nav search
Data Center Software Security Transformation DevOps Business Personal Tech Science Emergent Tech Bootnotes BOFH

Fruit flies' brains at work: Decision-making? They use their eyes

Before you scoff, consider how much we rationalise away what's in front of us

By Andrew Silver, 6 Sep 2017

Scientists hunting for the secret of how boffin scalpel-fodder favourite Drosophila melanogaster (aka the fruit fly) makes decisions have found that some of the brain circuitry active when it makes choices can be linked to what it has already seen.

The research is being undertaken in order to some day help better understand the much larger brains of mammals (like ours).

The problem with studying mammalian brains, such as those of primates or humans, is that they contain so many different neurons – the human brain has about 100 billion – of different types. Although plenty of prior experiments have recorded brain activity of mammals (for example, by using electrodes), Hokto Kazama, a systems neuroscientist at the RIKEN Brain Science Institute in Saitama, Japan, told The Register that it's very difficult to get to the scale of large swatches of individual neuron types.

However, there are plentiful special genetic tools for tagging the fewer neurons found in fruit flies, so that they can be much more easily and precisely observed during experimentation than mammals. Short-lived, genetically modified fruit flies with clear neuron tags are also easy to breed and maintain, he says.

Previously, neuroscientists have found that D melanogaster can remember its orientation, places or landmarks it has visited as it navigates, and even associate consequences (heat) with what it has clapped its compound eyes on. In the new study, appearing this week in Nature Neuroscience, Kazama's team ran a very similar task experiment to these while simultaneously recording the activity of neuron types linked in previous research to navigation.

The team collected a few hundred specially genetically modified D melanogaster fruit flies and tethered them to a plate in front of a white screen with their heads fixed, leaving their wings free.

For two seconds, the flies would see a black vertical bar in their left or right field of vision. The bar would disappear for eight seconds, and then after that they would see two black bars – one left, one right.

Consistently, the untrained fly would flap its wings in the direction of the bar on the opposite side to the one it had first seen, suggesting that it considered the original image when making its decision (for some unknown reason), Kazama says.

The flies were also genetically modified so that the selected type of neurons in their brain would glow a fluorescent green when active. One group of these neurons consistently glowed after the fly was shown the first bar, another as it made its decision of where to "turn" when shown the two bars.

This suggests the team discovered the neural activity linked to when the fly was using the visual information from its past (bar location) to make a decision (to turn).

"All of us wonder how we make the decisions we make," Ashburn, Virginia-based Howard Hughes Medical Institute's Janelia Research Campus neuroscientist Vivek Jayaraman, who was not involved in the study but researches fruit flies, told The Register.

He says there might be structural patterns in the neural activity of fruit flies that can act as a pointer for understanding structural patterns in mammalian brains, although he cautioned that there's more work to do before determining the significance of the found activity pattern and what it does to the fly's behaviour.

Kazama says the lab's next step is to try to determine how the fly combines the two separate groups of active neurons to aid in navigation. ®

The Register - Independent news and views for the tech community. Part of Situation Publishing