"In a recent Edge interview, Dan Dennett pitches the most fascinating new idea I've read in a long, long time: That our neurons are powerful computational building blocks in part because they've reverted to an older and slightly feral state.
Here's Dennett :
"Realize that every human cell in your body, including your neurons, is a direct descendent of eukaryotic cells that lived and fended for themselves, for about a billion years, as free-swimming, free-living little agents. They had to develop an awful lot of know-how and self-protective talent to do that. But when they joined forces to become multi-cellular creatures, they gave up a lot of that. They became, in effect, domesticated — part of larger, more monolithic organizations.
In general, we don't have to worry about our muscle cells rebelling against us. (When they do, we call it cancer.) But in the brain, I think, some little switch has been thrown in the genetics that, in effect, makes our neurons a little bit feral. It's like what happens when you let sheep or pigs go feral: they recover their wild talents very fast.
Maybe the neurons in our brains are not just capable, but motivated, to be more adventurous, exploratory, or risky in the way they live their lives. They're struggling amongst themselves for influence and for staying alive. As soon as that happens, you have room for cooperation, to create alliances, coalitions, cabals, etc."
Dennett traces this idea — of the "selfish" neuron — to computational neuroscientist Sebastian Seung. According to Seung and Dennett, it's precisely because of neuronal selfishness that the brain is able to "spontaneously reorganize itself in response to trauma or novel experiences." For example:
Mike Merzenich sutured a monkey's fingers together so that it didn't need as much cortex to represent two separate individual digits, and pretty soon the cortical regions that were representing those two digits shrank, making that part of the cortex available to use for other things. When the sutures were removed, the cortical regions soon resumed pretty much their earlier dimensions.
Or if you blindfold yourself for eight weeks, as Alvaro Pascual-Leone does in his experiments, you find that your visual cortex starts getting adapted for Braille, for haptic perception, for touch.
Why should these [idle] neurons be so eager to pitch in? Well, they're out of work. They're unemployed, and if you're unemployed, you're not getting your neuromodulators, so your receptors are going to start disappearing, and pretty soon you're going to be really out of work, and then you're going to die.
In other words, the selfishness of neurons incentivizes them to be useful — to hook up with the right network of their fellow neurons, which is itself hooked up with other networks (both 'up' and 'downstream'), all so they can keep earning their share of life-sustaining energy and raw materials.
Thus there is, in this view, an internal 'economy' in the brain, in which neurons must compete with each other for resources. This design stands in contrast to the standard, Von Neumann computer architecture, whose parts never have to worry about where their energy is coming from. Without resource contention, there's no need for selfishness. And this is, in part, why computers are less flexible and adaptable — less plastic — than brains.
Plasticity, says Dennett, is itself one of the most amazing features of the brain, and if you don't have an architecture that can explain it, your model has a major defect. I think you really have to think of individual neurons as micro-agents, and ask what's in it for them?
Neurons as agents: This could well be the single most important fact about our brains."====http://www.meltingasphalt.com/neurons-gone-wild/
Here's Dennett :
"Realize that every human cell in your body, including your neurons, is a direct descendent of eukaryotic cells that lived and fended for themselves, for about a billion years, as free-swimming, free-living little agents. They had to develop an awful lot of know-how and self-protective talent to do that. But when they joined forces to become multi-cellular creatures, they gave up a lot of that. They became, in effect, domesticated — part of larger, more monolithic organizations.
In general, we don't have to worry about our muscle cells rebelling against us. (When they do, we call it cancer.) But in the brain, I think, some little switch has been thrown in the genetics that, in effect, makes our neurons a little bit feral. It's like what happens when you let sheep or pigs go feral: they recover their wild talents very fast.
Maybe the neurons in our brains are not just capable, but motivated, to be more adventurous, exploratory, or risky in the way they live their lives. They're struggling amongst themselves for influence and for staying alive. As soon as that happens, you have room for cooperation, to create alliances, coalitions, cabals, etc."
Dennett traces this idea — of the "selfish" neuron — to computational neuroscientist Sebastian Seung. According to Seung and Dennett, it's precisely because of neuronal selfishness that the brain is able to "spontaneously reorganize itself in response to trauma or novel experiences." For example:
Mike Merzenich sutured a monkey's fingers together so that it didn't need as much cortex to represent two separate individual digits, and pretty soon the cortical regions that were representing those two digits shrank, making that part of the cortex available to use for other things. When the sutures were removed, the cortical regions soon resumed pretty much their earlier dimensions.
Or if you blindfold yourself for eight weeks, as Alvaro Pascual-Leone does in his experiments, you find that your visual cortex starts getting adapted for Braille, for haptic perception, for touch.
Why should these [idle] neurons be so eager to pitch in? Well, they're out of work. They're unemployed, and if you're unemployed, you're not getting your neuromodulators, so your receptors are going to start disappearing, and pretty soon you're going to be really out of work, and then you're going to die.
In other words, the selfishness of neurons incentivizes them to be useful — to hook up with the right network of their fellow neurons, which is itself hooked up with other networks (both 'up' and 'downstream'), all so they can keep earning their share of life-sustaining energy and raw materials.
Thus there is, in this view, an internal 'economy' in the brain, in which neurons must compete with each other for resources. This design stands in contrast to the standard, Von Neumann computer architecture, whose parts never have to worry about where their energy is coming from. Without resource contention, there's no need for selfishness. And this is, in part, why computers are less flexible and adaptable — less plastic — than brains.
Plasticity, says Dennett, is itself one of the most amazing features of the brain, and if you don't have an architecture that can explain it, your model has a major defect. I think you really have to think of individual neurons as micro-agents, and ask what's in it for them?
Neurons as agents: This could well be the single most important fact about our brains."====http://www.meltingasphalt.com/neurons-gone-wild/