Table 2.
Results of uni- and multimodal inversion, using a fixed number of spatial and temporal modes.
| MEG |
EEG | ||
|---|---|---|---|
| Mags | Grds | ||
| Number of sensors (range across participants) | 102 | 204 | 63–70* |
| Number of spatial modes (mi) (range across participants) | 50 | 50 | 50 |
| Number of temporal modes (ri) (range across participants) | 8 | 8 | 8 |
| Empirical (baseline) SNR (plus standard error across participants) | 3.8 (0.3) | 4.3 (0.3) | 4.6 (0.4) |
| Unimodal inversions | |||
| % Data variance explained (range across participants) | 94–98 | 89–95 | 93–99 |
| Conditional source precision (plus standard error across participants) | 59.4 (8.0) | 76.1 (7.0) | 95.3 (8.0) |
| Fused inversion | |||
| % Data variance explained (range across participants) | 91–96 | ||
| Conditional SNR (plus standard error across participants) | 29.5 (2.5) | 9.2 (0.9) | 6.4 (0.6) |
| Conditional source precision (plus standard error across participants) | 126.9 (15.3) | ||
Grds=Planar Gradiometers; Mags=Magnetometers.
*
Number of EEG sensors varied over participants owing to removal of bad sensors.