Beamformer

A plugin for Alakazam

Tools > 4. Frequency and Component Analysis > Beamformer · epoched → averaged, with a map of power change · Beamformer

Where power changes between a baseline window and an active window of the same length, at every location of a source model, by FieldTrip’s beamformers (Van Veen et al. 1997; Gross et al. 2001). LCMV compares power in the time domain, from the covariance of the trials; DICS compares power at one frequency, from their cross-spectra, which suits an induced oscillation or a frequency-tagged response. The result is the trials’ average, as Average makes it, with the map stored beside it; the view draws the map (Beamformers).

Figure 1: Beamformer: the bins, the method and its frequency, the two windows, the source model and the regularisation.

Table 1: Beamformer’s options.

Option	Default	Meaning
Bins	every ordinary bin	The bins mapped; their trials together make the filter.
Method	LCMV	LCMV (time domain) or DICS (frequency).
Frequency, Smoothing	10 Hz, 4 Hz	DICS only: the frequency, and the multitaper smoothing around it.
Active from/to, Baseline from/to	300 to 500, -200 to 0 ms	The two windows compared. They must be of the same length.
Source model	Cortical sheet	The template Cortical sheet, or a regular Volume grid inside the template brain.
Sheet vertices	5 124	The sheet: 5 124, 8 196 or 20 484 vertices.
Grid spacing	10 mm	The grid.
Regularisation	5 %	The diagonal loading of the covariance or cross-spectra, as a percentage of the mean sensor power, FieldTrip’s lambda.

The value at each location is the relative change, (active - baseline) / baseline, so 1 means the power doubled. A trial rejected anywhere in either window takes no part; the view says how many did.

Figure 2: LCMV power change from -200 to 0 ms to 300 to 500 ms in the N400 data, on the cortical sheet.

Figure 3: The same on a 10 mm volume grid, on slices of the template MRI through the peak.

References

Gross, Joachim, Jan Kujala, Matti Hämäläinen, Lars Timmermann, Alfons Schnitzler, and Riitta Salmelin. 2001. ‘Dynamic Imaging of Coherent Sources: Studying Neural Interactions in the Human Brain’. Proceedings of the National Academy of Sciences 98 (2): 694–99. https://doi.org/10.1073/pnas.98.2.694.

Van Veen, Barry D., Wim van Drongelen, Moshe Yuchtman, and Akifumi Suzuki. 1997. ‘Localization of Brain Electrical Activity via Linearly Constrained Minimum Variance Spatial Filtering’. IEEE Transactions on Biomedical Engineering 44 (9): 867–80. https://doi.org/10.1109/10.623056.