Classroom Behavior Rules: Reward vs. Punishment Simulator

Rule Set

Reward strength 0.55

Punishment strength 0.35

Rule consistency 0.70

Class size 20

Class Snapshot

Engaged Neutral Disruptive

Avg. engagement

--%

Disruption rate

--%

Learning outcome index

--%

How it works

Each student is an independent agent with two hidden state variables: engagement e ∈ [0,1] and resentment r ∈ [0,1]. At random intervals a student "acts" — on-task with probability equal to their effective engagement:

P(on-task) = e · (1 − 0.6·r)

The teacher notices any given act with probability equal to rule consistency C. When noticed:

on-task + reward:      e += R · (1 − e) · k_r
disruptive + punished: e -= P · e · k_p
                        r += P · k_res

Unnoticed disruption lets misbehavior spread a little (small downward drift in e); resentment r always decays slowly over time, while e relaxes gently toward a neutral 0.5 baseline when neither reward nor punishment is applied (extinction). This reproduces two textbook classroom-management effects: reward alone drifts engagement up slowly but is fragile without consistency, while heavy punishment suppresses visible disruption quickly but builds resentment that caps how engaged a student can ever become — exactly the "compliance without engagement" pattern seen in real classrooms.

Reward strength — how much a noticed on-task act raises engagement.
Punishment strength — how much a noticed disruptive act suppresses the current act and banks resentment.
Rule consistency — probability the teacher actually notices and reacts to any given act (the single biggest lever in real behavior-management research).

⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.