Skip links
Published on: Press

No crash for the fly.

The return of spring is accompanied by a parade of insects, annoying mosquitoes, irritating flies, and buzzing wasps, which we often crush without hesitation. Let’s make a bet that by the end of this article your attitude towards these insects will have evolved.

What if we started looking at these winged insects from a different point of view?

Admittedly they bother us, but some will point out that they play a role in the ecological balance. As nourishment, they feed migratory birds as they pass through the tundra. Without them, 50% of birds would disappear. As carriers, wasps help winemakers, as their stomach is perfectly suited to the development of saccharomyces cerevisiae, which allows for the dispersion of yeast to make the wine. (Stefanini, Dapporto, Legras, 2012) Just like the fly who, with its small size, is the only pollinator of cocoa, whose flower that smells like a mushroom, attracts it irresistibly! No fly, no chocolate.

But we can also see winged insects as agile aircrafts. They carry their millions of neurons, autonomous neural and optronic pathways, equipped with micro-sensors and micro-actuators, that solve general problems of sensorimotor control. What a source of inspiration for autonomous vehicles, cars or drones!  Not to mention the ridiculous amount of energy consumed by these insects to fly, detect food, avoid obstacles, and mate in flight.

We discussed in a previous article that nature has no intention. Let us now add a radical definition of what a nervous system is:  a signal-processing machine. And it is a machine that can be modelled.

This immediately brings us two advantages: irrespective of the molecules that make up the machine and irrespective of the ones which transmit the signals. This is the pillar of cybernetic thinking laid out by Norbert Wiener in 1948  (Wiener, 1948), which today has a formidable device to simulate models: the computer.

In summary, cybernetic thinking opens the door to modelling the nervous system as a signal processing machine and the computer allows us to simulate it. You do not need life for that.

The consequence is simple: if the nervous system is the place where the learning of living beings occurs and intelligence is built, then there is no obstacle to simulate it and to therefore create an artificial nervous system. It is irrelevant whether a computer is made of silicone, optic or even quantum. It is the approach, the promise and the hope of artificial intelligence.

Let us be inspired by biology and ask the fly to give us its manufacturing secrets.

How does it fly safely, even with unfavorable and changing winds? How does it reach out to its partner to mate during flight? And to land precisely where it decides to? Ask yourself, as an engineer. How would you design a device that measures neither its speed, its altitude, ground speed or rate of descent, that has neither GPS, nor altimeter or variometer?

Nicolas Franceschini in Marseille runs the laboratory that has been working to understand the fly’s eye since 1985. (Franceschini, 2008) He teaches us that each fly has two panoramic eyes composed of about 3,000 tiny facets that act as small lenses. The cone they form is like a well at the bottom which is composed of exactly eight receptor neurons, with two that overlap. Look closely at the illustrative image. This is not a painting by Yves Klein. It is a section of a fly’s eye.  You can distinguish seven points in each “pentagon”:  six neurons plus the two overlapping neurons.

Each eye is therefore composed of 48,000 photoelectric micro-receptors. The six neurons pick up six independent signals that will be used for the vision of movement. The two overlapping neurons will be used for color vision by an extremely subtle mechanism that allows the fly to orient itself even when the sun is hidden.

This team discovered that the secret of the fly’s vision lies in the ability of this insect to focus on a vision of movement of various elements of the landscape during its flight. Very long-distance objects pass at a low angular speed, while nearby objects pass at a very high speed. The fly therefore only analyzes the “optical flow.” It only distinguishes the difference in contrast between objects, and analyzes their distance through their passing speed. This is enough for everything it has to do. Nature has selected an extremely frugal device, very light, that consumes little energy to keep the fly alive.

But if we stick to the principle that there was no plan to build this mechanism, what path did nature take to get here? Is there auto-organization in the fly’s brain? The positive response was provided in 2019 by a Japanese team (Trush, Liu, Han, 2019). The brain of the fruit fly reveals a column structure that auto-organizes itself and provides the fly with its excellent ability to analyze the visual scene. Would you be surprised to learn that this organization is very similar to that of mammals?

We will talk about this in a future article that will shed light on this central mechanism of cognition.

So, will you keep crushing the flies?

Works cited

Durhuus, B., And Eilers, S. (2014). S.A. On the entropy of LEGO. I’m J. Appl. Math. Comput., 45, 433-448.

Franceschini, N. (2008). Color and movement in the life of an insect. CNRS Interdisciplinary Thematic School, (pp. 56-80). Mr. Roussillon.

Stefanini, I., Dapporto, L., & Legras, J.-L. (2012, August). Saccharomyces cerevisiae and social wasps. Proceedings of the National Academy of Sciences, 109(33), 13398-13403.

Trush, O., Liu, C., Han, X. (2019). N-Cadherin Orchestrates Self-Organization of Neurons within a Columnar Unit in the Drosophila Medulla. The Journal of Neuroscience: the official journal of the Society for Neuroscience, 39(30), 5861-5880.

Wermelinger, B., Duggelin, C., Freitag, A., Fitzpatrick, B., and Risch, A.C. (2019, July). Ants of the woods – biology and distribution in Switzerland. Notice for the Practitioner, WSL Federal Research Institute(63).

Wiener, N. (1948). Cybernetics, or, Control and Communication in the Animal and the Machine. New York: John Wiley and Sons, Inc.