SCIENTIFIC EDUCATIONAL CENTER science idea

Scientists at MIT have found that our brains are not optimized to find the shortest route. Instead, he chooses the path that most accurately indicates the final direction.

For a computer, route planning is one of the most computationally complex tasks, but humans cope with it surprisingly efficiently, and on different scales. However, observations show that the routes that a person chooses systematically deviate from the shortest path between two points, and these discrepancies are still not well understood.

To find out why people don't always take the shortest route, MIT scientists, along with colleagues from France, China and Italy, analyzed 552,478 GPS tracks from 14,380 pedestrians in two major US cities, Boston and San Francisco. It turned out that pedestrians, as a rule, choose not the most direct or shortest path - but the one that most accurately points to the end point. Moreover, the more often people deviate from the shortest path, the greater the distance between the starting and ending points. Interestingly, the paths chosen are statistically significantly different when the origin and destination are reversed.

Choosing a path based on the final direction is called vector navigation, and it is known that many animals - for example, cats, rodents, bats - use this approach when planning a route. Based on the results of the analysis, the scientists created a vector navigation model that predicted the real human route better than the model based on minimizing the distance traveled with stochastic effects.

Since the results are the same for two large cities in the United States with different layouts, scientists suggest that vector navigation may be a universal property of path planning for humans. It is likely that vector navigation, which requires less effort than calculating the shortest route, allows the brain to devote more energy to other tasks. The loss in time is compensated by the gain in the computing power of the brain.

Article published in the journal Nature Computational Science

PHOTO © Fotolia / Andrea Danti
Source: naked-science.ru, sci-dig.ru