Plain sticky notes
Email from Dr. Fetsch- summary of neuroscience journal article
Our goal is to figure out how the brain makes quick decisions based on vision, like deciding when it’s safe to cross a busy street. To do this we train monkeys to make simple decisions while looking at a computer screen, kind of like playing a video game with their eyes (Figure 1A). They look at some moving dots, then they move their eyes left or right to tell us they think the dots were moving mostly leftward or rightward (it’s difficult because some of the dots jump around randomly). We’re also interested in how we get a sense of confidence in our decisions: for instance, if you’re confident there are no cars coming, you might walk across the street, but if not, you’d probably run just to be safe. So for that question, we train the monkey to tell us when they are not confident, by moving their eyes upward. This works because the monkeys care about getting juice, which is their reward. If they choose left or right and the choice is correct, they get a large drop of juice. If they get it wrong, they get nothing. But if they take the ‘sure bet’ (moving eyes upward) they get a small drop of juice, which is better than nothing, so they learn to do that if they aren’t sure about left vs. right. The reason to do experiments with monkeys instead of people is that we can measure brain activity better in monkeys, using techniques that aren’t safe for humans yet. We measure the nerve cell activity in a part of the brain that allows the monkey to see and interpret the moving dots. These cells can be seen in Figure 2B, in green. If you learned about the parts of the neuron in science class, you can even see the axons and dendrites, which let the neurons send and receive information from other cells. The pictures are taken with a microscope, and the cells glow green because they are engineered to have a special fluorescent protein (they don’t normally look like that, it just helps us see them better). The coolest part is this: using the same sort of method used to make them glow green, we can actually turn these neurons on or off like a light switch, by shining a laser on them. That’s Figure 3B. With the laser on, the activity is much lower (the red curve). We did this while the monkeys were playing the video game, and it affected their decisions. For example, some neurons respond strongly to leftward motion, basically telling the rest of the brain that the answer should be “left”. If we turn those cells off with the laser, then the monkey says “right” more often (Figure 7A), and his confidence is also altered as if there was more evidence favoring “right” (Figure 7C). This means that the same neurons responsible for the decision are also responsible for confidence in the decision. Surprisingly, the effect on behavior disappeared after some time, meaning the rest of the brain was able to start ignoring the neurons we were messing with, and start using other neurons to do the task. This is a new example of plasticity, or how brain circuits can change over time. Our study was also one of the first to get this laser method to work well in monkeys, which is a step closer to getting it to work in humans. Eventually, scientists would like to use these tools to help people with diseases of the brain, like Alzheimer’s disease and Parkinson’s disease. So even though my work is not about curing those diseases, using animals to improve the technology can help other scientists and doctors find cures in the future. Let me know if anyone has questions!