They form elaborate societies and efficient industrial complexes, and are probably the most accomplished dancers in the insect world. But do honeybees (Apis mellifera) have "imagination"?

It might sound like a strange question, but it's one that has dogged researchers for decades. Work published this month in Animal Behaviour challenges an old result that has often been used to support the idea.

Honeybee society is divided up into several jobs. Those bees that venture out of the hive to look for food are called foragers.

Once a promising food source has been found, foragers return to the hive and use an interpretive dance to tell their colleagues where the food can be found. The waggles and turns of the honeybee dance communicate distance and direction to the honeybee recruits.

The new experiment was designed to determine whether honeybees carry around mental maps of their environment, and use those maps to make judgments about whether a food source is likely to be where a fellow bee says it is.

PUSHING THE BOAT OUT

More than 20 years ago, Fred Dyer and James Gould, then both at Princeton University in New Jersey, trained forager bees to recognise a food source on a boat in the middle of a lake.

After the bees returned to their hive to communicate their find, Gould and Dyer took note of whether the new recruits were willing to follow directions that led them to the middle of a lake - a highly unlikely place for food to be.

The researchers found that the recruits did not take the bait, suggesting that they had compared information from the waggle dance with their own mental maps of the local terrain and decided that the waggle dance was wrong.

The study became known as the "lake experiment." Although it has never been formally published in a peer-reviewed journal, that hasn't stopped it being invoked in essays and books as evidence that honeybees have cognitive abilities that allow them to form mental maps and even develop a kind of intuition.

"Experiments suggest that recruits, having attended a dance in the hive specifying the distance and direction of a food source, can evaluate the 'plausibility' of the location without leaving the hive," wrote Gould in a 1990 review on honeybee cognition. "This suggests a kind of imagination."

SHAKE, WAGGLE AND ROLL

Now, Margaret Wray of Cornell University in Ithaca, NY, and her colleagues have repeated the lake experiment.

But they found that the recruits blithely followed foragers' directions and journeyed out of the hive and to the lake just as often as they followed directions to a more plausible location located an equal distance away. "They weren't making a distinction," says Wray.

Wray is careful to note that the new results do not overthrow the idea that honeybees keep a mental map of their surroundings. Other studies have since addressed this question using different methods, and have concluded that the mental maps exist. But the findings do challenge a result that has been perpetuated for decades.

"(The original study) wasn't actually published, and people will admit that the results were not necessarily conclusive," says Wray.

"But it just sort of took on a life of its own. People said if bees can imagine their surroundings then perhaps they have imagination and then perhaps they have consciousness."

It is unclear why Wray's results contradict the earlier findings, but Wray suggests that perhaps small environmental differences, perhaps a stronger wind, for example, discouraged the earlier bees from flying over the lake. Gould, who is conducting fieldwork in Bermuda was unavailable for comment on the paper.

SEEING IS BELIEVING

Wray's experiments used scented food sources, which raises a potential problem, notes Jurgen Tautz, a honeybee researcher at the Julius Maximilian University of Wurzburg in Germany.

Bees can smell food from kilometres away if the wind is right, whereas their ability to see the food is limited to only a few metres, he says. That means a strong scent could negate the need to follow a dance at all.

Meanwhile, Tautz has developed a different explanation for the results of the lake experiment. In 2004, he and his colleagues conducted a similar experiment as a way of testing how honeybees measure distance.

The researchers hypothesised that honeybees measure distance by visually assessing changes in the landscape as they fly.

Flying over water could interfere with this system, the authors reasoned. "The surface of the water all looks the same," says Tautz. "As long as they fly over the water, they cannot measure distances."

Tautz and his colleagues found that the bees left the hive in response to the foragers' waggle dance, but then gathered along the water's edge.

For him, Gould's results have a simple physiological explanation - an inability to correctly assess and then communicate distance to the food - and not a cognitive one.

"Bees are very clever, there is no doubt about it," says Tautz. But although he is open to the idea of a cognitive map, he finds it surprising that a bee would have one.

"Usually they are focused on the most essential aspects of daily activity," he says. "Carrying around a cognitive map is really a luxury."

Copyright 2008. All rights reserved by New York Times Syndication Sales Corp. This material may not be published, broadcast or redistributed in any manner.