If bees can spot sugary rewards at a distance, it may mean that we need to re-evaluate experiments that assess their intelligence.
Bumblebees can identify sugary liquids before they even take a sip. This ability to spot a sweet reward at a distance could mean that some of the approaches we have used to assess bee intelligence need to be re-evaluated.
“Much of what we know about insect cognition comes from work on bees,” says Tomer Czaczkes at the University of Regensburg in Germany. In many such studies, bees choose between artificial “flowers” after being trained to associate one option with a reward such as sucrose water, and the other with plain water or a bitter solution.
A bumblebee taking a sip from one of two soaked filters affixed to a toothpick Melina Kienitz |
But little was known about whether bees could figure out the difference between these at a distance. If this was the case, some bees might be using their sugar-sensing abilities to take a shortcut – which could mean they are not being trained on a particular task, but simply flying directly towards a treat.
Czaczkes and his colleagues tested this in 90 buff-tailed bumblebees (Bombus terrestris). After familiarising the bees with how to feed on sugar water from several artificial flowers, the team put each bumblebee in a clear box containing a choice between sucrose solution and plain water in three set-ups: a pair of large droplets, a pair of cigarette filters soaked in the two solutions and two upright plastic tubes containing some solution and stuffed with cotton soaked in the solutions.
In the droplet and cigarette filter tests, the bees went to the sugary solutions more often than would be expected with random chance, suggesting they could quickly tell the difference between the two. However, the bees visited the tube with the sugar solution about as often as they chose the tube with plain water. The researchers think this may be because the tubes were suspended in an opaque cup, making it harder to see their contents . This suggests the bees are picking up on visual cues that betray where the sugar is, perhaps a small difference in colour or the way the solutions distort light.
“The animals care mostly, if not only, about the food – not about our task. It is always our job as scientists to make sure that the only way they can reach the food is by solving the task,” says Massimo De Agro’ at the Sant’Anna School of Advanced Studies in Italy. Offering bees what had appeared to be identical sucrose solution and water was an attempt to do just that.
Often, researchers will first train bees with sugar and water, but conduct further tests of their abilities without either, says Czaczkes. The bees’ performance in many tasks tends to drop between training and the first test.
“This implies some of the bees were learning the wrong thing. Had this [solution] differentiation at a distance been controlled for, our estimates for how well bees do in these tests would be higher,” he says.
Czaczkes says that the team is concerned that some failed tests – which tend to not get published – would have supported more advanced brain power in bees and perhaps other insects. “We may have been underestimating how smart insects are.”
Claire Hemingway at the University of Tennessee, Knoxville, is interested in what the sucrose concentration threshold may be for the bees’ detection powers. Can bees still tell the difference when the reward is more dilute?
There needs to be more testing to know how the bees’ discrimination abilities factor into their learning, and how relevant it is in nature and across many studies, says Aimee Dunlap at the University of Missouri-St. Louis. For instance, it is unclear how the bees’ attention to a hypothetical visual cue from nectar might stack up against brighter cues such as flower colour.
Journal reference: