New lucid aid: Aurora, The Dream-Enhancing Headband.

Quote:

---

Originally Posted by pointofbeing

Yes you have 3 electrodes, but the ground is passive, meaning you only have two active contacts, which is the minimum for a channel. The point I'm making is not that you guys are making a bad product, just that given that you only have one channel, it's going to be inaccurate and full of noise. This is not a fault of you, or your product it's a fault of science. Obviously you cant just wear a full cap and plug in all the electrodes it would be invasive and ungodly expensive. You are making a device that is reasonably priced and simple, but it's got flaws simply because of it's simple nature. You can't get accuracy from a single channel, nobody could. If you are detecting rem, how? How does a single channel detect rem not just once and a while but all the time? How does it deal with the potential 60hz noise?

---

Actually, it's not that difficult to detect REM using a single channel if you have the right algorithms. Also, cancelling 60hz (or 50hz in Europe) is easy using standard differential amplification techniques. Have a look at the open EEG project if you want to learn how it can be done.

I tried a very simple light/sound device from a friend of mine. It can be used for lucid dreaming, but has no REM detection. It plays music while lights flash gently above your closed eyelids. I went into a meditative alpha state, and had a very interesting experience. I am sure if I went to sleep with the light/sound device, I could use the music and flashing lights as a dream sign to make me lucid, (especially for naps and WBTB.)
This device can work to make people lucid much better than the light/sound machine I mentioned if it can sense REM well. It doesn't seem like a scam to me. I want one for Christmas.

The iWinks team is hosting an AMA right now on reddit.

For those of you who don't know, an AMA means "Ask Me Anything". Basically, you log in on reddi (or create an account), and ask what you want to know about the Aurora in the comment section of the link below, and they will reply to it.

To everyone a little curious about lucid dreaming technology, I will see you there: We are iWinks and we are developing technology to help improve your dreams! AMA : IAmA !

Quote:

---

Originally Posted by pointofbeing

Yes you have 3 electrodes, but the ground is passive, meaning you only have two active contacts, which is the minimum for a channel. The point I'm making is not that you guys are making a bad product, just that given that you only have one channel, it's going to be inaccurate and full of noise. This is not a fault of you, or your product it's a fault of science. Obviously you cant just wear a full cap and plug in all the electrodes it would be invasive and ungodly expensive. You are making a device that is reasonably priced and simple, but it's got flaws simply because of it's simple nature. You can't get accuracy from a single channel, nobody could. If you are detecting rem, how? How does a single channel detect rem not just once and a while but all the time? How does it deal with the potential 60hz noise?

---

Note that this device only has to detect REM, it's not a high-end research EEG device. In fact, they're just using the Neurosky's ASIC to do pretty much anything. No external amplification or DRL's.
Here's the datasheet: http://www.seeedstudio.com/document/...0Datasheet.pdf

They say the device spits out 512 bits per second. So if the ADC is 8 bits, then the bandwidth is only 84sps. That's only good for toy applications and not research applications (the max frequency or the nyquist frequency is 42hz).
So yeah, this device is basically a "toy" used only to detect REM and not a full research platform... If you want a research-grade EEG, check out openBCI or build your own :)

Where do i get it and how much does it cost?! Tell me nnnnnnoooooooooowwwwwww!!!!!!!

"Note that this device only has to detect REM, it's not a high-end research EEG device."

I know it just has to detect rem. The problem is to properly detect REM it needs proper placement of the ground reference, it should be placed on the mastoid, not in a straight line with the other senors on the forehead like they have. When the passive ground is on the mastoid, REM artifacts become very strong and clear, when thrown together in a single line on the forehead it becomes faint and noisy.

Quote:

---

Originally Posted by pointofbeing

REM artifacts become very strong and clear, when thrown together in a single line on the forehead it becomes faint and noisy.

---

I concur, at least that was my experience with the Zeo... Too high SNR... I'll have to check into the NeuroSky...

Here's the specs for the heart of NeuroSky: http://wearcam.org/ece516/neurosky_e...oard_tgam1.pdf

A whopping 10 bit ADC...

Hell, if you want to build your own for a few bucks, take a look here: http://www.bioera.net/bw/ae/active_prototype2.html