I've *very* serious doubts about if/how it could work, given the current understanding of eeg signals. Anyhow if you happen to buy one I'd like to suggest some experiments to understand! I'd love to do a reverse engeneering on that thing (my guess, simple sensors, lots of software .. should be cheap to build clones). - dasnake

really interested in brain computer interfaces. Did you try the hw? - luminous

@dasnake: yes, that's why I'm so curious about it: I'd love to know what you could do with that. :) --- They declare an SDK for Win and Mac. :) - Folletto

@sergio: I've tried a few, but not that one. I'm tempted. ;) - Folletto

http://www.youtube.com/watch... - Emme Ci

Nice video. :) - Folletto

Did you see the HW listed at the middle of this webpage? http://www.sensibilab.campuspo... - Matteo Balocco

This is an italian enterprise based in Lecco (Politecnico spin off, it sounds good) http://www.sxt-telemed.it/product... - Matteo Balocco

a similar, two players, game: I've been dreaming for this (http://www.mindball.se/product...) for years! - Emme Ci

@matteo: the sxt product is a polygraph, no eeg information, but I guess is enough to try to control something (ecg, skin conductivity and respiratory frequency are standard autonomic response to various mind condition, maybe more easy to use than direct eeg) - dasnake

Folletto, now you are responsible for opening me a whole new world of alienation... this is awesome! Anyone else planning to order an Arduino BT enabled board and the Neurosky, out there? - Matteo Balocco

Thanks Matteo for the links! :D I was thinking too about NeuroSky and Arduino, but I'm still thinking about what could be the best offer around. Here a comparison table: http://en.wikipedia.org/wiki... - Folletto

Yup, both SXT and Sensibilab seem just biofeedback sensors, without EEG. :/ --- Judging by the comparison table NeuroSky has just one sensor: nice, but might not be enough. However it's the only one with a Mac SDK and Arduiono hooks... - Folletto

Actually I've read that SXT has also a not-listed product called Kimera-2 (look at the Sensilab webpage I linked above). Maybe it could be worth asking them some further infos. - Matteo Balocco

I think that monitoring autonomic response is a lot more serious than try with eeg. The eeg signals are *very* local and needs a lot of post-processing to extract little useful information if any. - dasnake

Sure, but it depends on what you want to do. ;) I'm not looking for something to monitor the status of a person during a day: it's just one of the possibilities. I need something related to brain activity, in order to think also about "controlling" something. I can't do that with autonomic responses, since they're...autonomous. :D - Folletto

That's why you need autonomic responses! They are the autonomous variation of some body measurables (heart rate variability, respiratory frequency, skin conductivity, etc) to peculiar state of mind. It's like letting the body do the pre-processing of the signal. If you look at eeg directly, instead, you have a much more complex information - difficult to analyze, but also somehow a more partial information - there's not all you need there! If you still want to look at the brain directly it might be better to use near infrared spectroscopy, it's more clean (in terms of both signal and method), but it's still very local. I'd stick with polygraph-like signals to monitor things like the mood a of a person. - dasnake

@dasnake: wait a second, maybe I'm missing a point. With those data you can't *control* anything, since they're all automated, isn't it? You can't control a robot, I mean: left, right, up, down, like you can do with the OCZ device (that's sold as a gaming device). - Folletto

NeuroSky answered my mail telling me that: 1/ around the end of the year they will ship to Europe and 2/ all their software is compatible with Mac OSX (with the exclusion of MindSet Research Tools: NeuroView and NeuroSkyLab). - Folletto

@folletto: I have some doubts about the principles behind devices like OCZ and how they translate eeg signals into defined directions, so I'm not really sure if you can do the same because I don't know what they do, but I'm sure you can use autonomic responses for some kind of control and interaction. The automated part is the response to a state of mind, the state of mind itself might be voluntary. - dasnake

At this time I suppose that the other reaction times are too slow to be operable, while on the other side brainwaves are varied in a very short time. Estimating from the minimum possible delay, the brainwaves variate in the range of hertz, while the other bodily responses are under a wider array of stimulants: nerves, chemistry, etc, that require a longer actuation time, that is better measured in whole seconds. I think that it's the reason behind the fact that nobody tried to control anything with autonomic responses, but they're trying with brain waves. --- However I agree, even the brainwaves maybe aren't the right answer. ;) - Folletto

About OCZ, guessing from their manuals, uses an initial calibration and then uses different arousal states to be mapped to distinct keys. I've seen the tutorial to trigger "jump", "up" and "down" and it uses a single "wave" (even maybe it's compound) mapping jump on the higher state, up on high and down on the lower one. - Folletto

Ok, so you can use autonomic responses: in scientific researc you usually track them to identify arousal states. It's a lot more common than do eeg. - dasnake

I think that maybe we can't understand each other: yes, you can track arousal states, but afaik isn't neither voluntary or fast. :) --- In fact in research it's tracked EXACLY because it can't be controlled consciously :D - Folletto