Technology Quarterly

LarsWestergren writes "There is an unusually interesting Technology Quarterly available for free from The Economist where they discuss some of the more interesting new areas in the area of science and technology. Of most interest to Slashdot might be Open source's local heroes, or perhaps playing Pac-Man on thought-controlled computers. Among the other articles this month: Predicting microweather, transparent magnetic memories, smart robotic transplants, how to bake the perfect chip, and Benoit Mandelbrot - the father of fractals."

Comments

From the "thought-controlled computers" article...

[tcopeland]

Furthermore, it takes an expert about 45 minutes to attach a 128-electrode EEG to someone's head. Worse, it also requires the use of conductive gel, which hardly makes for a consumer-friendly product.

Yikes. But towards a possible solution:

"So we need EEGs where you don't need the gel," says Dr Muller. "You put some electrodes in a baseball cap and you have a wireless downlink. But there is still a hardware problem. Brain signals measured non-invasively, and without gel, have very small voltages. So you need a powerful low-noise amplifier."

Measuring data

[Isopropyl]

AROUND five exabytes (5 billion gigabytes) of information was created in 2002, up from around two exabytes in 1999, according to the latest "How Much Information?" survey produced by the School of Information Management and Systems at the University of California in Berkeley.

How does one go about measuring this? It seems wildly inaccurate; either they're using a complex algorithm to model data creation, or they're taking a shot in the dark.

Because of the difficulty of estimating such figures, however, all of their numbers have wide margins of error.

I'll say! Give or take, say, five exabytes or so...

Wow

[blackmonday]

...Match this with the heads up display from the earlier article and the possibilities are endless. Imagine checking your email using nothing more than your glasses and your brain wave mouse! Where do I invest my pennies?

what about the learning curve?

[whovian]

Did you know that your brain was preparing for that movement a full half-second before it actually took place? Even more spookily, your mind knew which hand it was going to lift before you made the conscious decision to lift it.

This sounds rather interesting, but it seems it would apply to people who have already learned a task. Therefore, the neural connections would already be "connected" and trained.

But what about teaching somebody a new task using an EEG hat or such? You'd then use this device to find out how the brain learns. I mean, originally....the first bootstrap, so to speak.

I don't think it would be entriely useless to apply to learning new experiences either. Although your brain would draw on that which was previously learned, it would still be trying to absorb a wealth of new information.

Thought-controlled Pacman

[attonitus]

This seems like more fun:

Human Pacman

Re:The article I liked best

[endoboy]

The NY Times recently has been featuring numerous full page (and even double page) adds on the subject of Damadian's exclusion from the Nobel prize... the cost is clearly in the 100's of thousands; I wonder who's paying the bill?

Re:From the "thought-controlled computers" article

[gl4ss]

i was at a demonstration recently about this.

it took like ~3-5mins to attach some 6 or so sensors(a the cap took some time first up front and then getting the contact good on all of them by wiggling). the biggest cap they had was 256 channel(also maximum for that system).

for demonstration i guess the most obvious things you could make out were the sensors related to eye movement. and the activation or something bump up in another test where the subject was shown pictures and told to press space bar everytime a motorcycle picture came up.

though, the enmg(?) was more fun, that's used to measure conduction in nerves(that they work, sometimes vital to know if nerves don't work or not when the patient is unconscious), it was more fun because of the shocking effect that could be used to shock your arm(so that you would swing your arm unwillingly, that would be fun in force feedback).

Uh oh

[E1v!$]

LIFT your left hand. Did you know that ... your mind knew which hand it was going to lift before you made the conscious decision to lift it. ..... Car manufacturers might even develop vehicles that integrate the driver's thoughts with the braking or steering system. In a crash, that half-second could be the difference between life and death.

So now, even before I realize I'm thinking about smashing into the car infront of me (because they keep jamming on their brakes for no reason) my car will do it? Road rage will take on new meaning.

"Honestly officer, it was the software...."

Believe it or not...

[NodeVoltage]

I had the same kind of idea a couple of years ago for a design project for school. I got the idea from a program I saw on Discovery, they were using the EEG as a lie detector kind of deal. I thought that was dumb and figured reading brain waves would be more useful for controlling things like a game :]. The equipment was a little to expensive for me and I was unable to secure a grant so I gave up on the idea, but not after doing a little research. I was pleased and jealous to read the article that someone was actually working on this. At least it clarifies some of my theories. The main problems I thought would be difficult to solve would be reading the small amplitudes of the beta brain waves, 5-50 mV (if I recall correctly), without that icky gel crap, filtering out any outside noise, deciphering the right brain signal (i.e. "move left"), adjusting the device to be able to read different persons brain signals (since everyone's brain is wired differently), and moving all this stuff off the "baseball cap" so it doesn't weigh 50 lbs. I had thought the hardest would be creating software to read the brain signals and to adjust to different people, but I guess they solved that. I had also thought of another side effect that I'm sure they must have thought of. If one was able to correctly read a persons brain signals, would it not be possible to reverse the process and send signals to the person?