The Counterfactual Garden

Curriculum field sheet — When the same behaviour has several plausible causes

Independent Sahilverse educational material. Fictional cases; no compulsory personal disclosure.

Opening provocation

A learner misses three assignments. One educator says motivation is the cause. Another says the work schedule. A third suggests study strategies. None yet has a justified causal conclusion. Ask what else each explanation would predict and what observation would count against it.

The mechanism

The room records three known fictional observations: new evening shifts, less sleep, and continued class attendance. Players select up to two interventions. Each choice reveals a hypothesised benefit, a trade-off and a way to check the pathway. The room deliberately refuses to calculate a confident net effect from missing data.

A serious disagreement

A purely individualistic explanation can overlook material opportunity and institutional constraints; a purely systemic explanation can erase the learner's choices. A plausible causal account must explain mechanisms, alternatives and the possibility that several causes jointly operate.

The second-order question

A strict late-work penalty may make deadlines salient but could increase noncompletion if the barrier persists. Changing shifts could improve sleep but threaten income. Teaching retrieval may improve practice quality while leaving time scarcity unchanged. Policies can solve one problem and create another.

Transfer exercise

In a fictional school or workplace, identify one downward trend and three potential explanations. For each explanation, write a predicted observation. Propose a small, consent-respecting intervention and a criterion that would lead you to revise the explanation.

Evidence boundaries

The causal arrows are hypotheses; no randomised study or empirical effect size is provided. The branch consequences are possibilities, not forecasts about a real person. Names and personal circumstances are invented.

Debrief questions

• Which causes are actually observed?
• What does a counterfactual assert that the current data cannot yet show?
• What would distinguish a habit problem from a resource problem?
• Which intervention offers information even if it fails?

Learning cycle

Predict → interact → describe the result → state a limitation → revise or retain → transfer to another context. Do not assign a grade for agreement with the lesson.