A fruit fly’s split-second twitch might have something to teach the next generation of AI.
Researchers at the University of Sheffield say the “lightning fast” reactions of fruit flies and other insects could “revolutionise” artificial intelligence systems, including self-driving cars. Their study of flies’ brains and eyes identified a “turbo boost” feature, known as high-frequency jumping, that lets insects react with remarkable speed and precision.
The researchers said the mechanism could help make robots and self-driving cars smarter and more energy efficient by using movement to gather information instead of relying on computers.
Prof Mikko Juusola, of the School of Biosciences, said: “Our findings reveal a fundamentally new way of thinking about how brains compute information.”
The University of Sheffield research found that house flies and fruit flies do not process visual information passively, as previously believed. Instead of simply watching their surroundings, insects twitch their bodies in sync with what they see.
The researchers said these small, jerky movements, including rapid eye movements called saccades, help insect brains receive clearer, faster information about their surroundings.
The study, published in Nature Communications, found that when an insect makes a sharp turn, its brain “jumps” into a higher gear, Juusola said, allowing it to focus on the most important, fast-moving information.
The researchers said this mechanism helps insects overcome physical and neural constraints that would otherwise limit perception, and supports behaviours including high-speed flight and predator avoidance.
Juusola, the study’s senior author, said: “We’ve demonstrated how even the smallest brains can solve complex problems at extraordinary speeds.”
Dr Jouni Takalo said future AI systems, especially those used in robotics, autonomous vehicles and real-time decision-making, could be revolutionised by adopting similar principles of movement-driven, adaptive information processing.
Takalo, who led development of the biophysically realistic statistical model behind the work, said: “The findings challenge traditional models of neural processing, which assume that information flows through fixed pathways with built-in delays.
“Instead, the results support a new framework where sight is a collective effort between an insect’s movement, its visual input and its brain’s response.”
Read more from BBC News.
