Ethical Gamification & Professional Motivation

When more activity can be a worse motivational outcome

A security course adds a daily streak, points and a public leaderboard. Completion rises for two weeks, but optional reflection quality falls and several weaker learners stop participating in the public ranking. A product dashboard could still call the change a success because sessions and completions increased.

That conclusion is too strong. Participation is not mastery, and visible activity is not proof of high-quality motivation. This lesson builds Ethical Motivation & Gamification Governance Contract v1: a bounded record connecting each mechanic to a learning goal, intended behavior, reward type, participation signal, mastery evidence, autonomy/competence/relatedness signals, coercion and equity risks, opt-out state, sunset condition, owner and provenance.

The strongest positive state is MOTIVATION_GOVERNANCE_REVIEW_REQUIRED. A mechanic can reach that state only when its learning purpose and evidence boundary are visible and its risks are reviewable. Missing or contradictory evidence remains a typed HOLD rather than being converted into a success because engagement went up.

Route orientation and ownership boundary

MKN-013 owns the motivation/gamification governance contract. MKN-001 is the frozen hard prerequisite. MKN-012 owns instructional visual/slide design and MKN-014 owns pacing/fatigue signals; neither outcome is absorbed here.

Foundation learners should trace one mechanic from learning goal to evidence and risk review. Repair learners should identify the first missing or unsafe field without redesigning the whole product. Experienced learners should defend why points, streaks and rankings are not inherently ethical or unethical: the decision depends on purpose, visibility, choice architecture, evidence quality, social context and what happens when the mechanic stops.

Checkpoints: G1 separate participation from mastery; G2 identify an autonomy/coercion or equity failure; G3 recompute the same mechanic when the course moves into an employer-sponsored cohort where completion is visible to managers.

Mental model: learning purpose → mechanic → motivational-risk check → evidence split → keep/modify/retire

Use this flow:

learning goal → proposed mechanic → intended behavior → autonomy/competence/relatedness risk check → participation signal ↔ mastery evidence → opt-out/equity review → keep/modify/retire → durable-outcome check

The key design move is to keep participation evidence and mastery evidence in different fields. A streak can increase daily return rate. That does not prove the learner can perform the target skill. A leaderboard can create social energy. That does not prove the energy is autonomy-supportive, equitable, or durable after the reward disappears.

The contract is therefore a governance object rather than a reward catalog. It can say HOLD, MODIFY, RETIRE or REVIEW_REQUIRED even when a mechanic produces a statistically impressive activity increase.

Foundation contract: sixteen frozen evidence dimensions

The frozen dimensions are motivation_case_id, learning_goal_ref, mechanic_id, intended_behavior, reward_type, participation_metric, mastery_evidence_ref, autonomy_risk, competence_signal, relatedness_signal, coercion_flag, equity_risk, opt_out_state, sunset_condition, review_owner, and provenance.

Minimum invariants:

1. Every mechanic has a declared learning goal and intended behavior.
2. Reward type is explicit; hidden pressure is not disguised as neutral UX.
3. Participation and mastery are stored separately.
4. Autonomy, competence and relatedness are review lenses, not automatic scores.
5. A coercion flag cannot be ignored because engagement increased.
6. Equity risk must be visible for social, competitive or access-sensitive mechanics.
7. Opt-out state must be explicit where the mechanic can create social or institutional pressure.
8. A sunset condition states when the mechanic should be retired, modified or re-reviewed.
9. A named review owner and provenance are required.
10. Changed audience or visibility requires recomputation instead of copying the baseline decision.

Deep explanation: motivational support is about quality, not only quantity

Self-determination theory — FACT. SDT distinguishes qualities of motivation and treats autonomy, competence and relatedness as basic psychological needs relevant to internalization, engagement and functioning. The theory is a lens for asking whether an environment supports volition, effectiveness and connection; it does not prescribe MAKNOW point values or ranking rules.

Gamification review — FACT. The frozen peer-reviewed gamification review describes purposeful use of game-design elements to support engagement/motivation and notes that poorly designed gamification can undermine basic psychological needs or create harmful competitive effects. It also describes mixed educational results rather than a universal benefit.

Participation-versus-mastery split — INFERENCE. A learning platform should audit streaks, points or completion separately from demonstrated competence. High participation can coexist with flat mastery or externally controlled behavior. This is a system-design inference informed by the motivational boundary, not a universal empirical law.

MAKNOW governance contract — DESIGN. Learning-purpose mapping, coercion/equity states, opt-out requirements and sunset rules are local design authority. They must be versioned and reviewable rather than attributed to the external theory.

Worked cases: public ranking, private progress and changed institutional context

Baseline. A professional security course adds private streak feedback, points and a public leaderboard. Completion rises 18%, but reflection quality is flat and several low-ranked learners stop using the leaderboard. The contract records participation_metric=UP, mastery_evidence_ref=FLAT, autonomy_risk=MEDIUM, equity_risk=HIGH_FOR_PUBLIC_RANKING, opt_out_state=AVAILABLE, and a sunset review after the next retention checkpoint. The public ranking is MODIFY_AND_REVIEW_REQUIRED; private progress can remain under review.

Wrong approach. Replacing private progress with public ranking without recomputing autonomy, relatedness, coercion and equity is rejected. The mechanic is not equivalent just because it uses the same underlying score.

Transfer. The same course moves into an employer-sponsored cohort and managers can see completion. A streak that was optional in the consumer context now carries institutional pressure. The baseline decision is not an oracle: visibility, opt-out credibility, coercion and equity must be recomputed.

Failure modes that must fail closed

Fail closed when:

• participation increases while mastery evidence is flat or missing;
• a public ranking replaces private progress without risk recomputation;
• a mechanic becomes manager-visible or employment-adjacent but the consumer decision is reused;
• opt-out is missing, fake or costly;
• coercion_flag=true is ignored;
• a high autonomy/equity risk has no mitigation or review state;
• there is no sunset condition, owner or provenance;
• a mechanic is kept forever because it once increased clicks;
• the mechanic is removed but the team never checks whether learning behavior persists without the reward.

A typed HOLD is a valid result. The system is allowed to say that it lacks the evidence required to promote a mechanic.

Alternatives and trade-offs: private progress, cooperative goals, public competition and no mechanic

Four common alternatives illustrate the trade space. Private progress can support competence feedback with less social comparison, but still risks becoming controlling if tied to punishment. Cooperative goals can support relatedness, but can burden learners unevenly. Public competition can energize some learners while creating autonomy, relatedness and equity risks for others. No gamification mechanic removes reward pressure but may also remove useful structure or feedback.

The right question is not “Which mechanic gets the most engagement?” It is: “Which mechanism supports the declared learning purpose with evidence no stronger than the observed outcome, while preserving meaningful choice and a credible exit path?”

Sunset conditions matter because a mechanic can be useful during onboarding and harmful after learners have internalized the professional routine. Retirement is therefore a first-class success path, not a product failure.

Experienced bridge: treat gamification as a governed intervention, not a permanent feature

For an experienced product or learning engineer, model each mechanic as a versioned intervention:

• mechanic_id and visibility are explicit;
• target behavior and learning goal are separate;
• reward exposure is logged without being treated as competence;
• participation and mastery streams remain distinct;
• motivational-risk notes are visible rather than compressed into one “engagement score”;
• opt-out events are first-class evidence;
• owner and review cadence are explicit;
• sunset/removal produces a follow-up check for durable behavior.

This allows the team to test whether a mechanic merely rents behavior with external rewards or supports a routine that persists after the reward is reduced or removed. It also makes employer-sponsored or high-power contexts visible instead of treating them as cosmetic deployment changes.

Guided and independent practice

Guided: Run lab://MKN-013/MKN-013-LAB1 on the baseline fixture. Explain why rising completion plus flat mastery does not authorize promotion of public ranking. Name the exact fields that force MODIFY_AND_REVIEW_REQUIRED.

Repair: Run one single-fault mutation. Fix only the failed dimension—opt-out, sunset, owner, provenance, coercion or evidence split—and rerun the gate.

Independent: Solve the employer-sponsored transfer fixture without seeing the baseline label. Recompute visibility, autonomy, coercion, equity and opt-out credibility from the changed context. The artifact must record baseline_oracle_reuse=false.

Verification evidence required for acceptance

A reviewer must be able to reproduce every decisive state from visible synthetic evidence. Verification checks:

• all sixteen frozen dimensions exist;
• participation and mastery are not collapsed;
• public visibility and institutional visibility trigger risk recomputation;
• opt-out is explicit and not penalized;
• coercion/equity risks cannot be silently overridden by participation gains;
• sunset and owner are present;
• provenance identifies the evidence and rule version;
• transfer is independently recomputed;
• baseline oracle reuse is false;
• no live learner/reward system is mutated.

The package passes only when its decision is reproducible without hidden model reasoning.

Assessment and transfer: prove the rule survives a change of power

S1 — Retrieval: Explain autonomy, competence and relatedness as review lenses and state why they do not prescribe a leaderboard policy.

S2 — Application: Build Ethical Motivation & Gamification Governance Contract v1 for the baseline course. Separate participation from mastery and select keep/modify/retire states.

S3 — Diagnosis: Replace private progress with public ranking. Identify the invalidated evidence fields and the safe non-success state.

S4 — Transfer: Move the course to an employer-sponsored cohort where managers can see completion. Recompute coercion, equity, opt-out credibility and sunset conditions without copying the consumer-product decision.

PASS requires a bounded, evidence-traceable decision. “Engagement increased” is never sufficient by itself.

References, freshness and authority boundaries

The frozen academic source, Gamification in Action: Theoretical and Practical Considerations for Medical Educators, remains available on the Self-Determination Theory site and frames gamification through autonomy, competence and relatedness while warning that poorly applied mechanics can undermine motivation. Current SDT theory pages likewise continue to identify autonomy, competence and relatedness as central psychological needs relevant to high-quality motivation.

Authority boundaries:

• FACT: SDT’s motivational distinctions and the peer-reviewed review’s description of opportunities and risks.
• INFERENCE: MAKNOW’s separation of participation signals from mastery evidence.
• DESIGN: mechanic allowlists, opt-out requirements, coercion/equity vocabulary, thresholds, labels and sunset rules.

The source was rechecked on 2026-09-13. Re-review by 2027-01-15 or earlier if the cited source or MAKNOW design authority materially changes.