The article is interesting, but I'm puzzled as to why we would want to fuse neurons and silicon chips outside of the brain when silicon circuits can easily simulate the things a neuron does. It seems to me that the important technology is fusing electronics and neurons in the brain, so that people's brains can interface with computers. In fact, this was accomplished several years ago, though the technology is still developing. So far I think most research has involved wiring electrodes to neurons in the motor cortex so that by thinking about "moving" in a certain way, a person (or some other animal) can in fact control a computer, which can in turn control a robot arm or something. Here's a 2004 article, Paralysed man sends e-mail by thought, on one example of the implementation of this technique. Here's another one (1999) on rats controlling a robot arm to obtain water: Real-time control of a robot arm using simultaneously recorded neurons in the motor cortex. The abstract gives a nice short summary with the interesting results of the study. It's interesting to note that at first, the rats move when they move the robot arm, but after a while they learn to direct their thoughts to control only the robot arm, and not their own muscles. Also, neural nets were used to decipher the patterns of neuronal activity in the rat brains that corresponded to the rats' volition/the desired patterns of movement of the robotic arm. Clearly a significant technology for paraplegics. Finally, while it's impressive to see computers or robots controlled "by thought", these aren't really brains interfacing with computer chips... they are just electrodes being stimulated by neurons firing; the data is then fed through a computer to determine what action is intended. When I saw the title of Leslie's article, I was really excited because I thought maybe something else entirely had been accomplished--like a chip in your brain that actually participates in information processing cooperatively with neurons. So if this technology were sufficiently sophisticated, for example, you could look at a math problem requiring lots of serial computation and then just control the chip to do it for you... and then just know the result, because that chip would interact with certain other neurons. I think we're a long way away from this... but in light of the advances with motor-cortex-neuron-recording, I think it's inevitable that the same technology will eventually be extended to other brain functions, like sensation and higher cognitive functions. At some point maybe people will be able to look at 16!/(pi ^ e ^ phi), take a fraction of a second to crunch that through their 40 gigahertz silicon brain implant, and spit out the answer. And do other crazy stuff too, of course. Like record a video of everything they ever see and then play it back later in their head just by thinking about playing it back. Hmm.. but I wonder; what exactly is the plan with these rat neurons glued to silicon chips? Is it something like those things I was imagining? Maybe they have already created hyperintelligent cyborg rats and/or humans and are just trying to keep it quiet, and that's why they didn't really say in the article what they were doing with the technology... maybe.