A multi-institutional team of researchers led by Virginia Tech VTC’s Fralin Biomedical Research Institute has for the first time identified specific patterns of brain chemical activity that predict the rate at which individual bees learn new associations, providing important insights into the biological basis of learning and decision-making.
The study, published in Sciences Advances, found that the balance of the neurotransmitters octopamine and tyramine can predict whether bees learn to associate odors with rewards quickly, slowly, or not at all.
Because the same ancient brain chemicals that guide learning in honeybees also influence attention and learning in humans, the findings may help scientists better understand why individuals learn at different rates and how these processes go awry in various brain diseases.
Specific patterns of brain chemical activity emerge before learning begins and again when learned behaviors first appear, indicating how quickly individual bees learn. The research helps explain how chemicals in the brain drive attention and enhance learning, and could have implications for basic biology, medicine and agriculture.
The study also highlights the ability to combine neuroscience and machine learning to study complex brain chemistry in the living brain in real time. By measuring the release of multiple neurotransmitters simultaneously, researchers can begin to understand how complex interactions shape learning across species.
Examining the bee brain
Bees buzzed around the lab of computational neuroscientist Reed Montague, a professor at VTC’s Fralin Institute for Biomedical Research. Montague collaborated on the project with Arizona State University professor and behavioral neuroscientist Brian Smith.
The study builds on Montague’s previous work on learning in bees. In a paper published in Nature in 1995 and cited more than 400 times, Montague and colleagues devised a computer model that predicts how signals from specific individual neurons might help bees forage in unfamiliar environments by learning which sights and smells are worth pursuing.
The research builds on the early contributions of the late neuroscientist Martin Hammer, whose work advanced scientific understanding of the neural mechanisms of learning and memory in insects.
Bees live relatively short lifespans in a complex social system, providing researchers with a model to study cognition in natural and laboratory settings.
Insects are amazingly sophisticated.
“Bees aren’t born into this world knowing what they need to know to find flowers and harvest nectar and pollen,” Smith says. Additionally, foraging bees have a lifespan of only a few weeks, and their lives are spent within a 3-5 mile radius of their colony. “This is a huge area for an animal with a small brain.”
The environment is also constantly changing, with flowers blooming and disappearing within hours, days, and weeks.
“That bee must be a learning machine,” Smith said. “You must be willing to forget what you learned yesterday and learn something new today. If you don’t do that, you will never be able to carry out your mission within the colony.”
In Montague’s computational learning model, bees learn from a series of successive predictions that can lead to a reward.
“Bees have a sophisticated system for pursuing this,” Montague says. “They can use the system to make prudent or risky choices.”
The model matched observed insect behavior in experiments. “I applied this to the bee brain and showed that you could theoretically guide bees from flower to flower in a way that perfectly matches the foraging statistics of bees,” he said.
From humans to bees
During his visit with Arizona State University’s Smith and colleagues, Montague shared the latest research on a new method he and his team at Virginia Tech have conducted to measure monoamines, including dopamine and serotonin, in real time and in less than a second in human patients undergoing deep brain stimulation therapy for Parkinson’s disease and essential tremor.
Smith was familiar with Montagu’s 1995 paper. His own research focuses on learning and memory in insects and mammals, including how animals learn about smells, research that can help inform neurological conditions.
Smith learned about Montague’s groundbreaking work on humans and asked questions.
Could Montague have implanted electrodes into the brains of bees?
Shortly after, Smith and Bee were on a plane headed to a research lab in Roanoke.
Montagu’s early research was theoretical.
“We didn’t even have a way to measure monoamines or the brains of bees,” Montagu said. “Now we’ve done this work in humans and turned it back into tiny electrodes that can be inserted into the bee’s brain while they learn and condition.”
Even though their brains are small, bees can teach us a lot.
“We’re pushing honeybees forward as a model for an incredibly sophisticated type of learning and memory task,” Smith said. “I’ve always been interested in measuring in real time when these neurochemicals are released to understand what kind of signals they generate that cause neural networks to be in one state or another to remember something or literally forget something.”
Related brain chemicals are involved in conditions such as addiction, major depression, and attention deficit disorder. These ancient chemicals and systems have evolved in honeybees over 130 million years.
“These are evolutionarily very old systems that are still present in our brains,” Montagu says. “We can condition bees to stimuli in the world that are associated with humans.”
Chemistry learning connections
Bees are a well-established model for studying learning because they rapidly form associations between odors and food rewards. In the Roanoke lab, researchers studied the proboscis extension response of bees that extend their feeding tubes when they learn that a particular odor predicts sugar.
Seth Batten, a senior researcher in Montague’s lab, had experience measuring neurotransmitters in several organisms. “But never inside a bee,” he said. “Not only was it a fun and rewarding engineering feat to create a system that could make measurements in such a small brain, but it was amazing to see how complex these organisms are and how quickly some of them learn.”
Some bees will learn the odor-reward pairing after just a few repetitions, while others may need to repeat it many times or may not learn within the same time frame.
The researchers recorded subsecond estimates of four key neurotransmitters (dopamine, serotonin, tyramine, and octopamine) important for sensory processing and learning in honey bees. The measurements were taken from the antennae, an early processing center for odors, using machine learning techniques that track multiple chemicals simultaneously.
They found that bees can be classified as learners and non-learners based on whether they develop conditioned responses to odors. Some of the bees that learned formed the bond in just three odor-sugar trials, while others required up to eight trials. The variation was strongly related to the timing and strength of the antagonistic signals between octopamine and tyramine.
Bees that emitted faster and stronger signals when first exposed to the odor tended to learn faster when the reward was introduced. This relationship held even though the odor had not yet been combined with sugar.
The same push-and-pull pattern between octopamine and tyramine reappeared when the bees first showed a learned response, but it still reflected how quickly the bees learned. Dopamine and serotonin did not show this pattern.
As learning progressed, neurotransmitter patterns continued to diverge between learners and non-learners. In learned bees, octopamine and tyramine responses changed significantly after the learned behavior appeared, whereas dopamine and serotonin levels gradually decreased during training. Non-learners showed little change over time.
The findings suggest that signaling between octopamine and tyramine plays a central role in setting learning sensitivity and regulating learning duration once an association is formed.
“From a biomedical perspective, understanding neural networks provides insight into how the larger brain works,” Smith says.
In addition to informing basic science and human and animal health, this research also has implications for food supplies through the role of bees as pollinators. “A lot of our agriculture relies on bees,” Smith says.
/Open to the public. This material from the original organization/author may be of a contemporary nature and has been edited for clarity, style, and length. Mirage.News does not take any institutional position or position, and all views, positions, and conclusions expressed herein are those of the authors alone. Read the full text here.
