Use profiles

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

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.

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 Preface. Here subject qualification comes from the relevant technical or advising practice. Method Engineering supplies Method-description content and whole-language rationale through ME.8/.23 and comparison through ME.22. Problem Structuring and Decision Support supplies the recommendation’s grounds, alternatives and reconsideration through PSD.13. Development Opportunity Construction and Development Direction Advising supplies engagement, authority and feasibility conditions through DOCA.7.

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 ME.23 account relates three missing results: ME.21 asks which source contributions were carried; ME.12 asks whether a maintained claim agrees with its basis; ME.24 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 ME.12 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 ME.23 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. PSD/DOCA 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 ME 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.

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 Preface’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.