PUBLISHED
The model organism Caenorhabditis elegans doesn’t have anything we would view as a brain. In fact, the nematode roundworm has just 302 neurons with which to process information, so efficiency is essential. Using just two of these, C. elegans has evolved a method of tracking the smell of food to its source that is remarkably efficient, not just in brain power but in time and energy. The newly revealed explanation of how this is done could have applications for miniature robots with limited processing capacity.
The rest of this article is behind a paywall. Please sign in or subscribe to access the full content.Our eyes can deceive us, but not nearly as often as our noses. To use smell to find food, animals follow a gradient of scent chemical concentrations in their environment. Given the potential for confusing factors, however, they will often start off on a path that sends them past their goal, rather than taking them directly to it.
Single-celled organisms’ solution to this problem is what mathematicians call the biased-random walk strategy. They will punctuate a journey by turning randomly to see if the scent is stronger in a particular direction, adjusting the rate of turning based on the chemical concentrations.
Dr Alon Zaslaver of the Hebrew University of Jerusalem and his graduate students note in Nature Communications that multicellular creatures have devised more sophisticated methods. True to its name, they found C. elegans' solution is particularly elegant.
C. elegans is known to squirm in the direction of a likely food source before either “pirouetting” or making gradual curvature corrections to its motion. However, before Zaslaver's work no one understood how the worm knew when to change course.
Zaslaver showed that one of C. elegans' precious neurons senses the amount of food scent, driving the worm towards its goal. However, when its direction is imperfect, a second neuron, which Zaslaver likened to the recalculating function on a navigation app, takes over.
The second neuron responds not to the intensity of the smell, but to the extent at which it is changing. In mathematical terms, it measures the magnitude of derivatives, or the steepness of gradients of a graph of smell intensity. If the derivative is negative, the scent is weakening, indicating a likely wrong direction. The more positive the derivative, the more likely it is this is the direction the worm should go.
When the team ran simulations comparing the two strategies, the worm's approach always beat the biased-random walk strategy. Zaslaver found that without either neuron, the worms could still find food but they take less direct routes.
Useful as the discovery could be for designing self-directing machines, one might ask why, with our 100 billion neurons, humans are too stubborn to change course when we are clearly on the wrong path.
[H/T: Haaretz]
