| name | educational-and-childrens-game-experience |
| description | Use when a game for children or learners needs age-appropriate comprehension, scaffolding, learning-feedback separation, privacy/safety, stopping cues, guardian/teacher needs, accessible play, or non-exploitative engagement. Use onboarding-and-first-run-design for ordinary apps. |
| metadata | {"portable":true,"category":"15-game-visual-experience","compatible_with":["claude-code","codex"]} |
Educational and Children’s Game Experience Design
Design voluntary, age-appropriate play that supports a stated learning decision while keeping learning, enjoyment, safety, privacy and commercial evidence separate.
Use When
- Children are an intended or likely audience, or a game claims a learning outcome.
- Comprehension, scaffolding, tutorials, feedback, stopping, guardian/teacher support, safety, privacy or purchases/ads must adapt by age and context.
- Educational content must remain playable without turning the game into a quiz wrapper.
Do Not Use When
- The product is an ordinary adult app onboarding flow; use
onboarding-and-first-run-design.
- The work is curriculum, pedagogy, clinical development or legal approval; require the accountable expert and use this skill only for game experience.
- The work is general HUD, art or game-feel craft without a child/learning concern; use the corresponding game specialist.
- Engagement telemetry is being used to claim learning; route to research design and require a learning measure.
Required Inputs
| Artefact or context | Source | Required? | Why |
|---|
| Age range, contexts, languages, abilities and guardian/teacher role | Product/research | yes | Prevents an invented “child user” |
| Learning objective, prior knowledge, construct, limits and accountable expert | Education owner | required for learning claim | Defines what can be taught or assessed |
| Core loop, content, failure, stopping, social, privacy, ads and purchase model | Product/SRS | yes | Exposes safety and coercion risks |
| Consent-approved research plan and representative evidence | Research/ethics | required for observed claims | Protects children and limits inference |
Workflow
- Define cohorts by age/developmental context, language, ability, prior knowledge, play setting and guardian/teacher involvement. Stop if a single generic child cohort would change design or safety.
- State the learning objective as observable performance, the evidence method, comparator/baseline, duration and transfer limit. Separate it from enjoyment, completion and retention.
- Bind learning to the core verb: model, guided practice, feedback, variation, independent use and reflection. Keep instructions close to action and fade scaffolds only after demonstrated comprehension.
- Design failure as recoverable information. Separate gameplay consequence, learning feedback, praise and assessment; avoid shame, public ranking, punitive grind and answer-pattern guessing.
- Specify age-appropriate text, voice, icon, controller/touch, captions, sensory settings, reading support, pause/save/resume, stopping cues, guardian/teacher summaries and safe social boundaries.
- Minimise data and social exposure. Treat ads, purchases, targeting, profiling and third-party SDKs as blocked until accountable privacy/legal/store review. Any permitted ad must be clearly identified, non-intrusive, age-appropriate, non-targeted as required, easy to close/decline and unable to interrupt play, learning, recovery or stopping.
- Test comprehension, independent play, learning, delight, frustration, accessibility, stopping and guardian/teacher usefulness as separate questions under approved child-research safeguards. Record limitations and adverse signals.
Decision Rules
| Condition | Action | Wrong-choice failure |
|---|
| Child succeeds only with an adult or persistent prompt | Keep/scaffold the support and test the intended independence level | Completion is misreported as comprehension |
| Engagement rises but learning, trust, stopping or wellbeing worsens | Reject or redesign the mechanic | Retention is treated as a child outcome |
| Error identifies misconception | Give actionable task-linked feedback and a safe retry | Praise, shame or generic correction teaches nothing |
| Ad, purchase, social or SDK child treatment is unresolved | Disable the surface and escalate to accountable review | Revenue or data collection outruns child safety |
| Cultural or educational authority is missing | Mark content provisional and block approval claims | A design team impersonates an expert or community |
Capability Contract
Read access to age, learning, safety, privacy, culture and evidence artefacts is required. Planning and review are read-only by default. Do not recruit/contact children, collect data, run studies, publish, purchase, enable SDKs or claim educational/legal/cultural approval without explicit authority and accountable expert oversight.
Degraded Mode
Without expert-defined learning objectives or child-safe research, provide only a provisional experience and test plan. Without privacy/store/legal or guardian decisions, disable ads, purchases, social and third-party data paths. Without representative-child evidence, mark comprehension, learning, delight, wellbeing and stopping not assessed.
Anti-Patterns
- Quiz pasted onto unrelated play. Fix: make the learning construct part of the core verb and feedback loop.
- Completion equated with learning. Fix: specify the construct, measure, transfer limit and separate evidence.
- “Kid-friendly” as an age specification. Fix: define cohorts, prior knowledge, language, abilities and context.
- Endless rewards, streak loss or nagging. Fix: use voluntary mastery, clear session endings, pause/save and neutral return.
- Public shame, punitive failure or pay-to-win rescue. Fix: private actionable feedback, safe retry and fair non-paying paths.
- Child ads/SDKs enabled by default. Fix: fail closed until privacy, store, legal, content and device evidence passes.
Outputs
| Artefact | Consumer | Evidence and acceptance condition |
|---|
| Age/cohort and learning-experience contract | Product, education and design | Cohorts, verb, scaffolds, feedback, evidence and limits are explicit |
| Child safety, privacy, stopping and commercial gate | Accountable owners and release | Data, social, ads, purchases, safeguards and disabled defaults are decided |
| Child/learning study and evidence matrix | Research, SRS and QA | Comprehension, learning, enjoyment, access, wellbeing and stopping remain separate |
Examples
examples/river-cue-learning-loop.md — a worked learning loop with scaffolding, safe failure, stopping, guardian evidence and an ad hold.
References
references/child-learning-experience-contract.md — cohort, learning, scaffolding, safety, stopping, research and ad schemas.../../05-ux-process-research-and-psychology/ux-research-and-usability-testing/SKILL.md — evidence and consent discipline.../../00-cross-cutting-ops-qa-a11y/design-ethics-and-anti-dark-patterns/SKILL.md and ../../00-cross-cutting-ops-qa-a11y/inclusive-and-assistive-design/SKILL.md — ethics and inclusion owners.../../../doctrine/design-doctrine.md and ../../../doctrine/references/wcag-2.2-criteria.md — shared quality/access floor.
