For many years, scientists had a notion about how an animal's brain uses internal clues to determine its location in relation to its environment, much like our brains do when our eyes are closed.

Based on brain recordings from rats, the hypothesis postulates that networks of neurones known as ring attractor networks keep an internal compass that tracks your location in the world. It was believed that a big network with many neurones was necessary for an accurate internal compass, but a small network with few neurones would result in errors due to the compass's needle drifting.

 Then scientists discovered an inbuilt compass in the small fruit fly.

"The fly's compass is very accurate, but it's built from a really small network, contrary to what previous theories assumed. So, there was clearly a gap in our understanding of brain compasses." Janelia Group Leader Ann Hermundstad stated.

Marcella Noorman, a postdoc in the Hermundstad Lab at HHMI's Janelia Research Campus, has spearheaded research that solves this conundrum. The new idea demonstrates that it is possible to generate a fully accurate internal compass with a relatively small network, such as in fruit flies.

The findings alter neuroscientists' perspectives on how the brain performs a wide range of functions, including working memory, navigation, and decision-making. 

"This really expands our knowledge of what small networks can do. They actually can do a lot more complicated computations than previously known," says Noorman. 

When Noorman arrived at Janelia in 2019, she was faced with the same question Hermundstad and others had raised: how could the fruit fly's small brain develop an accurate internal compass? 

Noorman began by demonstrating that a simple network of neurones could not form a ring attractor and that "extra stuff" - such as other cell kinds and more complex biophysical features of the cells - was required to make it function. 

To accomplish this, she removed all of the "extra stuff" from existing models to see whether she could create a ring attractor with what remained. She didn't think this was possible.

However, Noorman struggled to prove her hypothesis. That was when she realised she needed to take a different strategy.

"I had to flip my mindset and think, well, maybe it's because you can generate a ring attractor with a small network," according to her, "and then figure out what specific conditions the network has to satisfy to make that happen." 

 

Study’s revelations


By altering her assumption, Noorman showed that it is possible to construct a ring attractor using as little as four neurones, as long as the connections between them are carefully regulated.

Noorman collaborated with other Janelia experts to test the novel theory in the lab, revealing physiological evidence that the fly brain may produce a ring attractor.

"Smaller networks and smaller brains can perform more complicated computations than we previously thought," according to Noorman. "But, to do so, the neurones have to be connected much more precisely than they would otherwise need to be in a larger brain where you can use a lot of neurones to perform the same computation."

"So there's a trade-off between how many neurones you use for this computation and how carefully you have to connect them," she explains.

Next, researchers intend to investigate whether the "extra stuff" may improve the ring attractor network's robustness, as well as whether the base calculation can be used as a building block for more complex computations in larger networks with multiple variables. Additional tests could also help researchers understand how neurones in the network are connected and how sensory stimuli influence the network's perception of head direction. 

Noorman, a mathematician turned neuroscientist, has found it difficult yet enjoyable to figure out how to transform biology into a solvable math problem.

"The fly's head direction system is the first example of neural activity that I'd ever seen, so it's been fun to actually figure out and understand how that works," she adds.