Male houseflies chasing each other in midair can track a target with no measurable delay, suggesting their brains predict where the opponent will go rather than reacting to where it is. This matters because even the fastest engineered tracking systems suffer from sensorimotor delays—the time it takes to sense a movement, process it, and respond. In aerial combat between flies, both participants must switch instantly between chasing and evading, pushing their nervous systems to the limit. The researcher wants to understand what visual information the fly uses to make these split-second predictions, and how individual differences in sensory or movement biases determine who wins or loses. The research is fundamentally curiosity-driven. The fly brain is the size of a poppy seed and uses extremely little energy, yet it performs predictive tracking that current robotics cannot match. If the rules behind this prediction-feedback system can be extracted, they could inform neuromorphic computing and edge computing applications—systems that process data locally rather than in the cloud. Past fundamental research on insect vision, for example, directly inspired the motion detection algorithms used in modern drones and self-driving cars.
View original technical description
Aerial contests between male insects fighting for territory and mates demand rapid reactions and sensorimotor coordination. During competitions between symmetrically capable opponents, we should expect to observe insects pushed to their flight performance limits. Properties like flight speed and turn rate are regularly in direct opposition, influencing the most viable steering strategies. A pursuer needs to be able to rapidly adapt its flight course to counter the manoeuvres of evaders, relying on short sensorimotor delays and high manoeuvrability. I aim to use this behavioural paradigm to examine how animals can minimise response delays and how aerial mobility is constrained. Uniquely in dogfighting, both participants must be capable of chasing and evading. This swapping of roles should lead to novel manoeuvres adapted to take advantage of sensory or movement biases. I intend to establish the territorial housefly Fannia canicularis as a pliable laboratory-based model species to investigate how individual differences mediate success or defeat during aerial combat. I will also investigate the visual information that flies use to minimise the delay in their reactions to the movement of their opponent. My preliminary recordings of F. canicularis show that the fly body tracks its target without apparent sensorimotor delay. This pattern suggests that the fly system can conduct tracking behaviour predictively to reduce response delay. This prediction-feedback tracking performance rewrites how we understand pursuit behaviour and has relevance to edge computing applications and robotics, but we don’t yet know what information they use to make these predictions. The brain of a fly is approximately the size of a poppy seed and has an extremely low power requirement, but it is exquisitely tuned to produce fast and accurate behaviour. Understanding how flies process sensory stimuli into predictive responses may reveal simple rules that can be adapted to improve visual feedback in engineered systems and inform neuromorphic approaches to computing.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know