Stars and mastery signals: making progress legible without lying

Evidence grade: STRONG on the feedback principles, design-craft on the signal systems. How evaluative signals affect learners is well-researched (the feedback page holds the fuller treatment); this page applies it to the specific artifacts learning games ship — stars, ticks, gold states, progress bars — where the design question is what each signal claims and whether the claim is true.

Every learning product emits signals — stars, checkmarks, percentages, glowing cards — and each is, functionally, a claim about the learner: you can do this; this is finished; this needs work. The feedback literature's central warning applies directly: signals that direct attention to the self (am I good?) underperform and sometimes harm, while signals that direct attention to the work and the next step help. Butler's classic result — comments improved performance, grades didn't, and grades-plus-comments performed like grades alone because the grade ate the comment — is the design constraint in one sentence: the evaluative signal dominates whatever accompanies it. Design the dominant thing carefully.

The properties of honest signals

Criterion-referenced, never normative. A star that means "you met the standard" informs; a rank that means "you beat Sam" ego-involves — same pixel budget, opposite psychology.
Checkable. A child should be able to say what a signal certifies ("gold means I remembered it after the break") and the system should be unable to emit it otherwise — signals awarded for attendance or persistence-of-clicking are decorative badges, and children audit them faster than adults do; the first unearned star devalues every earned one.
Honest in both directions. Mastery signals must be loseable where the claim stops being true — a skill that fades past its review and stays gold is a signal lying to the child and the teacher at once. The cost (children dislike losing gold) is the feature: the signal means something because it tracks reality. The softener is fictional framing (spells fade; recharging them is normal) — normalize the mechanic, never fake the state.
Terminal. A completed signal should complete: finished work that keeps soliciting ("just one more star!") converts achievement into a retention treadmill. Done must be allowed to feel done.

The teacher-facing half

Mastery signals do double duty: the same stars a child sees are assessment evidence a teacher reads, which imposes a second honesty constraint — signals must mean the same thing everywhere. A dashboard where gold means "demonstrated and retained" supports placement and grouping decisions; one where gold means "watched the video" launders usage into achievement, the digital-native mistake rebuilt as UI.

What the evidence doesn't say

It doesn't demand grade-free everything — Butler's result warns about evaluative dominance, and a mastery system is evaluative by design; the fix is criterion-reference and next-step orientation, not signal abolition.
It doesn't settle progress-bar psychology at primary age — goal-gradient effects are real in adult research; with children we apply them gently and watch.
It doesn't make signals teaching — the signal certifies learning that practice engineering produced; polishing signals can't rescue hollow reps.

In the classroom

Translate signals into criterion language when discussing progress: "gold means it stuck" beats "you got gold."
Never rank the signals publicly — a class wall of star counts is a leaderboard in craft paper.
Trust loseable signals more — an app whose mastery states never regress is describing its optimism, not your students.

How Wiz Kids applies this

Stars = demonstrated competence (with par grading efficiency); gold = retention proven across the spaced schedule, and it fades when reviews lapse because the claim lapses; completed lessons stop asking; nothing is normative anywhere. The teacher dashboard reads the same signals with the same meanings — one truth, two audiences.

References

Butler, R. (1988). Enhancing and undermining intrinsic motivation: The effects of task-involving and ego-involving evaluation. British Journal of Educational Psychology, 58.
Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1).
Kluger, A. N., & DeNisi, A. (1996). The effects of feedback interventions on performance. Psychological Bulletin, 119(2) — attention-to-self vs attention-to-task, the meta-analytic base.

© Glu IO Pty. Ltd. — Wiz Kids (wiz.kids). Link freely; republication requires permission — see terms. Found an error in our reading of the research? We correct fast: tell any teacher piloting Wiz Kids.