AI: Mapping the brain

This is a little more disconcerting to me than a lot of other similar projects. Mostly, people who are advocating this approach seem scattered and, at best, in the very early stages of their approach (except for the gentleman over at IBM).

FTA: It’s a strong claim, but Meier is coordinating the EU-supported FACETS project which brings together scientists from 15 institutions in seven countries to do just that. Inspired by research in neuroscience, they are building a ‘neural’ computer that will work just like the brain but on a much smaller scale.

and...

How does it do that? Nobody yet knows, but a team within FACETS is completing an exhaustive study of brain cells – neurons – to find out exactly how they work, how they connect to each other and how the network can ‘learn’ to do new things.

Luckily, the admit that they too are also in the early stages, but they seem more organized, they seem to have a better plan and, well, they seem to be taking a combination approach- not trying to simulate the brain directly, nor trying to build analogs for the brain a piece at a time, but cobbling the two methods together.

Sciences moves us one step closer to magic.

Headline: Brain wave patterns can predict blunders, new study finds

FTA: From spilling a cup of coffee to failing to notice a stop sign, everyone makes an occasional error due to lack of attention. Now a team led by a researcher at the University of California, Davis, in collaboration with the Donders Institute in the Netherlands, has found a distinct electric signature in the brain which predicts that such an error is about to be made.

Hopefully this won't be far behind...


OK, so the remembrall is likely to remain in the realm of fiction... its still really cool.

Human Brain Simulations by 2018

Nine years seems optimistic to me, but Dharmendra Modha's one of those guys you have to take seriously. He's leading the DARPA funded SyNPASE team and the IMB Almaden Research Center. He's also the guy who a few years ago created a simulation of a rat's brain. It's also important to note that though a simulation would be incredible, its not necessarily AGI. Its fairly unclear to me what the outputs of the simulation would be, if any. I think this is purely for modelling purposes. None-the-less it would be an amazing feat that would seem to line up with Kurzwiel's predictions for AI breaking in the 2020's.

FTA: Although the brain is still not well understood, Modha said, "there is enough quantitative data for us to be able to begin putting together the pieces." He predicted that by 2018 computers will be able to simulate the workings of the human brain, a breakthrough that will provide researchers with unprecedented insight into how the complex organ operates.

And the same article also provides a little shout out to BMI/BCI: In addition to boosting computer performance, enhanced understanding of the brain will enable people to communicate directly with machines, whether they are robots or mechanized prosthetic limbs. Primates have already proved that such brain-machine interfaces are possible, Miguel Nicolelis, co-director of Duke University Medical Center's Center for Neuroengineering, said during the conference. The researcher and his colleagues last year successfully implanted electrodes in the brain of a monkey in North Carolina that enabled him to control a robot on a treadmill in Kyoto, Japan.

Nicolelis and his team have developed a microchip they expect will allow human brains to communicate with robots using only brain signals and enables the bots to return messages directly to the brain, without the use of sight or touch. Nicolelis said that he hopes the technology will be sophisticated enough to implant into a human brain by 2012 and enable a completely quadriplegic patient to walk again.

'Daddy, why were poets sad?'

'Because they didn't understand protein P11, sweetie.'

FTA: ScienceDaily (Mar. 6, 2009) — It takes weeks or months for the effect of most antidepressants to kick in, time that can feel like an eternity to those who need the drugs the most. But new research suggests that a protein called p11, previously shown to play a role in a person’s susceptibility to depression, activates a serotonin receptor in the brain known for producing a rapid antidepressant response. If scientists could develop drugs to target this receptor, they might produce an effect in as little as two days.


Two days vs. several weeks? Not bad. As usual, I'm not in favor of anti-depressants being administered casually.

Mind Reading... this should end well.

FTA: Wearing a headband fitted with fibre-optics that emit light into the pre-frontal cortex of the brain, they were shown two drinks on a computer monitor, one after the other, and asked to make a mental decision about which they liked more. "When your brain is active, the oxygen in your blood increases and depending on the concentration, it absorbs more or less light," Luu says. "In some people, their brains are more active when they don't like something, and in some people they're more active when they do like something."

Small steps, guys. Small steps. This is just the begining. We're only going to get better, faster and more accurate at this sort of thing.

Still, I'm found of most things with a potential application to BMI . I'm really, really looking forward to immersive VR and a robot butler that'll get me a beer.