Reference:

清華大學 首頁故事
National Tsing Hua University Home Page Story

A cross-disciplinary research team composed of Director Ann-Shyn Chiang of the Brain Research Center at National Tsing Hua University, Professor Chung-Chuan Lo of the Institute of Systems Neuroscience, and Distinguished Chair Professor Ting-Kuo Lee of the Department of Physics has uncovered how fruit flies rapidly decide whether to “eat” or “flee.” Their study found that the fruit fly brain transmits olfactory information through a “hybrid” neural network: “generalist” neurons randomly process common odors such as pheromones and floral scents, while “specialist” neurons process the most important odors for fruit flies—those associated with food.

Previously, the scientific community believed that neural connections in the fruit fly brain were entirely random. However, Director Ann-Shyn Chiang discovered that the connections contain both random links and dedicated “special lines” for processing critical information, achieving a highly precise and efficient division of labor. This breakthrough not only reveals the neural computation mechanisms of insect brains but also opens new avenues for cross-disciplinary applications in neuroscience and artificial intelligence. The findings were recently published in the top-tier international journal Science Advances.

At the center of the fruit fly brain lies the mushroom body, a crucial hub for processing sensory information and memory in insects. Over the past two decades, scientists have debated how mushroom body neurons are connected. For instance, 2004 Nobel Prize in Physiology or Medicine laureate Richard Axel, who sampled about one-tenth of the neurons in his research, concluded that the connections were random. In contrast, Director Ann-Shyn Chiang believed that there should be specific connectivity patterns.

An Academician of Academia Sinica and an internationally renowned neuroscientist, Director Ann-Shyn Chiang led the NTHU team to combine connectomics analysis, in vivo imaging, and computer simulations. They discovered that olfactory neurons in the mushroom body form neither purely random nor fixed connections, but a hybrid pattern with both characteristics. Certain neuron clusters exhibit clear connection preferences, encoding different odors through both dispersed and convergent patterns, thus balancing olfactory sensitivity with diversity.

The mushroom body in the fruit fly brain functions like a miniature central processing unit (CPU), integrating various external signals to help the insect make decisions in complex environments—such as avoiding danger, finding food, or seeking mates.

The team’s research revealed that fruit fly olfactory neurons operate in two roles: “generalists” and “specialists.” Generalist neurons randomly receive a wide range of odors—from food and pheromones to floral and plant scents—and broadcast them to multiple neuron groups, like a radio signal. Specialist neurons, by contrast, process only specific crucial odors, such as fruit or other food scents, transmitting them via dedicated pathways to target neurons, like a direct telephone line.

Professor Chung-Chuan Lo of the Institute of Systems Neuroscience further explained that the fruit fly’s hybrid neural connection pattern resembles a high-strength password, combining randomness with order: “It’s like creating a password from a familiar word plus random characters—such as adding numbers or symbols before and after your name. It might seem illogical at first glance, but it’s easy to remember and hard to crack.”

Director Ann-Shyn Chiang noted that identifying the hybrid “random” and “ordered” characteristics in fruit fly neural connections not only clarifies how the brain processes information in both distributed and centralized ways, but may also offer new clues for treating neurodegenerative diseases such as Alzheimer’s and Parkinson’s, while inspiring the design and application of AI neural networks.

The first authors of the paper are Li-Shan Cheng, a master’s student in the Institute of Physics at NTHU, and Ching-Che Charng, a doctoral student in the Institute of Systems Neuroscience. The corresponding authors are Distinguished Chair Professor Ann-Shyn Chiang and Professor Chung-Chuan Lo of the Institute of Systems Neuroscience, along with Distinguished Chair Professor Ting-Kuo Lee of the Department of Physics. The team also includes Kuan-Lin Feng, a postdoctoral researcher at the Brain Research Center, who conducted fruit fly biology and behavioral studies, and Ruei-Hwang Chen, a doctoral student in the Institute of Systems Neuroscience, who used functional imaging experiments to observe how olfactory signals are transmitted in neural networks.

Director Ann-Shyn Chiang emphasized that Li-Shan Cheng and Ching-Che Charng served as the theoretical analysis core of the team, while Ruei-Hwang Chen and Kuan-Lin Feng were the experimental scientists responsible for validation. “It was the interdisciplinary collaboration—combining expertise in physics, neuroscience, and behavioral studies—that was key to solving the mystery of fruit fly neural connections,” he said.

This research on fruit fly olfactory neural connections was led by NTHU’s Brain Research Center, in collaboration with China Medical University, the National Health Research Institutes, Academia Sinica, and the Kavli Institute for Brain and Mind at the University of California, San Diego. It was jointly supported by the Ministry of Education, the National Science and Technology Council, and the Peng Education and Welfare Foundation.

1. Director Ann-Shyn Chiang (front right) of the Brain Research Center and Professor Chung-Chuan Lo (front left) of the Institute of Systems Neuroscience, with doctoral student Ruei-Hwang Chen (back row, left), master’s student Li-Shan Cheng of the Institute of Physics, doctoral student Ching-Che Charng of the Institute of Systems Neuroscience, and postdoctoral researcher Kuan-Lin Feng of the Brain Research Center, who together decoded the olfactory code of the fruit fly brain.

2. The NTHU interdisciplinary research team that cracked the fruit fly brain’s olfactory code and published their findings in Science Advances. From left: postdoctoral researcher Kuan-Lin Feng, doctoral student Ruei-Hwang Chen, doctoral student Ching-Che Charng, Professor Chung-Chuan Lo, Director Ann-Shyn Chiang, and master’s student Li-Shan Cheng.

3–4. The NTHU interdisciplinary research team that cracked the fruit fly brain’s olfactory code and published their findings in Science Advances.

5. First authors of the fruit fly brain neural connectivity study: Li-Shan Cheng (right), master’s student in the Institute of Physics, and Ching-Che Charng, doctoral student in the Institute of Systems Neuroscience.