October 30, 2020

Can insects learn math? | Science

Can insects learn math? | Science

Magazine Science Advances today publishes the results of an investigation that shows that European bees are able to use symbolic representations to solve simple mathematical operations such as addition and subtraction.

Despite having a brain with just under a million neurons, these insects are able to solve complex problems such as the understanding of the concept of zero.

Bees are a great model to research in neuroscience. That is why in our most recent experiment we decided to test the ability of this insect to solve simple arithmetic operations.

Addition and subtraction

As children we learn that the symbol plus (+) means that we must add two or more quantities, while the symbol less (-) indicates that we must subtract them. To solve these types of problems, we need to use our short-term and long-term memory. Once we have learned and stored the rules of addition and subtraction in our long-term memory, short-term memory allows us to manage the quantities involved in each operation we perform.

Although it is not easy to solve arithmetic operations such as adding and subtracting, this ability has been fundamental for human societies. For example, archaeological evidence shows that Egyptians and Babylonians already used arithmetic operations around the year 2000 BC.. These would have been useful for counting cattle and recalculating the number of heads remaining after a sale.

The scene (copied by the Egyptologist Lepsius) shows a group of scribes counting heads of cattle. In the second row we can see 835 heads on the left, and around 2235 goats on the right. In the lower row you can see 760 donkeys on the left and 974 goats on the right. / Wikicommons

One might ask if the development of arithmetic thinking requires a brain as large as that which primates possess or if other animals are also capable of performing arithmetic operations when faced with similar problems. We have tried to answer that question using the European bee as a model.

How do you train a bee?

Bees are central foragers, which means that a worker will return to the same place as long as there is a source of food. We used a concentrated sugar solution throughout the experiment to make sure they returned continuously.

Each time a bee chose the correct number (see details below) it received a reward of sugar water. On the contrary, if the bee made a wrong choice, it received a punishment in the form of a bitter solution made of quinine.

In our experiment we used this simple method to teach bees to add or subtract during sessions of four to seven hours per individual. Each time a bee was satisfied with the sugar solution, it returned to the hive and later returned to continue training.

Addition and subtraction in bees

The bees were trained individually with a Y-shaped labyrinth (see figure below). During the experiment, insects flew to the entrance, where they saw a group of between one and five elements. These represented geometric shapes (for example, a square) that could be of two different colors depending on the arithmetic operation to train: blue to add an element (+ 1) or yellow to subtract it (- 1).

Y-shaped labyrinth used to train bees. / Scarlett Howard

From this information, the bee had to fly through a hole into the decision chamber, where he had to choose one of the arms of the labyrinth. One arm contained one more element seen in the entrance and the other one less. In each round of the experiment the animal had to choose the arm to which it should fly according to the initial number of objects seen at the entrance and the operation for which it was being trained.

The position of the correct group was changed randomly in the labyrinth, to prevent the bees from relating one of the sides with the correct answer. At the beginning of the experiment, the bees randomly chose the arm of the labyrinth to which they were going. After one hundred trials they learned to choose the correct solution to the problem.

At the end of the training, we presented the bees with a number of elements they had never seen before. The bees chose the correct answer between 64 and 72% of the time. This proportion is statistically different from what would be expected if the answers had been chosen at random. Therefore, the insects of our school of bees They learned to use arithmetic operators to add and subtract.

Why is this difficult for bees?

Adding and subtracting is difficult because it requires processing the information to two levels and simultaneously. On the one hand, the understanding of numerical attributes; on the other, its manipulation to operate.

The bees also used their short-term memory. Since the number 1 to be added or subtracted was not present when performing the operation, the insects had to learn this abstract concept during the training.

By demonstrating that bees can combine simple arithmetic and symbolic learning, we identify new areas of knowledge to explore, such as, for example, the ability of other animals to add or subtract.

Artificial intelligence and neurobiology

Artificial intelligence and how computers can learn to solve new problems autonomously are two topics of great interest today.

Our findings show that small brains are able to learn symbolic arithmetic operators, suggesting new ways to incorporate short-term and long-term memory interactions, thus reducing learning times in artificial intelligence systems.

Also, our results show that there are several ways to understand mathematical symbols as a language of operators. This could help to understand how different cultures independently developed numerical skills.

This article has been published simultaneously in English.The Conversation

Scarlett Howard, PhD candidate, RMIT University; Adrian Dyer, Associate Professor, RMIT University Y Jair Garcia, Research fellow, RMIT University

This article was originally published in The Conversation. read the original.

The Conversation


Source link