Table 2. Comparison between model and empirical performance in grip force tracking.
| Model data | Empirical data | ||||
| Low force | High force | Low force (5N) | High force 10% MVC | ||
| Relative hold error | Controls | 0.75 (ref) | 62% | 9.8 (ref) | 61% |
| SCZ | 134% | 106% | 145% | 108% | |
| CV hold | Controls | 0.03 (ref) | 62% | 0.028 (ref) | 57% |
| SCZ | 133% | 106% | 129% | 89% | |
| Release duration | Controls | 73 ms (ref) | 98% | 73 ms (ref) | 99% |
| SCZ | 140% | 138% | 143% | 127% | |
Relative error during the hold period (Relative hold error), Coefficient of variation of force during the hold period (CV hold), release duration for controls and for patients (SCZ). For each variable, the ‘low force, control subject' data was taken as reference (ref) and the comparative values of the three other conditions expressed as %reference. Note that the large difference between the relative hold error (the reference values of the controls) between the model and empirical data is arbitrary, since the modeled force is dimensionless. This is the reason for the percentage-based comparison.