Can bees really do math? Researchers from RMIT University in Melbourne, Australia have discovered that bees are capable of doing basic mathematics.
In their study, the researchers found that bees can be taught to recognize colors as symbolic representations for addition and subtraction. They were then able to use this information to solve math problems presented to them by the researchers. And while addition and subtraction may come easily to us, they are actually complex operations as they require two levels of processing.
“You need to be able to hold the rules around adding and subtracting in your long-term memory, while mentally manipulating a set of given numbers in your short-term memory,” says RMIT associate professor Adrian Dyer. “On top of this, our bees also used their short-term memories to solve arithmetic problems, as they learned to recognize plus or minus as abstract concepts rather than being given visual aids.”
The experiment involved training the bees in a Y-shaped maze, where the bees received a reward of sugar water when they made a correct choice, or received a bitter quinine solution if they made an incorrect choice. Because honeybees will return to a location if it provides a good source of food, they were able to keep going back to the experimental set-up to collect nutrition and continue learning.
When a bee entered the maze, they would see a set of elements between 1 to 5 shapes. These shapes would either be blue (meaning the bee had to add) or yellow (meaning the bee had to subtract). The bee would then fly into a decision chamber, where it could choose to fly to the left or right side of the maze. One side had the correct solution of either plus or minus one, while the other side had the incorrect solution. After more than 100 learning trials, the bees learned that blue meant +1 and yellow meant -1. They then were able to apply these rules to new numbers.
“Our findings suggest that advanced numerical cognition may be found much more widely in nature among non-human animals than previously suspected,” says Dyer. “If maths doesn’t require a massive brain, there might also be new ways for us to incorporate interactions of both long-term rules and working memory into designs to improve rapid AI learning of new problems.”
Image Credit: RMIT University