Ladder-walking locusts show big brains aren't always best

Dec 24, 2009

Scientists have shown for the first time that insects, like mammals, use vision rather than touch to find footholds. They made the discovery thanks to high-speed video cameras - technology the BBC uses to capture its stunning wildlife footage - that they used to film desert locusts stepping along the rungs of a miniature ladder.

The study sheds new light on insects' ability to perform complex tasks, such as visually-guided limb control, usually associated with mammals.

According to lead author Dr Jeremy Niven of the University of Cambridge: "This is another example of insects performing a behaviour we previously thought was restricted to relatively big-brained animals with sophisticated motor control such as humans, monkeys or octopuses."

Because insects such as bees and flies spend a lot of time flying, most research has concentrated on how insects use vision during flight. Many insects that spend a lot of time walking, such as stick insects, crickets and cockroaches have relatively small eyes and use long antennae to 'feel' their way through the environment.

spend time both walking and flying, and have short antennae and large eyes, which made Niven wonder whether they used vision to find footholds.

To answer this question, the team built a miniature locust-sized ladder and filmed the locusts walking along it. They counted the number of times the locusts missed steps, comparing the number of mistakes they made in different situations.

This video is not supported by your browser at this time.
Credit: Jeremy Niven, University of Cambridge

"By combining all these different experiments, we showed that locusts use vision to place their legs. We showed that when locusts can't see one front leg they stop using that leg to reach to the next ladder rung, favouring the leg they can see," Niven explains.

"Big-brained mammals have more neurons in their visual systems than a locust has in its entire , so our results show that small brains can perform complex tasks. Insects show us how different animals have evolved totally different strategies for doing similar tasks," he says.

As well as illustrating how insects can achieve similar results to mammals by using simpler mechanisms, the findings deepen our understanding of locusts' neural circuits.

This is important because locusts have been a model organism for studying limb control for the past 40 years. Insects such as the locust have been crucial to many breakthroughs in neuroscience, and are often the inspiration for limb control in robotics.

Explore further: Research helps steer mites from bees

More information: 'Visual targeting of forelimbs in ladder-walking locutsts',Jeremy E. Niven et al, Current Biology, 24 December 2009.

add to favorites email to friend print save as pdf

Related Stories

Why locusts abandon a solitary life for the swarm

Dec 18, 2008

By applying an old theory that has been used to explain water flow through soil and the spread of forest fires, researchers may have an answer to a perplexing ecological and evolutionary problem: why locusts switch from an ...

Locust breeding expected to rise

Dec 30, 2005

Locust breeding is expected to accelerate in the months ahead, the U.N. Food and Agricultural Organization said in a bulletin Friday.

Recommended for you

Research helps steer mites from bees

14 hours ago

A Simon Fraser University chemistry professor has found a way to sway mites from their damaging effects on bees that care and feed the all-important queen bee.

Bird brains more precise than humans'

14 hours ago

(Phys.org) —Birds have been found to display superior judgement of their body width compared to humans, in research to help design autonomous aircraft navigation systems.

User comments : 1

Adjust slider to filter visible comments by rank

Display comments: newest first

zevkirsh
not rated yet Dec 25, 2009
yea, insects reproduce by the millions each day, if their brain get it right 98% of the time, the species thrives. primates reproduce maybe 1nce a year. if we make a mistake 2% of the time, the species goes extinct. plus,
there's a whole lot less brain power needed to balance a body weighing a fractionth of an ounce with muscle's having unparraleled efficiency and power to weight ratio.

human's weight many thousands of times the weight of an insect with relatively weak and useless muscles in comparison.

point being, its way easier for a neural net to be 'good enough' when the hardwear (body, skeleton , muscle, lungs, vasculuture) is extremely fault tolerant and powerful.

thinking that 'copying' an insects brain and putting it to use in the AI of a robot weighing half a pound and using 'muscle' actuators that aren't nearly as responsive and dense as actual insect muscle, may very well be useful, but we should not forget to see that we are not creating a true model brain copy .