’s decisive premises and choose differently because of a preference. Keep the explanation if it served that understanding use; return the preference to the decision. Persuasion would require a different declared purpose and its own grounds.

EXD.6:6 - Bias-Annotation

Authors can prefer a new artifact because they invested effort in it, or prefer a fluent response because it resembles their own account. Use the sufficient incumbent, the subject criterion and the actual recipient conditions to constrain that judgement.

A human-learning result and an AI reconstruction probe have different evidential reach. Neither can silently stand in for the other.

EXD.6:7 - Conformance Checklist

Recognition concerns a choice whose outcome could change the retained explanation or next repair.

For assurance, inspect whether the incumbent and candidate have a stated use; their material differences are identifiable; correctness and recovery criteria are clear; preparation, help and initial responses are preserved at the needed grain; the evidence supports the claimed outcome; and the retained result is worthwhile under its stated burden and reopen condition.

A missing observation is not a zero score. A numerical ESI or effort value requires its declared measurement basis.

EXD.6:8 - Common Anti-Patterns and How to Avoid Them

EXD.6:9 - Consequences

The author can choose a finite repair and stop work that adds no worthwhile contribution. The comparison can also reveal a missing subject input or an uncertainty about recipient conditions. Generalization beyond the observed or stipulated use remains a separate question.

EXD.6:10 - Architectural Rationale

Comparison connects constructive choices with receiving value. It uses existing characterization and choice methods while making the explanation-specific account, form, recipient and help distinctions explicit. Its result can send work to any construction pattern or preserve a sufficient answer; it is not a mandatory final stage of every exchange.

EXD.6:11 - SoTA-Echoing

FPF C.2.8 supplies the common structural characteristic; C.11.CRC/C.11 supply the contribution and marginal-choice questions. C.37.CF and HCD.19 provide concrete comparison and repair practices at their subject scopes. Adopt those supplied operations and retain their distinct answer criteria.

Mengelkamp et al., 2025 distinguish comprehension and metacomprehension outcomes for informative narratives and exposition under reading-goal conditions. Their mixed results qualify any fixed genre or goal recipe. Fichtel et al., 2025 likewise distinguish more co-constructive behavior from a significant mean objective-understanding advantage.

The serious alternative is to select the most fluent, extensive or visually attractive explanation without a receiving comparison. In the worked case, the smallest account/form/help distinction can change the repair while leaving a sufficient incumbent available. Reopen when the task, subject basis, recipient, assistance, budget or relied-on evidence changes.

EXD.6:12 - Relations

EXD.1 establishes the receiving question. EXD.2, EXD.3 and EXD.5 construct candidate repairs. EXD.4 exposes changes during the exchange.

FPF C.2.8, FPF C.11.CRC and FPF C.11 retain their defining questions. The source and dependency account names the actual Core’s Method-development family and HCD receiving uses.

EXD.6:End

Use profiles

The profiles reuse the Architectural Reference’s subject, account, expression, recipient and assurance distinctions and the six Methods. Each changes the receiving use, supplied contributions and evidence question.

EXD.Profiles:1 - Instructional explanation

Use this profile when a learner needs to reconstruct and use a relation that a learning task requires. A lesson, example or demonstration is insufficient when the learner can imitate its surface but cannot make the required connection. Start from the target performance and its criterion, including the help permitted. A learner who already performs the relevant task, or someone seeking only a sufficient immediate answer, may need no additional explanatory exercise.

Human Capability Development supplies the task through HCD.6, feedback tied to the task criterion and retry through HCD.9, task-specific assistance through HCD.10, and instructional-material evaluation through HCD.19. Explanation Design constructs the relation-bearing account, example, expression and recipient contribution within that work. It neither infers a capability deficit from every failed explanation nor treats today’s assisted correction as a durable capability result.

For the pooled-mean task, suppose the learner can add and divide and initially writes “(3 + 8)/2 = 5.5, because there are two averages.” The required question is equal weight per individual. EXD.1 preserves that target; EXD.2 shows why count multiplied by mean recovers each group sum. EXD.3 can place count, mean and sum together, while retaining the same subject claims.

EXD.5 asks what the denominator should count under this target. Preserve the initial explanation. If needed, point to the eight individual observations and ask the learner to reconstruct the total sum and count. Record the corrected calculation, 54/8 = 6.75, with the help supplied. A retry with three observations of mean 4 and nine of mean 8 requires (12 + 72)/12 = 7. The product criterion is the reconstructed relation between group sums, group counts and the individual mean; a correct number alone can result from copying.

A changed target remains a changed target in instruction. If the learner now asks for equal weight per group, the corresponding mean is 6 in the retry, and EXD.4 returns to the selected question. The instructor can separately decide whether the original learning objective still requires the individual-weighting task.

Recognition concerns a particular learning obstacle and a plausible constructive move. Assurance concerns the product actually obtained at the stated help conditions. An unaided changed task can support a claim about that performance; retention needs a later opportunity, and broader transfer needs a task that tests the intended changed relation or context. Observations on one learner or a constructed example do not establish a population effect.

Guidance can reduce an irrelevant search burden while concealing what the learner can do independently. An initial contribution can reveal a missing relation while imposing unnecessary effort on an unprepared learner. Choose the opportunity and assistance by the task and preparation; compulsory initial failure, repeated self-explanation and automatic fading are poor substitutes for that decision.

Fiorella’s differentiated explaining, visualizing and enacting contributions, the 2023 relational-mapping study, Zhang and Fiorella’s error-focused feedback, and Tetzlaff and colleagues’ assistance synthesis inform EXD.2/.5 at their stated limits. Mayer’s multimedia principles and Cromley and Chen’s synthesis inform EXD.3’s conditional form choices. These sources support candidate instructional moves and relevant distinctions; they do not certify this particular lesson. The practical gain is a repair directed at the needed relation with interpretable assistance and retry evidence. EXD.6 and HCD.19 decide whether changing the material is worth its burden or whether a different task or support question owns the next action.

EXD.Profiles:2 - Technical and advisory explanation

Use this profile when a practitioner or advice recipient needs to understand a rationale well enough to select an entry, assess a condition or make an informed next move. The subject may be a technical arrangement or a recommendation. The difficulty can be an omitted relation, an ambiguous question or a changed premise; disagreement and lack of authority require their own treatment.

The common account and recognition/assurance distinction remain as stated in the Architectural Reference. Here subject qualification comes from the relevant technical or advising practice. Core’s Method-development patterns supply Method-description content and whole-language rationale through A.3.2.DE and E.8.ML and comparison through C.37.CF. Decision Design supplies the recommendation’s grounds, alternatives and reconsideration through PSDA.14. Decision Design supplies engagement, authority and feasibility conditions through PSDA.19.

An engineer asks why source allocation, coherence checking and reconstruction have separate entries and proposes one review procedure. EXD.1 distinguishes wanting one check from wanting the architecture rationale. The supplied E.8.ML account relates three missing results: F.0.2.MA asks which source contributions were carried; A.3.1.CC asks whether a maintained claim agrees with its basis; F.0.2.RC asks which promised source contribution is no longer recoverable. Their different questions and source returns justify direct entries. A diagram may group the entries for navigation without establishing that their subjects are parts of one Method.

If the engineer only needs to check one maintained claim against an available source, direct A.3.1.CC can suffice. If a required source disappears, return that affected reliance while retaining independently supported results. EXD.3 can change the expression of these same relations; EXD.6 keeps the current E.8.ML account unless a relevant comparison justifies a repair. A publication-authorization question has different grounds from understanding this rationale.

In a second, stipulated case, Mira asks why a bounded coordination assignment is recommended before immediate enrolment in a management course. The available rationale includes successful recent coordination tasks, a sponsor offering protected time for four months, and no current evidence that a credential is required. Select those decisive grounds and the live alternative rather than inventing a biography or capability diagnosis.

If Mira understands these grounds but values the course’s social setting, EXD.4 treats the contribution as relevant preference or disagreement. Decision Design work can reconsider the recommendation. If the sponsor withdraws protected time, the changed premise also returns there. A recital exercise contributes nothing to the immediate advising question. The explanation can succeed in making the rationale usable while Mira declines the recommendation.

Recognize a usable explanatory move by the recipient’s actual question and the supplied rationale. Assure subject claims against their sources and conditions, and examine the response needed for this use: for example, distinguishing which Core’s Method-development family entry answers the missing question, or recognizing how loss of protected time affects the advice. Verbal agreement does not prove understanding, and understanding does not supply consent, resources or decision authority.

The principal costs are unnecessary elicitation, an overlong justification and exposing more detail than the recipient needs. The principal bias is to interpret an unfamiliar question or disagreement as inability. Fisher’s consultation analysis and Axelsson and colleagues’ contextual feedback distinctions inform the dialogue; they are adapted to these constructed cases without claiming a universal sequence or outcome. The practical gain is a qualified rationale and correct next entry or return. A separate instructional need can select HCD and the instructional profile, but is not inferred merely from receiving advice.

EXD.Profiles:3 - Human–AI explanation

Use this profile when a person uses an AI system to prepare, discuss or revise an explanation, or seeks an explanation about an AI system. First distinguish these subjects. An AI’s role in producing text does not make its own operation the thing explained. A user asking why a deployed system produced an output may require evidence about that actual system, its inputs and operation.

The Architectural Reference’s account, expression, recipient and use conditions still govern the work. The human can contribute questions, corrections, constraints and preferences. AI-produced content and claimed causal grounds require qualification for the relevant subject. Fluent output, an expressed confidence and a system’s own story about its reasoning cannot establish an otherwise missing causal warrant.

In the pooled-mean exchange, an AI assistant can state the finite-list definition, recover the sums and counts, and answer 6.75 for equal individual weight. If the user changes to equal branch influence, EXD.1/.4 select the corresponding question, answer 5.5 with its label and retain the scope of the first answer. If the user asks only for the established calculation, further elicitation may add no value.

If the user instead asks, “Why did our deployed predictor choose this value?”, the arithmetic case supplies no causal explanation of that model. Recover the relevant model, input and available diagnostic or subject evidence. If that basis is unavailable, name what is missing and limit the answer accordingly. A more polished narrative or the model’s unsupported self-report cannot repair that lack.

For instructional use, the assistant can ask for a focused explanation and preserve the initial response, hint and retry as EXD.5 specifies. The resulting exchange shows what the person produced with that assistance. An AI answering the learner task itself can help inspect a public instruction or construct a candidate response; it is not an observation of the human learner. The profile likewise makes no inference that an AI recipient has human attention, memory or learning mechanisms.

Recognize a promising exchange when the question and subject are selected, available grounds constrain the content, and contributions can change the next move. Assure the actual claims and response at the relevant system, version, access and help conditions. When testing an explanatory interaction, distinguish conversational behavior, immediate objective task performance, confidence, later retention and decision quality.

Current social XAI work supplies joint question formation, incremental explanation and adaptation. Liao and colleagues and Schmude and colleagues inform elicitation anchored in a user’s task, without treating a fixed question bank as complete. Fichtel and colleagues’ 2025 study is a consequential limit: more co-constructive behavior under its prompting conditions did not yield a significant mean objective-understanding advantage. Neither the existence of an interactive system nor a more collaborative style settles the receiving outcome.

Adaptation can make the exchange more relevant while adding turns, reading burden and opportunities for unsupported content. Treating a human user as a fixed knowledge level can miss a changed purpose; treating every correction as a prompt to add detail can retain the wrong question. The useful result is a warranted answer, task-specific repair or honest subject return with interpretable interaction evidence. EXD.6 governs worthwhile local comparison; the instructional and technical/advisory profiles supply their respective stronger receiving questions.

EXD.Profiles:End

Explanation Design Principles Framework Sources and Edition Notes

EXD.Sources:1 - Relied-on contributions and their limits

This edition connects formal and pragmatic explanation, empirical work on learning and interaction, and constructive practice through their contributions to a recipient’s task. The sources below support the stated distinctions and Methods. Worked examples in this framework are constructions unless explicitly reported as source observations.

Older books by Brown and Wragg (2001), LeFever (2012), Khalifa (2017) and McCain (2022) identify lines of instructional practice, explanatory craft and philosophical inquiry in the field’s background. Their publisher descriptions and contents supplied lineage orientation. Unread chapters are not premises for the operative Methods above; the later sources supply the named receiving contributions and limits.

The Architectural Reference explains the connected language and the consequences of these sources for its use.

EXD.Sources:End

Name and citation

EXD.Edition:1 - Citation

Anatoly Levenchuk. Explanation Design Principles Framework. Use the release date at the beginning of this file when citing this edition. Cite a particular Method by its EXD.1–EXD.6 address and title, together with this edition. The framework’s FPF publication repository is the public discovery route.

EXD.Edition:2 - Naming settlement

This local naming settlement concerns this six-pattern Domain Principle Framework and its three use profiles. Its subject rule is FPF E.4.DPF; its naming methods are FPF F.18 and FPF F.17.

NameCard EXD-NAME-01. The governed value is the Explanation Design DPF specified by this edition’s Architectural Reference, six EXD Methods and Use profiles. Its kind is Domain Principle Framework as constrained by E.4.DPF. Its Tech designation is Explanation Design Principles Framework; its Plain designation is Explanation Design. EXD is the local pattern-address prefix.

The effective local reference scheme fixes the governed practice as constructing and adapting explanations for a recipient’s use. It distinguishes subject, question, purpose, warranted account, expression, recipient, contribution and use conditions; supplying subject Methods and their outcomes remain external. The local sense of “Explanation Design” is this DPF for that practice, including performed interaction. The expression does not name every theory of explanation or a new FPF kind.

“Explanation Practice” was considered because it includes performance, but underexpresses the constructive repertoire. “Explanation and Understanding” was considered because it names both sides of the relation, but suggests broader coverage than this governed practice. The selected name is made precise by the opening scope and the direct dialogue entry. Existing FPF and supplying-DPF designations suffice for their own values; no parallel name family is introduced. This is an initial naming settlement, with no predecessor rename or cross-framework equivalence asserted.

Reopen the name when its actual use systematically hides performed interaction, when the governed practice changes, or when an actual naming collision prevents unambiguous citation. A new source citation alone does not change this value’s identity.

EXD.Edition:3 - Public term row

The FPF F.17 public row below makes the single local naming settlement available for reuse. Public citation of the framework and its prefix needs this one stable designation; the six ordinary action titles do not require six additional term families. This is the bounded F.14 admission reason. The row’s claim content consists of the identity, sense, basis, use and currentness claims stated here; the row and the naming settlement are distinct epistemes about the same governed framework.

No F.9 Bridge occurrence is asserted: the row publishes the same local naming claim and does not establish a cross-scheme equivalence. Publication and substantive fitness are answered by their own evidence; the row makes neither claim merely by being filled.

EXD.Edition:End