Table 1.
Classification performance for the platform condition on the test data.
| Platform off | Platform on | |||||||
|---|---|---|---|---|---|---|---|---|
| Accuracy (%) | False positive rate (%) | False negative rate (%) | d’ | Accuracy (%) | False positive rate (%) | False negative rate (%) | d’ | |
| Pre-ICA | 78.3 (0.33) | 19.8 (0.20) | 23.6 (0.65) | 1.69 (0.01) | 71.6 (0.56) | 35.4 (1.52) | 21.4 (0.60) | 0.75 (0.08) |
| ICA | 81.0 (0.49) | 21.4 (0.99) | 16.6 (0.73) | 1.59 (0.07) | 73.7 (0.52) | 26.8 (1.91) | 25.8 (1.75) | 1.26 (0.12) |
| Pre-ICA + Kalman Filter | 77.8 (0.55) | 20.2 (1.05) | 24.2 (1.94) | 1.68 (0.07) | 72.8 (0.53) | 36.6 (1.58) | 17.8 (1.94) | 0.69 (0.08) |
| ICA + Kalman Filter | 83.4 (0.27) | 17.2 (0.53) | 16.0 (0.60) | 1.90 (0.04) | 73.1 (0.8) | 28.4 (2.61) | 25.4 (1.79) | 1.18 (0.16) |
The standard error is given in parentheses below each mean value of the 10 nested cross-validation iterations.