What if plants could learn, remember what they learned, and teach what they learned to other organisms? Would this be considered a form of intelligence?
This behavior is what evolutionary ecologist Monica Gagliano and biologist Simon Garnier observed in their experiments with plants and even single-celled organisms.
“One of the typical comments that I would get about that study was, ‘I don’t believe you.’ And I’m like, ok, you don’t have to. But can we talk about my data?”
— Monica Gagliano, evolutionary ecologist & researcher
Monica Gagliano used mimosa plants in her study because they visibly react to physical touch. Mimosa leaves contract when the plant is disturbed in “unusual” ways as a means of protection. This behavior enables them to look smaller and less appetizing in front of a hungry plant-eater, for example.
But this adaptation comes with a cost: with contracted leaves, the mimosa plants lose about 40% of their photosynthesis capacity. This is a major decrease in the function that keeps the organism alive, particularly if there is no real reason for it (i.e. there is no real danger to the plant). The benefit does not outweigh the cost if the plant contracts its leaves when there’s no real predator preying on them.
In Gagliano’s study, she would drop the mimosa plants from a set height to elicit the visual cue that the plant is disturbed – contracted leaves. She repeated 60 consecutive drops per plant.
Half the plants were dropped (or “trained” as she puts it) in environments where there’s an abundance of light. The other half were dropped in a lower-light environment.
What she found:
- Both sets of plants “learned” that being dropped was not a real danger, demonstrated by the fact that they stopped contracting their leaves after a few drops.
- The plants that trained in a low-light environment learned faster than the plants training in light abundance, because they couldn’t afford to dawdle. Less light means less opportunity to forego photosynthesis capacity, resulting in a quicker learning curve. This fits the common knowledge that man adapts to his environment. “The capabilities of the average person could be doubled if their situation demanded it” might also go for plants?
- Both sets of plants “remembered” what they learned. After being left alone for 3 days, they did not contract their leaves when being dropped again. And then after not being touched for 6 days, once again they didn’t contract their leaves when dropped.
“But how long can they remember?” she wondered.
Gagliano left the plants alone for a full 28 days before coming back to “disturb” them. And when they were dropped from the same height again, they displayed that they “retained what they’d learned.” They didn’t contract their leaves, as if they “remembered” being dropped last month, and that it wasn’t dangerous.
You may be wondering, “Where is that memory being stored?” And it’s what the panel host at the World Science Festival asked of Simon Garnier after he described his experiments with slime mold.
Slime mold is a single-celled organism that at times links up with other slime mold organisms to fuse together and become one slime mold. It has no brain, but unlike the cells in our bodies — which each have a single nucleus — the cell of a slime mold can contain millions of nuclei.
As Garnier described, if you take a chunk of slime mold (multiple single-celled organisms fused together) that has passed through a dangerous barrier to get to food, and then mix that chunk with another chunk of slime mold that hasn’t gone through the same barrier, there will be a transmission of that memory into the inexperienced slime mold. Once fused with the experienced slime mold, the part of the organism that hasn’t been through the barrier will know the fastest, safest way to pass through it and reap the reward on the other side. In essence, this brainless goo has learned something, remembered it, and transmitted that memory to another slab of brainless goo.
Slime mold is also famous among a certain subset of scientists for solving mazes quickly, efficiently, and repeatedly. There was a brilliant experiment that demonstrated slime mold replicating the most efficient way to move through the Tokyo subway system. To repeat: A brainless goo found the fastest route through the railway of the most populated city on planet earth, in about 26 hours.
“Would you define this as ‘intelligence’?” the panel host asked Garnier.
Garnier responded: “I try not to use the word ‘intelligence.’ When I can, I use the word ‘problem-solving,’ because I can define this as a problem… The problem in this case is balancing things, like the cost of building the networks, but also the robustness of the networks. So that’s a problem, a trade-off, they need to be able to solve. The question of whether or not that is something we consider ‘intelligence,’ depends on how… We have a history of testing and measuring intelligence. We have created a battery of tests, and some of the tests are about optimizing this trade-off. So the question is now, we either have to discard all these tests because the slime mold passed it — decide that these tests don’t actually test intelligence — or we have to admit that they are intelligent. But I don’t want to pronounce on it because I want to be reviewed too.” [Everyone laughs] “But it’s essentially a question of – they passed a test that we have given to brained animals, and we have considered this a test of intelligence.”
I included the full quote to display that Garnier carefully tip-toed around the question for a reason that became revealed as the panel went on: These researchers face discreditation from the science community if they purport that plants are intelligent. Taken out of context, a statement like this could be considered too “out there” to be taken seriously by researchers.
Due to this context, it was somewhat courageous for Gagliano to reflect that plants have the same “neurotransmitters” that humans have. But because they don’t have brains, and therefore no neurons, they can’t have the “transmitters.”
Humans came up with the term “neurotransmitter,” which locked us into a certain framework.
Neurotransmitters are the messengers that carry our thoughts — not just between neurons, but also between the mind and body. In humans, without neurotransmission, we have no thought. And without brains, we have no neurons. Thus, we concluded that without brains, there can be no thought.
So what if you allowed yourself to consider the questions of whether plants can “learn” even when they don’t have brains, and if they can “think” without neurotransmitters?
The scientists never explained where the memory is stored. But to this day, after hundreds of years of research, the science is inconclusive on where human memories are stored and how they’re retrieved. Does that mean humans don’t remember? Or have we simply not learned enough about our own complex neuroscience yet?
Gagliano reminded us that memory is an intrinsic part of learning. If one is to “learn” a useful fact and then to forget it the subsequent day, was it really learned? Being able to demonstrate in practice what one has learned in theory may be the ultimate evidence that one has truly learned.
The slime mold effortlessly and repeatedly moves through the maze in the most efficient way. The mimosa plant keeps its leaves fanned out after being disturbed on iteration after iteration. What other examples of non-animal “learning” and “memory” are occurring throughout the biological world that we’ve never observed?
To finish Gagliano’s quote that opened this essay: “Are you a philosopher, or are we doing science? Because if we’re talking about philosophy, then yeah, maybe you’re right. Depending on what your vision of learning and memory is, then maybe there is something there to discuss. But if we are just talking about the science of… Here is my experiment, here is my data, let’s talk about that.”
I’m reminded of Ludwig Wittgenstein’s idea that “philosophy is just a byproduct of misunderstanding language.” These researchers record behavior in plants and single-celled organisms which looks a lot like learning and remembering, but observers may counter with, “That’s not intelligence.” In a technical, linguistic sense, these observers are right: The chemical reactions in plants (that exactly match chemical reactions in humans) – which represent “thinking” – can’t be constituted as “thoughts” because the existence of a thought infers the existence of a neuron. And plants do not possess neurons.
But as Gagliano pointed out, the language we come up with locks us into certain frameworks. Perhaps if we never named them “neurotransmitters” after all, we wouldn’t be debating with each other about whether or not plants can think, and whether they are intelligent. Maybe the behavior they display would be sufficient to conclude plants and single-celled organisms do possess a form of intelligence. If only we had not linguistically locked ourselves into a certain understanding of intelligence.
So perhaps the interpretation of what these organisms have demonstrated is merely a debate of semantics.
Thanks to Claud Richmond for the photo.
