EEG with fNIRS, and the fight for space on the scalp

Everyone asks about channel count. The question that decides the setup is how much scalp you have.

An fNIRS channel is a pair of optodes roughly 30 mm apart, and that spacing sets how deep you are sensitive. EEG electrodes sit on the 10-20 positions. Both want the same head, and there is only one head.

So every electrode you add costs you an optode position, and the other way round. That trade is the whole design.

Decide which one is primary

Before the purchase order, not after.

If EEG is primary

g.SENSOR fNIRS adds 8 fNIRS channels to a g.Nautilus, g.USBamp or g.HIamp that you already have.

One amplifier records both. No LSL, no trigger to split, no per-device offset to measure, because there is one clock and one converter.

g.tec puts preparation for fNIRS plus 32 EEG channels at about 10 minutes with active electrodes. That number decides how many subjects you actually run.

If fNIRS is primary

Take NIRSport2 for mobile work, NIRScout for a lab bench, or an Artinis Brite.

Put the EEG electrodes into the fNIRS cap. Artinis recommends this and the reason is practical: their neoprene cap absorbs external light and is stiff enough that the optodes do not move.

Then place electrodes on the 10-20 positions and fit the optodes between them. Both g.tec and Artinis give the same advice, because electrodes are the smaller and more forgiving of the two.

When the optode template will not fit between the electrodes, Artinis states the options plainly. Break the template into smaller sub-templates, or cut the number of EEG channels.

For high density on both at once you need combined holders that carry an electrode and an optode in one position. Those exist only for certain EEG devices. On the g.tec and NIRx combination, the g.GAMMAcap carries 8, 16, 32 or 64 electrode holder rings alongside 32 optode holder rings for 16 sources and 16 detectors.

Where we would not spend the money

If the fNIRS is there to support an EEG study, prefrontal activation sitting next to your ERPs, then 8 channels on the amplifier you own beats buying a second system. It costs less, it removes the synchronisation work, and it is ready in ten minutes.

Buy the dense fNIRS system when the fNIRS is the finding and the EEG is the support.

What we would not do is sell you 64 channel EEG and a dense optode array on the assumption you can run both at full density. Without combined holders you cannot, and the discovery usually happens on the day of the first pilot.

Interference, and which direction it runs

fNIRS is optical, so it does not disturb the EEG.

The electronics inside the optodes can. Artinis reports that lowering skin-electrode impedance minimises this interference, and that with low enough impedance it is eliminated. The fix is the electrode preparation you should be doing anyway, which makes this the cheapest problem on the page.

The cap

Dark, so ambient light never reaches the detectors. g.tec makes the same point about cap material.

Stiff enough to hold the optodes in fixed positions, because an optode that shifts by a few millimetres is measuring a different piece of brain.

Light. You are putting two sets of sensors on one head and the movement artefacts are shared between them.

Synchronisation

This is a smaller problem than it is with eye tracking, because the haemodynamic response unfolds over seconds while the EEG resolves milliseconds. You are not chasing sub-millisecond alignment.

You do still need the event markers written into both files. One device gives you that for free. Two systems need LSL and the same trigger word sent to each, which is the architecture we set out in eye tracking with EEG, and where the timing breaks.

In real time and closed loop work, g.tec runs the EEG amplifier as the master device because it has the higher sampling rate.

Short separation channels

A short channel measures the extracerebral signal, which is scalp blood flow rather than cortex, and lets you regress that out of the long channels. NIRxWINGS2 records the systemic physiology alongside.

Budget for these at the start. Adding them later means rethinking the optode layout, and the layout is the thing you already fought the electrodes for.

The parts

If you are still choosing between the two fNIRS platforms, that comparison sits on our fNIRS systems page.

NBT supplies g.tec, NIRx and Artinis in Israel, and installs them.

Send us the area of interest, the number of EEG channels you need and what you plan to report, and we will tell you which of the two routes fits. Which one is carrying the result, the electrical signal or the haemodynamic one?