Even the humble fruit fly craves a dose of the happiness hormone, according to a new study from the University of Sussex showing how they can use dopamine to learn in ways similar to humans. Computer science experts at the University of Sussex have developed a new computational model that demonstrates a long-awaited link between insect and mammalian learning, as described in a new article in Nature Communications.
Dr. James Bennett and colleagues built in anatomical and functional data from recent experiments and modeled how the anatomy and physiology of the fruit fly brain can support learning according to the RPE (reward prediction error) hypothesis.
The computational model shows how dopamine neurons in an area of the brain of a fruit fly known as the mushroom body can generate signals similar to dopamine neurons in mammals, and how these dopamine signals can reliably instruct learning.
Scientists believe that determining whether flies also use predictive errors to learn could lead to more humane animal research, allowing researchers to replace animals with simpler insect species in order to study the mechanisms of learning in the future.
A better understanding of mental health problems
By opening up new ways to study neural learning mechanisms, the researchers hope that the model could also be helpful in providing a better understanding of mental health problems such as depression or addiction, which are based on the RPE hypothesis.
Dr. Bennett, a research fellow at the University of Sussex’s School of Engineering and Informatics, said, “With our computational model, we were able to show that data from insect experiments did not necessarily conflict with predictions from the RPE hypothesis, as previously assumed.
Building a bridge between insect and mammal studies to learn
“Bridging the gap between insect and mammalian studies for learning could open up the possibility of harnessing the powerful genetic tools used to conduct experiments on insects and the smaller size of their brains to determine brain function and disease in mammals, including humans. to understand . ”
Understanding how mammals learn has come a long way thanks to the RPE hypothesis, which suggests that associative memories are learned in proportion to their inaccuracy.
The hypothesis has had considerable success in explaining experimental data on learning in mammals and has been extensively applied to decision making and mental illnesses such as addiction and depression. However, scientists have struggled to apply the hypothesis to insect learning due to conflicting results from various experiments.
Simplification through simulation can implement learning
The University of Sussex research team created a computational model to show how the main features of the anatomy and physiology of the mushroom body can implement learning according to the RPE hypothesis.
The model simulates a simplification of the mushroom body, including different types of neurons and the connections between them, and how the activity of these neurons promotes learning and influences the decisions a fly makes when certain decisions are rewarded.
Five novel predictions about the influence of neurons
To better understand learning in the fly’s brain, the research team used their model to make five novel predictions about the influence of different neurons in the mushroom body on learning and decision-making, in the hope that they will fuel future experimental work.
Dr. Bennett said, “While other models of the mushroom body have been made, to the best of our knowledge, no other model has ever shown connections between dopamine neurons and another group of neurons that predict behavior and lead to rewards. For example, if the reward is the sugar content of foods, these correlations would enable the predicted sugar availability to be compared with the actual sugar ingested, which could help to learn more accurate predictions and appropriate behaviors when searching for sugar.
“The model can explain a variety of behaviors that fruit flies display when the activity of certain neurons in their brain is either silenced or artificially activated in experiments. We also suggest connections between dopamine neurons and other neurons in the fungal body that have not yet been reported in experiments but that would help explain even more experimental data. ”
Thomas Nowotny, Professor of Computer Science at the University of Sussex, said, “The model combines learning theory and experimental knowledge in a way that allows us to think systematically about how fly brains actually work. The results show that learning in simple flies can be more similar to learning than previously thought. ”
Read the original study, “Learning with Gain Predictive Error in a Model of the Drosophila Mushroom Body” in Nature Communications.
