Tripedalia cystophora lives in the mangroves of the Caribbean. It has only 1,000 neurons capable of activating simultaneously. And yet, according to researchers from the University of Kiel in Germany, it is capable of learning from its experience. These results, published on September 22 in the journal Current Biology, lead to a much broader reconsideration of the phenomenon of learning in the animal kingdom.
“If you are an animal and you need to navigate your environment, you have to learn to associate certain cues with their consequences. Otherwise, you die without leaving any offspring,” explains Christie Sahley, a neuroscientist at Purdue University. “This is a fundamental process that does not necessarily require the presence of a highly developed brain.”
Researchers usually consider two forms of learning: the first, called “non-associative,” includes phenomena such as habituation: if you tap an animal once, it will back away, but if it sees that it doesn’t hurt, it will gradually stop moving away.
The second type of learning is called “associative.” More complex, it involves connecting certain cues present in the environment to each other. The most famous example is the famous Pavlov’s experiment: a dog who is given food every time it hears the sound of a bell ends up salivating at the sound of the bell alone. It has associated the tinkling with the imminent arrival of food.
24 eyes and 1,000 neurons
According to Ken Cheng, an animal behaviorist at Macquarie University in Australia, very few experiments to date have demonstrated associative learning in animals as simple as jellyfish. Ken Cheng has just written a commentary on this subject in the same issue of Current Biology. In 2021, when conducting a study on associative learning in Cnidarians (a zoological group that includes jellyfish, corals, or sea anemones), he reviewed all the contributions in the field and found only a handful of studies on associative learning. And they were all about sea anemones.
Why so little interest in this matter? Jan Bielecki, a neurobiologist at the University of Kiel in Germany and co-author of the new study, explains that scientists too often design their experiments as if they were testing typically human abilities, which is a mistake. “You cannot judge a fish on its ability to climb trees,” he deplores. “The parameters used must make sense to the animal. You have to meet it at its level in order to study it.”
Hence the idea of observing whether associative learning is possible in small jellyfish with four ocular structures called “rhopalia,” each containing six eyes and around 1,000 neurons (each rhopalium takes turns acting as the non-centralized nervous system of the jellyfish). To do this, the team set up an experiment that studied a reflex behavior of jellyfish for protection. Indeed, this variety lives in mangroves where it moves in sometimes murky waters and has to try to avoid the underwater roots of trees. Its eyes help it distinguish the shapes of roots, but when it touches one anyway, its umbrella automatically retracts in a reflexive protection response (the umbrella refers to the gelatinous corolla from which the jellyfish’s tentacles emerge).
The jellyfish never bumps six times
In their experiment, scientists placed the animals in tanks painted in three different ways: in the first tank, highly contrasting black and white vertical stripes representing closely spaced tree roots were drawn; in another tank, moderately contrasting gray and white vertical stripes giving the optical illusion of tree roots located well beyond the walls of the tank; and in the last tank, a plain gray background without any contrast effect.
The jellyfish crossed the tank adorned with black and white stripes without ever bumping into the edges: believing that they were seeing roots very close, they kept their distance. But, of course, it was impossible to test their ability to learn from their mistakes, since they did not make any.
On the other hand, for the visual system of jellyfish moving in tanks with gray and white stripes, the roots seemed distant; they moved too close, so they sometimes collided with the edge of the tank. However, mistake after mistake, they learned to keep their distance from the obstacle. They required between three and five collisions, on average, to acquire this new behavior.
Scientists ask for more evidence!
Does this make jellyfish “learning beings”? Not so fast, temper some voices like that of Catharine Rankin, a behavioral neuroscientist at the University of British Columbia. She would like to see additional tests conducted to better understand what these animals are actually doing and assess their true capabilities. “I would like to see them demonstrate ‘extinction,'” she says. In other words, if a jellyfish sees a visual signal associated with the edge of the tank, but the obstacle is removed and it no longer bumps into it, will it gradually stop avoiding this obstacle? That is the question.”
Christie Sahley, who has studied learning in many species that are more “primitive” than jellyfish, is also curious about how long the jellyfish would be able to remember the association between the gray stripes and the risk of impact with the edge.
Despite these legitimate expectations, scientists generally agree that this new study provides valuable information on the general mechanisms of learning in a wide variety of animal species. These relatively rudimentary organisms offer an interesting opportunity to study the internal molecular mechanisms of neurons that explain these adaptive capabilities. It is certainly much easier to do so than with human or mouse brains, which contain vastly greater quantities of neurons with far more complicated reciprocal interactions.
“We know, in any case, that it is not necessary to possess a hippocampus or cerebral cortex to learn,” explains Ken Cheng. “Even though they lack these nervous organs, these animals are still capable of learning. So now we need to see what happens with even simpler beings. Even with simple cells.”
#Les #méduses #comment #apprendre #sans #cerveau
publish_date] pt]
