By Marta Serafinko
Sept 30 (Reuters) - When a bird lunges at a butterfly in mid-flight, it may see something that is not there — the insect's wings flashing beneath may not be moving where they appear to be moving. By the time the predator realizes the illusion, the butterfly already may have escaped.
Researchers studying this natural defence have found that contrasting wing patterns can work together with wing movements to create misleading motion signals, making it harder for predators to judge a butterfly's
speed and direction during flight.
"The patterns on butterfly wings combine with their wingbeats to create optical illusions, making it look like the butterfly is moving in a different direction to its actual course," said Jolyon Troscianko, a professor at the University of Exeter in England and a leader of the study published in the journal Nature. "This is likely enough to save a butterfly from the split-second moment when a bird would try to catch it."
The idea that stripes and other bold markings can create "motion dazzle" has long been proposed by scientists. But experiments involving animals including zebras, snakes and lizards have produced mixed evidence, and many studies did not account for the movements of wings or limbs.
"Butterfly wings are flexible surfaces that clap together and fling apart, and also pivot forwards and backwards in a figure-of-eight as they fly. This movement means that a stripe across the forewings can be visibly and slowly moving down on the down-stroke," Troscianko said.
"But on the up-stroke, this stripe is rapidly folded away and twisted in line with upwards flight direction. Together these make the downwards motion of the butterfly much more visible than its upwards motion, even though it is flying up," Troscianko added.
To investigate the idea, the researchers combined high-speed filming of real butterflies, computer modelling, simulated evolution and touchscreen experiments.
"Stripes are particularly good at interfering with the circuits that detect motion," Troscianko said. "The barber pole illusion is a good example — stripes moving sideways can look like they're moving up or down instead."
The patterns also appear to increase uncertainty about a butterfly's position during an attack.
"This increased 'confusion' effect would be almost impossible to overcome without evolving a fundamentally different visual system," Troscianko said, suggesting it may have been pivotal to the success of the butterflies as a group.
Not all butterflies appear to use this strategy, however.
"The least motion-confusing butterflies are typically brown or red," Troscianko said, suggesting that some species may rely more on camouflage or on warning colours that may signal to predators that the insect is toxic, poisonous or has a bad taste.
To test whether the effects could influence attacks, the researchers recruited 100 human volunteers and carried out 3,000 virtual butterfly-catching experiments. Participants watched videos and attempted to catch butterflies by tapping a screen. Butterflies that were predicted to produce stronger visual confusion caused people to aim farther behind them.
"Predation experiments with live birds and butterflies would have been ideal in many ways, but likely impossible and unethical in practice," Troscianko said.
Because the neural circuits involved in motion vision are quite fundamental in vertebrates, Troscianko said, the researchers hope the findings can translate between different species.
Butterflies face multiple predators, and motion confusion may not fool them all equally.
"For now, we've only looked at bird vision," Troscianko said. "In principle, similar effects could happen in predatory insects like dragonflies and robber flies. However, they have much faster vision than birds, so the illusions might not work for them. Indeed, dragonflies and robber flies do successfully hunt butterflies quite often."
Beyond biology, the team's motion-vision tools could have practical applications, from improving how movement is displayed in flight or combat simulators to developing patterning that protects personnel, vehicles, drones or robots in motion when being targeted visually.
"I'm most excited to explore how useful our models could be for improving road safety and reducing discomfort and stress in the built environment, where flashing, moving and striped objects can cause harm or danger," Troscianko said.
(Reporting by Marta Serafinko in Gdansk; Editing by Will Dunham)
