Electrode Cap vs. Ultracortex
Hi all,
New here and currently deciding between purchasing the Electrode Cap and the Ultracortex.
I want to use 16 channels to measure the brains response to entrainment, more so in the context of (shoe-string budget) cognitive psychology research than BCI.
My main requirement is to get the highest possible frequency response (ideally up to 80 Hz with the Wifi-board), and I assume that the Cap with wet electrodes will be best due to its lower impedance. Is that correct?
It would also be good if the frequency response is stable for around an hour. That is the power recorded for Alpha at T0 and T60 should be equal if the brains actual Alpha power is equal (i.e. the relative coupling between measured and actual frequency response should stay consistent, maybe hard to test in practice...). I'm not planning to move around during this time so I don't think electrode movement will be a problem but does sound feasible? Maybe gel evaporation could be a problem for wet electrodes.
Thanks all!
Comments
Gavin, hi.
You mentioned:
Gamma EEG recorded from the scalp is very weak, even at 40 Hz. There is also potential contamination with scalp muscle EMG. There will not be any EEG to find at 80 Hz. Can you clarify what sample rate you want to work with? Without the Shield you are limited to 250 Hz with 8 channels and 125 Hz with 16 channels. With the Wifi Shield it may be possible to get 16 channels at 250 Hz or 500 Hz. Please read the disclaimer on the Wifi Shield product page:
https://shop.openbci.com/collections/frontpage/products/wifi-shield?variant=44534009550
Here are some references on the EMG EEG contamination topic:
https://www.frontiersin.org/articles/10.3389/fnhum.2013.00138/full
I also attached Whitham's 2007 paper, mentioned as one of the references in the above review.
As far as best recording quality, yes the cap with gel will be superior to the dry electrodes. And gel sessions can last say up to an hour. Any gel based cap will work, your department already might have one.
Regards, William
Rereading your initial question, it does seem your project is interested in alpha band, typically 8 to 12 Hz. I was thrown off by your mention of frequency response at 80 Hz. Perhaps you can clarify. You don't need high sample rates to catch alpha band.
Note that even without the Wifi Shield, you can record to the SD card on the Cyton at the complete 250 Hz rate and 16 channels. So the question is, do you need live data, or just want to record.
Good to know!
Thanks for the advice William!
I'll go with the cap/wet electrodes then, I figured these would be the best option but wanted to confirm. I'm actually working independently on this and unfortunately don't currently have any colleagues to mooch supplies from, otherwise, I'd probably try and borrow equipment to start off.
In terms of getting a good 80Hz signal, I realize this may be a bit ambitious. While I agree there will be EMG noise, I think this will be less problematic in an entrainment paradigm, as the entrainment cue is fixed at a single frequency, while the EMG noise is spread across a range of frequencies. For instance, the frontiers paper you sent (thanks for that, it's a good reference) mentions:
Still, my choice of 80Hz was somewhat arbitrary - I'm interested in looking at entrainment properties across all frequency bands, and simply thought that twice 40Hz (with its current clinical relevance) would be good to aim for. Getting any gamma would be great for now!
I do want real-time data so I will try and get the WiFi-shield working for this (I had seen the warning on this, hopefully, I can push it into working for me!). If it doesn't then I'll just live with 125 Hz for the now. Also, the Cyton has five spare input pins D11/12/13/17/18 (well, D12/12/17 if I use the Wifi board, I saw it takes two pins) - is it able to also stream these (particularly the three analogue pins) as well as the 16 channels at the full rate? I plan to use one at least for the pulse sensor, but out of interest, is it possible to attach additional electrodes to the analogue pins, or do those pins have a different reference/bias voltage to the bioamplifier pins?
The ADS1299 pins, the primary input pins, operate in the microvolts to millivolts range. The PIC32 / chipKIT processor analog input pins have a range of 0V to 3.3V full scale. These are accessed with the "analog read mode" as described in the SDK. The values read are stuffed into the Aux channel locations in the data packets. These are 16 bit values. Yes Aux channels stream at same rate as main channels.
Oh right, that makes perfect sense that the extra pins aren't amplified, that obvious retrospect in retrospect. Good to know that they stream at the same rate as the main channels. At the moment I haven't fully decided on the protocol I'll use, so I'm glad to know there is quite a bit of flexibility on the Cyton board.