Uncovering the Mechanics of Memory and Learning
Scientists have come up with an idea that could help us understand why certain animals learn things. They tested it on C. elegans in an experiment, and wrote about the results in a scientific journal called the Proceedings of the National Academy of Sciences (PNAS). This might help us to answer some questions about learning that have been around since Ivan Pavlov’s famous experiments with dogs!
Ilya Nemenman, an Emory professor of physics and biology, said that his lab did some research and learned that associations between stuff are not just made up of one strong connection. They found out that it’s actually a combination of multiple different pathways that all play a part. And this is true in worms— which means the same might be true for bigger animals too, including humans.
William Ryu, a professor at the University of Toronto’s Donnelly Centre, explains that memory formation is not as straightforward as some may think. Instead, it involves many steps like creating both positive and negative associations between events simultaneously. His laboratory conducted research to understand this idea further.
Ahmed Roman, who was a graduate student at Emory University, wrote the paper and is now a postdoc at The Broad Institute. Konstaintine Palanski, who used to be a student at the University of Toronto, also contributed to the paper.
Pavlov’s Experiment
About a hundred years ago, Ivan Pavlov did some experiments with dogs. He discovered that if he made a sound every time before giving food to the dogs, they would start salivating (producing saliva) when they heard that specific sound. Even without the food in sight, the dogs would still know it was coming by simply hearing the special sound.
About 70 years after Pavlov’s experiment, psychologists used the same idea to create a model called the Rescorla-Wagner model. This model shows how strong certain connections can get between events when one event (in Pavlov’s case it was a sound) can be used to predict another event (which was receiving food in Pavlov’s dog). The predicted event will bring less surprise and its connection with the other event will become stronger.
The understanding people had of reinforcement learning in animals helped give ideas to modern theories. This then allowed artificial intelligence systems to use these algorithms. But even now, some parts of Pavlov’s experiments are still a secret.
Pavlov used to train dogs by giving them food after ringing a bell. For many trials, the dogs kept salivating whenever they heard the bell ring – even if there was no food given. However, when the same process kept on repeating, eventually the dogs stopped salivating (even though they still heard the bell) and forgot about their association with the food. This is called ‘extinction’.
Pavlov noticed something curious after he tested his dogs. If he waited a little while and then retested them, the dogs would start salivating when they heard the bell again, even though there was no food around! Neither Pavlov nor more recent studies could explain why this strange thing had happened or accurately show it with numbers.
Uncovering Mysteries of the Brain
Researchers have been using a tiny roundworm called C. elegans to explore mysteries of the brain. It’s only one millimeter long with 1000 cells, and 300 of those are neurons that help it learn. Scientists use this simple system to study the animal’s behavior. They can also apply what they learn from studying the worm to more complicated systems because its neural circuitry (the way it processes information) is just complicated enough.
Have you ever heard of C. elegans? It’s a type of worm that can be trained to like either cooler or warmer temperatures. Scientists test this by placing the worms in a container with different temperatures, but no food. If the worms have been trained to prefer cooler temperatures, they will move to the cool side of the petri dish and if they are trained for warm temps, they’ll go to the warmer side.
People disagree about why worms move towards a certain temperature. Some think it’s because they are expecting food, while others think it’s because they just like that temperature. It is difficult to answer this question because it takes worms a long time to move around the petri dish (which is 9 centimetres long) in order to reach their preferred temperature.
Discovering the Constant Learning of Worms
Nemenman and Ryu wanted to find a way to accurately measure how learning changed over time. So, the researchers used something called a microfluidic device that turned nine-centimeter petri dishes into four-millimeter droplets. This allowed them to quickly do experiments with hundreds of worms, each in their own droplet.
We did an experiment to see how worms move when they encounter different temperatures. We could see quickly which part of the droplet, either the cold side or warm side, they were attracted to and if their preferences changed as time passed.
Researchers discovered that worms moved to the cold side of a droplet if they expected food to be there. Over time, however, this preference weakens and ceases to exist when no food is present.
“We were so surprised because the worms then would avoid being anywhere near the temperature that originally offered them food,” noted Ryu.
The worm then starts moving between temperatures that are either hotter or colder.
The researchers thought that the worm does not just forget about how cold temperatures are linked to food, but starts to connect the cold side with no food. This then causes it to move to a warmer area. Over time, the worm starts connecting the warm temperature with no food and combining this knowledge with its memory of good things that happened when it was cold, makes it go back to cooler places.
Ryu tells us that the worm is constantly learning. The worm is pulled back and forth between feeling cold and feeling warm, because it wants to experience both positive situations and negative ones.
Uncovering the Mystery Behind the Worm’s Learning Pathways
Nemenman’s team made equations that explain how two things can change over time. One of the two things (called “positive or excitatory association”) would bring a worm closer to one temperature, and the other thing (named “negative or inhibitory association”) would move it away from that temperature.
Imagine you lose your keys and you look for them in the desk where you usually keep it. But if you don’t find them at first glance, then you go around different places looking for it. And eventually, when still unsuccessful in finding your lost keys, you go back to that same desk figuring maybe you didn’t search hard enough the first time. Nemenman says it’s similar with how a worm gravitates toward; It all depends on when exactly you take the measurements.
The scientists repeated their experiments in various ways – they changed the temperature of the worms before they started the experiment, and they made them go without food for different amounts of time. In the end, all the worm’s behavior was accurately predicted by using a specific math equation. Afterwards, they tested their hypothesis by genetically modifying the worms and turning off something called “insulin-like signaling pathway”. This route is known to make a negative impact on things.
By changing the biology of a certain type of worm, our experiments showed changes in the worm’s behavior which is what we had predicted using our theoretical model. This suggests that our model accurately depicts how this kind of worm learns.
We would like others to explore whether these results apply to larger animals from different species.
Ryu has come up with a new way of learning that looks at different factors. He explains that this way of learning helps to explain things that other theories struggle to understand.
This research was funded by the Natural Sciences and Engineering Research Council of Canada, the Human Frontier Science Program, and the National Science Foundation. They studied a model of thermal preference in a type of worm called C. elegans to discover two kinds of learning pathways – an exciting one and a blocking one. This can all be found in an article called “A dynamical model of C. elegans thermal preference reveals independent excitatory and inhibitory learning pathways” published in March 2023 in Proceedings of the National Academy of Sciences with a DOI: 10.1073/pnas.2215191120.
1. Finding out how to control echinococcosis in Linzhi, Tibet Autonomous Region of China
2. Seeing if E-learning can help medical workers treat tuberculosis people better
3. Using models and procedures to make cystic echinococcosis stay away
4. Checking if E-learning is a good way for medical staff to keep learning about helping tuberculosis patients: evidence from the China-Gates Foundation Tuberculosis Control Program
1. Scientists have designed and implemented a remote management system in order to keep an eye on, and control, a disease called echinococcosis.
2. Specialists are finding ways to improve the healing process in rabbits with conjunctival injuries through the use of chitosan/gelatin electrospinning membranes.
3. Doctors are consulting each other concerning laboratory tests used for diagnosing a condition known as myasthenia gravis in 2022.
4. There is guidance available for using a medication called selinexor in cases of hematological system diseases in 2022.
4. Using street-view images to make it easier to predict dengue fever at different places in cities
5. Looking at how the Middle East respiratory syndrome coronavirus changed from 2012 to 2020.
