This is a place for open discussion of brain and mind tracking.
The Neurosky Mindwave and related technologies present interesting opportunities for those of us wanting to both map brain/mind activity and use it as an interface to control devices. I am interested in currently existing applications for visualizing Neurosky data and I will post some links to things I have found, but it would be great to hear from others.
My other interest is in mapping states of consciousness and providing a kind of feedback that would be useful in mindfulness mediation. Mindfulness meditation is the practice of holding one’s attention in the present moment, having a direct experience of reality without the symbolic/conceptual overlay that the mind churns out almost constantly. But anyone who has practiced this way knows you drift back and forth between observational awareness and conceptual thinking. What would be entirely practical is the use of a EEG device to gently alert you when you are drifting back into conceptual thinking, because it happens subtlety and without notice. This would be like weight training for this invisible muscle of consciousness.
I will be doing some of my own testing to see how well the delineation between these states is picked up by Mindwave when I get my headset later this week. I am wondering if anyone has tried anything like this already?
Krisp,
I think I ended up with using a 30 sec average of the meditation and attention signals from the Mindwave for plotting the results. Data collection was done while I was practicing a meditation tape. I didn't try to capture how often I strayed from the meditation in those trials. You could try using the blink detection to intentionally mark your data when you notice your thoughts straying. Falling asleep is another state change that is obvious from the data.
Have fun!
I am not entirely clear about what you are saying here, can you describe your process a bit more?
You’ll see that the signal is quite noisy. Averaging the measurements over 30s or more should help with that.
Ahh. So realtime measurement wasn’t practical in your experience? Were you using the bundled visualization software?
Good God, look at this description from Vitamin-R:
“One feature that would be of great interest would be if we could figure out based on your current attention and meditation levels, the exact moment when you are losing focus/ become distracted and react immediately by gently nudging your attention back to the task; perhaps by playing the ticking clock sound or by speaking the objective at just the right moment.”
Looks like they are inviting private beta testers too: http://www.publicspace.net/Vitamin-R/mindwave.html
I had problems with the first headset (it was defective) that interfered with all of my early experiments. I looked at Vitamin R at that time so it explains why I did not pursue that path. When I got a working headset I decided to stay with the bundled software (visualization and meditation journal). They were sufficient for my evaluation work.
I may at a later time go back and look at the raw data directly or connect through the Thinkgear connector. The device is capable of realtime data. As mentioned above the data is noisy, so expecting a reliable fast signal on focus or meditation change may be unrealistic with the way they process the data.
One approach to improve this may be a calibration for each user where the machine learns what unique responses you give and remembers your specific brain wave personality. I am thinking of something similar to the speech recognition learning phase. I don't know if experts recognize patterns in the data they are looking at that would support this type of approach. It is certainly not trivial to explore.
Thank you.
The developer of Vitamin R basically confirmed what you are saying. The signal is noisy and a single dry sensor is not enough to ferret out these fluctuations in attention - the job would be more appropriate for an fMRI. They have in fact dropped Mindwave support in version 2 of their software. In the mean time I will be looking at OpenVibe to make the best of the data that is available.
I contacted Mindwave a while ago about application development for their product and I was directed to their forum for assistance. My impression from that correspondence was that they were focusing on mobile apps instead of desktop apps. I am not surprised Vitamin R is dropping support.
Emotive Systems uses multiple sensors, presumably to get spatial resolution on brain signals for more control options.
OpenVibe is not compiled for a MAC so it is not my choice of software. It does look interesting and certainly more flexible than my approach. I still feel this is a long way from the killer app hardware/software combination.
I am looking forward to your results.