PhysOrg has a brief article on the various ‘mind reading’ headsets that are in the pipeline and could make it onto the gaming market this year.
The article mentions several systems that are apparently close to release and notes some of technology which is intended to allow ‘thought control’ of games:
Emotiv, a company based in San Francisco, says its mind-control headsets will be on shelves later this year, along with a host of novel “biofeedback” games developed by its partners.
Several other companies – including EmSense in Monterey, California; NeuroSky in San Jose, California; and Hitachi in Tokyo – are also developing technology to detect players¬¥ brainwaves and use them in next-gen video games.
The technology is based on medical technology that has been around for decades. Using a combination of EEGs (which reveal alpha waves that signify calmness), EMGs (which measure muscle movement), and ECGs and GSR (which measure heart rate and sweating), developers hope to create a picture of a player¬¥s mental and physical state. Near infrared spectroscopy (NIRS), which monitors changes in blood oxygenation, could also be incorporated since it overcomes some of the interference problems with EEGs.
I’ll be intrigued to see how well they work, but I suspect they’ll be more of a novelty than a genuinely useful addition for avid gamers, at least at first.
This is largely because the main technology for reading brain activity is EEG.
Even with thousands of pounds worth of kit, neuroscientists get participants to do the same task over and over and then average the results to get a reliable waveform.
This is partly because this technology is a relatively crude measure of the total electrical activity that happens over a large area (so on any one occasion the wave will be influenced by a number of other brain functions going on at the same time), and partly because the electrical activity from something as small as the eye-blink muscles drowns out the signal from the brain.
It’s interesting that the article mentions near infrared spectroscopy as another possible way of reading brain function (as used by Natalie Portman).
This involves beaming near-infrared light into the head, where it penetrates the skull and gets absorbed by brain to differing degrees, depending on how much blood is in the area. The amount of light that bounces back can be used to infer blood saturation and, hence, brain activity.
However, changes in blood flow lag behind the activity of the neurons by up to 5 seconds (and interestingly, this varies as we age). This is because blood is ‘called in’ to replenish the local nutrients that are instantly available but in short supply.
Similarly, systems that measure skin conductance or heart rate (a proxy measure for arousal or stress) have a similar problem with lag.
So gamers wanting to control games at the ‘speed of thought’ are likely to be disappointed. EEG is too noisy, NIRS is too slow.
What the headsets might do well, however, is something quite different.
The MIT Affective Computing group have spent several years looking at how computers could present information differently depending on the emotional state of the user.
According to Jonathan Moreno’s book Mind Wars this is also something that the US Military has great interest in, and you can also see how it would enhance games.
The readings from the headset will probably do a better job of keeping track of the easier to measure and relatively slow moving responses like arousal and stress, and these could be used by game designers to enhance your experience (maybe to slow things down if you’re too stressed and under-performing to avoid frustration, or to pump-things up at tense moments).
One of the most interesting possibilities is what might happen when hackers got hold of the systems.
Suddenly, they’ll be thousands of people with standard kit for reading physiological responses and, to a certain extent, brain function.
As soon as someone finds a way to reliably read a novel type of brain function, even with this limited technology, everyone will be able to use it.
Furthermore, it might lead to some fascinating home cognitive neuroscience experiments and demonstrations. Imagine having a home NIRS system – rock on!
3 thoughts on “Playing mind games, off the shelf”
I’ve been writing on these devices and other novel input devices on my blog, Myomancy for a while.
My main interest is in how these devices could be used in education. Neuro-feedback devices are the obvious options but the evidence on their effectiveness is not strong.
I’m excited about other technologies such as how simple head tracking [ http://www.myomancy.com/2008/03/wii-head-tracking ] could be used to teach posture and balance skills. Or how the iPhone’s technology could change education [ http://www.myomancy.com/2008/02/how-the-iphone-will-change-education ].
For pure tech-porn, check out these Foldable Interactive Displays [ http://www.myomancy.com/2008/03/more-future-training-technology ]. Nothing to do with brains or education but an exciting bit of technology.
I wouldn’t say that NIRS is slow. It’s measured at 10 Hz and if blood flow is holding up response you can measure the decrease in deoxyhemoglobin, which is fairly instantaneous.
You’re quite right. NIRS itself isn’t slow, but blood response in the brain is (e.g. the several second lag from neuronal activity to increase blood flow). As NIRS measures blood saturation, all in all, the technology as a whole is too slow to give an online reflection of cognitive activity.