Guided Walkthrough¶
Guided procedure — instantiates Temporary Scaffold and Fade
Walks the performer through the real task live the first time, then hands off more of the decisions each pass until they run it unsupported.
A guided walkthrough takes the performer through the real task, live, the first time — a guide beside them narrating, directing, and catching errors as the actual work happens — and it instantiates scaffold-and-fade by handing off more of the decisions each pass until the performer executes unsupported. What makes it this mechanism and not worked-example fading is that it is co-execution of the genuine task in real time, not study of a finished exemplar: the work is really being done, and the scaffold is the guide's live involvement, withdrawn by transferring decisions rather than by thinning a shown solution.
Example¶
A new bank teller has to process their first complex transactions — a large cash deposit that triggers a hold decision, then a wire. On the first one, the mentor stands beside them and effectively co-drives: "count it twice; now this amount triggers a hold — here is how you decide the hold period, you do that part." The task is real, the customer is real, and the mentor is handing the teller one decision while steering the rest. On the next transaction the mentor hands off more — the teller decides the hold and enters it, the mentor only confirms. By the fifth, the mentor narrates nothing and intervenes only if something goes wrong. The handoff is not on a timer; the mentor releases each decision when the teller demonstrably takes it. The close is an independence check: the teller processes a comparable transaction with the mentor watching silently but present, ready to step in — and if a decision falters, the mentor takes back just that decision, not the whole task.
How it works¶
- Do the real task together, live — not a simulation of it or a study of a finished version — so the performer meets the genuine cues and consequences.
- Enumerate the decisions the task contains and hand them off one at a time, the guide retaining the rest.
- Release each decision contingently — when the performer demonstrably takes it — rather than on a schedule, tracking which decisions have moved.
- Close with an unsupported run under observation (an independence check) and, on failure, reclaim only the specific decision that faltered.
Tuning parameters¶
- Handoff order — which decisions transfer first (usually low-consequence, reversible ones before high-stakes ones). The wrong order either overwhelms the performer or bores them.
- Release evidence — how much demonstrated competence is required before a decision is handed over. Demanding evidence prevents premature release; lax evidence hands off too soon.
- Guide presence at close — silent-but-present versus fully absent for the independence run. Presence contains a real customer's risk; absence gives a truer read of independence.
- Reclaim granularity — on failure, take back the single decision or the whole task. Fine reclaim keeps momentum and dignity; coarse reclaim is safer when the stakes are high.
When it helps, and when it misleads¶
Its strength is that it lets a beginner do consequential, real work safely on day one, and because the task is genuine the transfer problem is smaller than with an artificial exemplar. It rests on contingent tutoring[n1] — the guide gives more help after a failure and less after a success. The failure mode is the guide who never actually lets go: the walkthrough feels productive because the work gets done, but the performer is riding the guide's decisions and collapses when alone — hidden dependency dressed as training. The classic misuse is calling a permanent "always shadow a senior" arrangement a walkthrough when no decisions are ever released. The guard is to make the decision handoff explicit and evidence-based, and to insist on a genuinely unsupported run before calling the task owned.
How it implements the components¶
task_barrier— targets the whole-task, unfamiliar-in-real-conditions barrier: the performer cannot yet carry the live task end to end.responsibility_transfer_marker— the one-decision-at-a-time handoff is the explicit, trackable transfer of responsibility from guide to performer.independence_check— the observed unsupported run tests whether the performer carries the real task alone.
Its nearest twin is Worked Example Fading; the walkthrough is live co-execution of the real task, not study of a modeled solution, so it does not thin a shown solution's steps (support_requirement, fading_schedule) and its handoff is contingent on the performer taking decisions rather than a preset fade. It does not build varied repetition for transfer (transfer_verification); that is Faded Practice Drill.
Related¶
- Instantiates: Temporary Scaffold and Fade — a walkthrough is the live-co-execution form of the archetype's support.
- Sibling mechanisms: Prompt or Hint · Checklist or Job Aid · Template or Partial Solution · Worked Example Fading · Coaching Cue · Embedded Tooltip or Contextual Help · Faded Practice Drill
Editorial Notes¶
Form Classification¶
Form family: Communication, Facilitation & Learning
Rationale: Guided Walkthrough operates as a designed message, facilitated interaction, ritual, or learning activity that changes shared understanding because it walks the performer through the real task live the first time, then hands off more of the decisions each pass until they run it unsupported.
Independent corroboration: The frozen evidence defines Guided Walkthrough as 'Walks the performer through the real task live the first time, then hands off more of the decisions each pass until they run it unsupported', so its operative form is Communication, Facilitation & Learning.
Nearest alternative: Protocol, Workflow & Routine — The walkthrough builds capability through live guided performance and fading support; decision handoffs organize the learning interaction.
Review outcome: Independent reviewer agreement; medium confidence.
Origin Attribution¶
Primary origin: Education & Pedagogy
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Contingent tutoring and scaffolding directly support live performance followed by progressive handoff.
Related originating lineages:
- Psychology — Learning research materially establishes contingency on success and failure.
Review resolution: Both reviewers agree that education_pedagogy is primary: Contingent tutoring and scaffolding directly support live performance followed by progressive handoff. I retain psychology only as formative lineage, not as a list of later applications. I resolve origin_mode as cross_disciplinary_synthesis because the artifact joins distinct disciplinary contributions. I resolve domain_reach as multi_domain because it transfers across several fields but is not a domain-free primitive. Encyclopedia synthesis is false because the exact generalized packaging is already established enough that encyclopedia-specific synthesis is not required.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
[n1] Contingent tutoring / the contingency rule (Wood, Bruner & Ross, and later Wood): effective one-to-one help is contingent — the tutor increases control after the learner fails and cedes control after the learner succeeds — which is the fine-grained version of handing off decisions as competence appears. ↩