MindWave · subject-independent (no calibration)
| Condition | Value (%) | Spread |
|---|---|---|
| Rest vs Math · RF · GroupKFold | 71.78 | 95% CI 70.52–78.07 |
| Rest vs Math · XGBoost · LOSO | 73.03 | 95% CI 71.66–79.24 |
| Medium vs Hard · RF · GroupKFold | 52.98 | 95% CI 50.06–58.01 |
| Medium vs Hard · RF · LOSO | 53.29 | 95% CI 50.46–57.92 |
Takeaway. Rest-vs-Math separates well across unseen subjects, but that contrast also changes eyelid state. Medium vs Hard (eyes open in both) collapses to ~53% — below the 56.99% majority-class accuracy baseline.
Caveat. Majority-class accuracy baselines: Rest vs Math 67.49%, Medium vs Hard 56.99%. Balanced accuracy chance is 50%.
Source: reports/modeling_results.md