91ֱ��

The research could, in the future, help us identify and treat the mechanisms of neurological disorders such as epilepsy, autism and Alzheimer’s.

The study also has applications in developing algorithms for machine learning and artificial intelligence by enhancing our knowledge of how biological brains process information.

When a neuron fires, it excites or inhibits other neurons. These two opposing forces need to be carefully balanced in the brain for it to function properly. This balance requires a process known as homeostatic plasticity, where the brain compensates for changes in neuronal activity. Relatively little is currently known about how this process works in living brains.

Researchers from the Neuroscience Institute have revealed how homeostatic plasticity works in a part of the fruit fly’s brain called the mushroom body - the memory centre for smell. 

Published in the Proceedings of the National Academy of Sciences, found that this brain structure compensates for prolonged imbalance through a combination of increased excitation and decreased inhibition, with these two mechanisms contributing differently for different types of neurons.

Dr Anthi Apostolopoulou, co-author of the paper and postdoctoral researcher from the Department of Biomedical Science, explains “When excitation and inhibition aren’t balanced, you risk getting runaway positive feedback loops. 

“With too much excitation, one neuron excites other neurons, which excite other neurons, some of which excite the first neuron again - this is what's thought to happen in an epileptic seizure. With too much inhibition, the whole brain would shut down. 

“Yet the brain is constantly changing as you learn from the environment. A lot of this learning involves strengthening and weakening connections between neurons, which risks upsetting the balance between excitation and inhibition. Homeostatic plasticity is one tool the brain uses to prevent imbalances from spiralling out of control.”

The fruit fly's mushroom body allows a fly to learn to associate a particular odour with reward or punishment. The memories are stored in neurons called Kenyon cells, which need to encode the odour "sparsely" to allow the fly to learn to distinguish between different odours. 

The Kenyon cells are excited by inputs carrying odour information, but are inhibited by a neuron called APL to suppress their overall activity. If there's too much excitation, the Kenyon cell activity won't be sparse enough to discriminate between odours. If there's too much inhibition, the Kenyon cells might be totally silent and not detect the odours at all. 

This study investigates how the fly brain sets up and maintains the circuitry so that Kenyon cell activity is in the correct 'Goldilocks' zone. 

Dr Andew Lin, co-author and Lecturer from the Department of Biomedical Science, explains “Our findings establish the fly mushroom body as a new model system for homeostatic plasticity in vivo, and we plan to use it to investigate homeostatic mechanisms in more detail - for example, what genes are involved. 

“As in flies, homeostatic plasticity also occurs in the human brain, and it’s believed that problems in homeostatic plasticity might underlie neuropsychiatric disorders like epilepsy, autism or schizophrenia.

“While it’s difficult to study in detail what happens in the brain inside a living mammal, this is easier in flies because they have simple nervous systems and scientists have invented many genetic techniques for studying fruit fly brains.

“Many of the same genes and general principles govern brain function in humans as in fruit flies, so our work could someday lead to new treatments for disorders like epilepsy, autism or schizophrenia. With computational colleagues in the Neuroscience Institute, we’re also using what we’ve learned from fruit fly brains to inspire new algorithms for machine learning.”