The Google of the future might be inspired by the fruit fly. Scientists at the Salk University and The University of California San Diego have learned that the fruit fly brain has evolved a masterful ability to perform searches that are similar to search engines. Flies use this ability to help them pinpoint smells that resemble odors they’ve happened on before. It allows them to prepare their response to those smells so that they understand whether they should advance towards it or fly away.
According to Science Daily, a new understanding of the way fruit flies search for smell could help develop computer algorithms in the future.
Computers perform similarity searches by narrowing the scope of information linked to each item. They add a marker called a “hash” to each item and assign a similar hash to items that have shared characteristics. When doing a search, a computer program parses through the hashes instead of the original item to get a result. It’s like a “digital shorthand.”
The scientists involved in the study found that flies and all animals are constantly performing similarity searches. So, they started digging deeper into the fruit fly’s brain mechanisms and found insight into how computer search algorithms can be improved.
— ScienceDaily (@ScienceDaily) November 10, 2017
“In the natural world, you’re not going to encounter exactly the same odor every time; there’s going to be some noise and fluctuation,” said Navlakha Saket Navlakha, assistant professor in Salk’s Integrative Biology Laboratory and lead author of the new research. “But if you smell something that you’ve previously associated with a behavior, you need to be able to identify that similarity and recall that behavior.”
As Science Daily notes, if a fruit fly recognizes that the odor of a rotting banana means it’s time to eat, it knows that it must have a similar response when it finds a similar smell. This occurs even if it has never encountered that smell before.
When a fruit fly first encounters an odor, 50 neurons spark in a unique combination that’s associated to that smell. But instead of limiting the number of markers associated with that smell, they expand the scope of the hashes. Those 50 neurons morph into 2,000 neurons which disperse the input signals so that each smell has a more complex “fingerprint.”
The fruit fly’s information processing system ensures that it identifies odor similarities more efficiently with more identity markers, which is an ability that could inform the creation of better search algorithms.
“It’s nice to see new insights coming from neuroscience being used to influence computer algorithms,” said Kristin Branson, a computational biologist at the Howard Hughes Medical Institute, in an interview with The Verge. “This is one of the reasons we’re studying the brains and doing computations at the circuit level — to try to find the better algorithms that evolution has found and incorporate them.”
Did you know that the fruit fly was able to do Google searches for smells? How do you think this research will inform the creation of better search engines in the future? Let us know your predictions in the comments below.
[Featured Image by Robian/ Getty Images]