Feedback audit

487. Lesson identity

Module

1.12 — Player feedback

Lesson

Feedback audit

Academic type

Systems

Schema type

Mixed

Order

2 of the module

Estimated time

20–35 minutes, including the practical audit

This lesson closes the Stage 1 perception work by applying an evidence-based audit to player-facing confirmation. The audit uses ACT → RESPOND → CHANGE → AGAIN.

488. Learning objective

After this lesson, you can complete a feedback audit that identifies one missing confirmation and verifies one sufficient confirmation, with each judgment supported by observable evidence.

489. Why this matters

A game can accept an input and change its internal state without communicating the full result to the player. A feedback audit separates three questions that are easy to confuse:

1. Did the player act?
2. Was there evidence that the input was accepted?
3. Could the player perceive and attribute the resulting consequence?

A click or button animation may answer the second question without answering the third. The audit prevents you from treating an internal change—or a local response alone—as proof that the complete action-result relationship is clear.

490. Prior knowledge

You should already be able to:

• Distinguish a player action from the game change that follows it.
• Explain why an important action needs player-facing confirmation.
• Apply the previous lesson’s principle: if the player did not receive the confirmation, the action did not happen from the player’s perspective.

491. Core concept

A feedback audit is an evidence-based check of whether a player can perceive the relationship between an action and its resulting change.

Audit an interaction in four passes:

1. ACT: What exactly does the player do?
2. RESPOND: What immediate evidence shows that the input was received or accepted?
3. CHANGE: What resulting state or situation can the player perceive?
4. AGAIN: After restoring the relevant starting conditions, does another observation show the same legible relationship?

A confirmation is missing when the observed evidence does not allow the player to perceive and attribute the relevant result. A confirmation is sufficient when the evidence allows the player to connect the action to its consequence without guessing. Sufficient confirmation may use more than one cue, such as a local switch movement plus a visible door opening.

This lesson uses only those two final classifications: missing and sufficient.

492. Evidence model

Record what was observed rather than what the designer intended.

Do not infer a visible result from hidden implementation state. Do not call a confirmation sufficient merely because the input produced a click. Input acceptance and consequence confirmation are related but distinct evidence.

493. Worked comparison

Consider two recorded observations of a switch-and-door interaction.

494. Observation A — missing confirmation

• ACT: The player presses the interaction control while facing the switch.
• RESPOND: The switch does not move, and there is no sound, text, or other local cue.
• CHANGE: A door outside the player’s view opens. The observation provides no linked indicator or subsequent view of that door.
• AGAIN: The interaction is reset. A second trial produces the same lack of perceptible local or remote evidence.

Audit result: Missing confirmation. A change may have occurred, but the player receives no observable evidence that connects the action to the door.

495. Observation B — sufficient confirmation

• ACT: The player presses the interaction control while facing the switch.
• RESPOND: The switch visibly moves and produces a distinct click, showing that the input was accepted.
• CHANGE: The door is visible in the same view and opens immediately as the switch moves. A line indicator between the switch and door also changes from dim to bright.
• AGAIN: The interaction is reset to the original switch and door states. A second trial reproduces the switch movement, click, linked indicator, and visible door opening.

Audit result: Sufficient confirmation. The local cues confirm input acceptance, while the visible door movement and linked indicator make the consequence perceptible and attributable to the action.

“Sufficient” does not mean large or dramatic. It means the available evidence supports the correct action-result explanation without guesswork.

496. Common mistakes

497. Mistake 1: treating internal change as player-facing evidence

“The door-open variable changed” is implementation evidence, not necessarily evidence available to the player.

498. Mistake 2: confusing input acceptance with consequence confirmation

A switch click can show that an input was accepted. It does not, by itself, show that a distant door opened.

499. Mistake 3: calling persistence a repeated trial

Pressing an already activated switch does not reproduce the original activation. For AGAIN, reset a stateful interaction before the second trial. If no reset occurs, describe the observation only as a persistence check.

500. Mistake 4: changing a classification to complete the worksheet

Classify the evidence impartially. Never relabel an interaction merely to obtain one example of each category.

501. Guided practice: audit supplied evidence

The following records are fixed interaction artifacts. Treat their listed observations as the available evidence. Do not add imagined cues or hidden implementation details.

502. Artifact 1

Trial 1

• A player faces a floor button and presses the interaction control.
• The button does not move or produce sound or text.
• A bridge extends behind an opaque wall and remains outside the player’s view.
• No indicator links the button to the bridge.

Trial 2 after reset

• The same action again produces no local cue.
• The bridge again remains outside the player’s view, with no linked indicator or later observation supplied.

503. Artifact 2

Trial 1

• A player faces a floor button and presses the interaction control.
• The button visibly depresses and emits a short tone.
• A bridge in front of the player extends immediately.
• A bright line from the button to the bridge activates during the extension.

Trial 2 after reset

• The button and bridge begin in their original states.
• The same action reproduces the button movement, tone, bright line, and visible bridge extension.

For each artifact:

1. Write a precise ACT statement.
2. List only observed evidence under RESPOND.
3. Record the perceptible result—or the absence of a perceptible result—under CHANGE.
4. Use the reset second trial as AGAIN evidence.
5. Classify the confirmation as missing or sufficient.
6. Explain why the evidence supports the classification.

Use this worksheet:

Artifact: ______________________________
ACT: __________________________________
RESPOND: ______________________________
CHANGE: _______________________________
AGAIN: _________________________________
Classification: missing / sufficient
Evidence-based reason: _________________

You may instead audit directly observed interactions from a prototype or recorded gameplay. If you use your own source, observe at least two actions and preserve notes, screenshots, timestamps, or a short recording. Select one evidence-supported missing example and one evidence-supported sufficient example only if your observations genuinely contain both. Otherwise, use the supplied artifacts for the absent category. A hypothetical design may be discussed as planning, but it is not evidence from a completed audit.

504. Practical assessment

Complete Practical assessment: Two-record feedback audit. Submit two audit records: one supported example of missing confirmation and one supported example of sufficient confirmation. Each record must include specific ACT evidence, perceptible RESPOND evidence, observable CHANGE evidence, a reset repeat observation, and a defensible classification.

505. Validation

A complete audit allows another reader to answer:

• What did the player do?
• What evidence showed whether the input was accepted?
• What consequence could the player perceive?
• Could the player attribute that consequence to the action?
• Was the interaction reset before the repeated trial?
• Why does the evidence support the classification?

A passing submission distinguishes input acceptance from confirmation of consequence and does not rely on imagined or hidden evidence.

506. Key takeaways

• Audit from the player’s observable point of view.
• Separate evidence of input acceptance from evidence of the resulting consequence.
• Use ACT → RESPOND → CHANGE → AGAIN to record the relationship.
• Reset stateful interactions before treating a second observation as a repeat trial.
• Classify evidence impartially before proposing a revision.

507. Next lesson

Continue to The four parts on one artifact. You will reuse an audited action to label input, rule, presentation, and feedback, identifying where the feedback leg sits in the intended cycle.

508. Knowledge check