- Researchers simulate a miniature brain to analyze how bees count things.
- They found that bees can count up to 5 items with just four brain cells.
- The study shows that intelligence doesn’t necessarily require large-size brains.
Many experiments demonstrate that animals possess a similar “number sense”. Not only large-brained animals or birds, but also frogs, fish, and even insects with tiny brains were shown to be able to make decisions on the basis of numbers. In particular, bees exhibit incredible counting-like abilities.
While bees have relatively simple brains compared to human brains, they can perform complex learning tasks like remembering the best locations to collect pollen and nectar. In the earlier studies, researchers were able to teach bees to identify several different colors.
Recently, a research team at the Queen Mary University of London simulated a simple miniature brain to understand how bees count things. They found that bees can count small quantities with just four brain cells.
How Bee’s Brain Counts Number of Items?
The human brain takes a quick look at all the items (small quantity) and counts them together. However, a bee’s brain functions in a different way: to count a small number of items, it inspects one item closely, and then it inspects the next one closely and so on.
This type of behavior makes a complex counting task much easier, enabling bees to exhibit magistic cognitive abilities with minimum brainpower.
Earlier researches have proved that bees can count up to 5 items and can select the larger or the smaller number from a set. They can even choose 0 (none) from a set, if trained to select ‘less’.
Actually, bees don’t understand numerical concepts: they use certain flight movements to inspect items closely. This shapes their visual input and makes it easier to perform a task (i.e. count the number of items) using minimal brainpower.
Courtesy of researchers
The study shows that the intelligence of insects and small animals could be mediated by fewer nerve cells, as long as these cells are connected together in an efficient manner.
Implications for AI
Employing such insect-inspired scanning behaviors could enhance artificial intelligence (AI) applications because the efficiency of AI systems such as autonomous robots and facial recognition or pattern matching apps, relies on robust, computationally inexpensive algorithms.
The human brain contains 86 billion nerve cells that receive, process, and transmit information through chemical and electrical signals. In comparison, bees contain only 1 million nerve cells, thus they have little brainpower. To solve tasks, a bee’s brain has to implement computational algorithm very efficiently.
Researchers modeled the input to the simulated brain by examining the bee’s point of view as it flies close to items and inspects each item at a time. As per the report, the simulated brain accurately determined the number of items on display.
Overall, the study suggests that intelligence doesn’t necessarily require large-size brains, but can be achieved with small neural networks embedded in the microprocessor that is the bee brain.