Skip to main content

How Fruit Flies Teach Us to See: Understanding Neuronal Networks for Stable Vision in Changing Light

Published on November 2, 2024, 1:24 p.m.
How Fruit Flies Teach Us to See: Understanding Neuronal Networks for Stable Vision in Changing Light

Have you ever noticed how your eyes quickly adjust when you move from a shadowy area to bright sunlight, like driving through a forest? It turns out, there’s a fascinating science behind this rapid adaptation. Researchers at Johannes Gutenberg University Mainz (JGU) have unveiled the neuronal networks and mechanisms that enable animals to perceive contrasts reliably, even as light levels fluctuate dramatically.

Professor Marion Silies, who led the study, explains, "In these situations, it’s not enough for the photoreceptors in our eyes to adapt; we also need an additional corrective mechanism." Her team previously identified a corrective mechanism known as 'gain control' in the fruit fly, *Drosophila melanogaster*, which operates just downstream of the photoreceptors. Their latest research, published in *Nature Communications*, details the algorithms and neuronal circuits that allow these flies to maintain stable visual processing despite rapid changes in light.

So, why is stable visual processing so crucial? Our vision must function accurately in a variety of contexts—whether we are navigating through our environment or tracking an object that moves from light to shadow. This ability is essential not only for humans but also for thousands of animal species that depend on sight for navigation. Interestingly, self-driving cars face similar challenges with fluctuating light conditions, often relying on radar or lidar technology to maintain accurate contrast detection. Silies and her team sought to understand how animals achieve this naturally, without any technological aids.

Combining theoretical and experimental methods, the researchers studied the compound eyes of fruit flies, which are composed of about 800 individual units known as ommatidia. The contrast between an object and its background is determined in the visual circuitry behind the photoreceptors. When lighting conditions change suddenly—like when an object moves into the shadow of a tree—contrast responses can vary dramatically. Without gain control, these differences would disrupt subsequent stages of visual processing, causing the object to appear differently than it should.

In the study, led by Dr. Burak Gür, the team utilized two-photon microscopy to pinpoint where stable contrast responses are first generated within the visual circuitry. They identified specific neuronal cell types located just two synapses away from the photoreceptors. These cells only respond locally to visual information. To accurately compute contrast, background luminance information requires narrow spatial pooling, a concept further explored through computational modeling by co-author Dr. Luisa Ramirez.

Silies shares, "We started with a theoretical approach that predicted an optimal radius for capturing background luminance across specific regions in visual space, while simultaneously searching for a cell type that could facilitate this process." Eventually, they discovered a cell type, designated Dm12, that pools luminance signals over a defined radius, effectively correcting the contrast response even in rapidly changing light conditions.

In conclusion, the Mainz-based neuroscientists have unraveled the algorithms, circuits, and molecular mechanisms that allow vision to remain stable amidst rapid luminance shifts. Silies, who has dedicated over 15 years to studying the visual system of fruit flies, suggests that similar mechanisms for luminance gain control likely exist in mammals, including humans, given the availability of the necessary neuronal substrates.


Source: Johannes Gutenberg Universitaet Mainz

Journal Reference:

  • Burak Gür, Luisa Ramirez, Jacqueline Cornean, Freya Thurn, Sebastian Molina-Obando, Giordano Ramos-Traslosheros, Marion Silies. Neural pathways and computations that achieve stable contrast processing tuned to natural scenes. Nature Communications, 2024; 15 (1) DOI: 10.1038/s41467-024-52724-5

Leave a Comment

Name
Email
Body
... ...

Get Exclusive Insights

with Every Issue

JoinShalyamNewsletter

Stay ahead in education, research, and innovation—straight to your inbox.