Gero Miessenböck tells us he has a dopplegander – Dr. Gero from Dragonball Z. This sinister character has his skull removed and can control things with his brain. And Miessenböck believes we can control the brain through light. But that’s where the resemblance stops – Miessenböck promises he’s doing this to learn, not to build a robot army.
If we could record the activity of all neurons, we would understand the brain. But this is so difficult, it’s probably as hard as understanding the brain as a whole. We don’t know how to break the code of the brain, how to decipher the signals the brain sends out.
If we want to break the code, we need to play with it and experiment. And that means controlling some of the neurons in the brain. The thought is that if we can control some of the neurons in the brain, we can understand something about how they work.
Galvani’s discovery that the nervous system works through electric impulses is still, in a sense, state of the art. But there are huge limitations – there are only so many wires we can stick into the brain. To create something less invasive, Miessenböck proposes reengineering the brain so we can control some neurons to be receptive to light.
We can make the proper receivers using genetic material. He shows us a neuron that has a light-activated pore, which fires when it detects light. Different mixes of genes could allow different parts of the brain to respond differently to light. We can write messages to the brain, and he suggests, control brains.
He shows us a fly that responds to a flash of light – two neurons have been changed so they are triggered by light. But how do we know the fly isn’t reacting to the flash of light? His assistant carrying out the experiments removed the fly’s head. They can survive for a day without heads, he tells us, but they don’t do much, just stand around grooming themselves. “The only trait that survives decapitation is vanity.” Even without their heads, the flies respond to the light as a control mechanism.
Behavior, he tells us, is a control loop – actions react to disturbances in their environments and feedback allows their action to change. In an experiment, flies walk through a chamber filled with one odor. At a point in the pathway, it crosses a path with another odor, and the fly has a choice. And since the fly is an intelligent being, it’s got an internal narrative around this decision, a critical voice. “You can think of this as the Catholic Church if you’re an Austrian like me. Or the superego if you’re a Freudian. Or your mother, if you’re a jew.”
When the flies moved into the other odor, specially engineered flies with a hundred light addressable cells were activated. The targeted cells are ones that produce dopamine. In one set of flies – those with neurons on the left activated – the fly won’t cross the threshhold between odors. Through experimentation, they were able to discover that “the critic” was a complex of only 12 cells, which then stimulate a part of the fly’s brain called “the mushroom body” which controls movement.
The system is simple enough that it can be replicated through a small circuit – sensors report to a critic which fires actors, which if strong enough will move the fly.
“I don’t know about you, but I find it very exciting to see vague emotional understandings evaporate” in the the face of discovering mechanical systems. But we’ve still got a mystery – the critic, and how intelligent behavior springs from these signals.