In a lab at Stanford University School of Medicine, mice see things. And not because they were given medication.
With the new laser technology, scientists have induced certain hallucinations in mice by switching on some neurons with light rays. The results reported the researchers on Thursday in the journal Science.
The technique promises clues to how the billions of neurons in the brain understand the environment. Finally, research could also lead to new therapies for mental disorders, including uncontrollable hallucinations.
In the early 2000s, Dr. Stanford-based psychiatrist and neuroscientist Karl Deisseroth and other scientists made neurons in the brain of living mouse mice to turn them on when exposed to a flash of light. This technique is called optogenetics.
In the first wave of these experiments, the researchers used light to study how different types of neurons function. But Dr. Deisseroth wanted to be able to pick out every single cell in the brain and turn it on and off with light.
So, he and his colleagues designed a new device: instead of just lighting a mouse's brain in light, it allowed the researchers to emit tiny beams of red light that could hit dozens of individual brain neurons at the same time.
[ Like the Science Times page on Facebook. | Sign up for the Science Times newsletter.
To try out this new system, Dr. Deisseroth and his colleagues on the perception of the visual world by the brain. When light enters the eyes of a mouse or a human, it causes nerve endings in the retina to send electrical impulses to the back of the brain.
In a region called the visual cortex, neurons quickly recognize edges and other patterns, which then assemble the brain into a picture of reality.
The scientists inserted two genes into neurons in the visual cortices of mice. A gene made the neurons sensitive to the red laser light. The other caused neurons to produce a green flash at power-on, allowing researchers to track their activity in response to stimuli.
The engineered mice were shown images on a monitor. Some consisted of vertical stripes, others of horizontal stripes. Sometimes the stripes were light, sometimes blurry. The researchers taught the mice to lick a whistle only when they saw vertical stripes. If they performed the test correctly, they were rewarded with a drop of water.
When images were shown to the mice, thousands of neurons in their visual cortices flashed green. A population of cells was turned on in response to vertical stripes; other neurons responded when the mice were displayed horizontally.
Researchers selected a few dozen neurons from each group to reach the target. They again showed the mice the stripes, and this time they also shed light on the neurons of the corresponding group. By switching on the right neurons, the mice were able to recognize streaks better.
The researchers then turned off the monitor and left the mice in the dark. Now the scientists turned on the neurons for horizontal and vertical stripes without the rodents being able to see anything. The mice responded by licking the whistle as if they were actually seeing vertical stripes.
Anne Churchland, a neuroscientist at the Cold Spring Harbor Laboratory who was not involved in the study, warned that this type of experiment could not say much about the inner experience of a mouse.
"Oh, wow, I saw a horizontal bar," she said.
Dr. Churchland said that further research was needed to better understand why the mice behaved in response to the red flashes of light. Have they seen the horizontal stripes more clearly or were they less distracted by misleading signals?
One of the study's most notable findings came about when Dr. Deisseroth and his colleagues narrowed their red light beams to fewer and fewer neurons. They kept making the mice lick the whistle as if they were watching the vertical stripes.
In the end, the scientists found that they could trigger the hallucinations by stimulating only two neurons. Thousands of other neurons in the visual cortex would follow the direction of these two cells and flash green when active. Dr. Deisseroth and his colleagues came to the conclusion – like a snow bank that is about to become an avalanche.
But it does not take a fancy optogenetic device to burn a few neurons. Even if they do not get stimulus, sometimes neurons just shoot at random.
This raises a mystery: If only two neurons are needed, why do not we hallucinate all the time?
Perhaps our brain wiring prevents this from happening. Deisseroth. When a neuron fires randomly, others may send a signal to calm it down.
Dr. Glickfeld suggested that attention could be crucial to trigger the avalanche of neuronal action at just the right time. "Attention allows you to ignore much of the background activity," she said.
Dr. Deisseroth hopes to see what other hallucinations he can trigger with light. In other parts of the brain, he may be able to make mice perceive more complex images, e.g. B. the face of a cat. It could make neurons produce phantom sounds or even phantom smells.
As a psychiatrist, Dr. Deisseroth treated patients with visual hallucinations. In his role as a neuroscientist, he wants to learn more about how individual neurons lead to these images – and how to stop them.
"Now we know where these cells are, what they look like, how they look like shape," he said. "In the future work we can get to know them much better."