Last weekend, I trekked up to Boston for the Open Source Hardware Association Summit. The talks were amazing. It’s a great group to be a part of, and I love watching it grow. This post is not about the OSHWA Summit, however. It is about brains. My brains. And, yes, I do have a working set of brains. That was proved on Thursday afternoon, the day before, in a hotel right near Logan Airport.
My client in the DARPA EEG project came down from New Hampshire for an arranged meeting to view and test-drive a legit commercial EEG. We will need to test our equipment against something real, so he is looking to purchase a multi-channel rig. Serendipity timed everything just right, and I got dropped off at the hotel at exactly the appointed hour. I was going to be the test subject!
That’s me getting loaded up with electrode goo! We used 8 active electrodes, and the EEG hardware is mady by Cortech Solutions. Most of the electrodes were on the back of my head, and some on the sides. That’s a reference electrode there in the front, and a ground clip on my earlobe.
Once we got the cap on, plugged in, and signals showing up on the laptop, we ran a few tests. First, I blinked. You could see in the signals a discreet spike of noise that got smaller in the electrode signals that were placed further back on my head (further away from my eyelid muscles). Next, I gritted my teeth. This produced a high-frequency signal that overwhelmed any other signals. Cool! The afternoon included a Brain Computer Interface demonstration and I controlled the demonstration laptop with my brain! Here’s how it worked.
I wish I had pictures of this. The computer screen was set up in a grid of letters and numbers and command keys (space, backspace, enter, etc) the screen would flash entire rows or columns randomly, and my job was to count or otherwise notice when the letter or key I wanted to use lit up. As a beginner, I was set up with ~30 row/column flashes to ensure the computer was reading my brain correctly. The computer was looking for something called ‘evoked potentials’. My noticing the target key, causes a predictable change in my brainwave pattern. Then the computer locates the correct key by matching which row and column flashes have the best match to the expected evoked potential signal. After training with the system by spelling the word ‘water’, I spelled -on my own, with only my brain and eyeballs- the word ‘tight’. Damn that was cool.
This kind of computer control is something that we will be able to do with the EEG system that we’re building for Open BCI. The site is under construction, and will be fleshed out much more by the time of the NYC Maker Faire, where I will be hanging out at the DARPA EEG booth and participating in a talk on Saturday. More on the details of that in another post.
Brains, who knew how much fun they are?!