Do Plants Practice Grid Computing?

Roland Piquepaille writes "According to Nature, plants appear to 'think' and seem to optimize their 'breathing' by conducting simple calculations through a distributed computing scheme. "David Peak and co-workers at Utah State University in Logan say that plants may regulate their uptake and loss of gases by 'distributed computation' -- a kind of information processing that involves communication between many interacting units." Nature adds this is similar to signals exchanged by ants to find the best source of food for an ant community. In their paper, the researchers added that their results were "consistent with the proposition that a plant solves its optimal gas exchange problem through an emergent, distributed computation performed by its leaves." This overview contains more details and references. It also includes a picture of the tiny pores on the surface of a cactus leaf, called stomata, which permit the plant to breathe when they're opened."

Comments

Um, ok

[skintigh2]

So, plant behavior kinda sorta looks similar to what a distributed computing system might look like, therefore plants are distributed computers?

"I saw a picture of a Mars rock that looked like a human face, therefore there are people living on Mars."

Or is this just a buzzword-filled way to say the obvious: there is no central brain in a tree; each leaf controls it's own pores and uses chemical signals from surrounding pores and leaves for help. We already knew that trees "communicate" with each other on when it is time to start changing color. Perhaps I should write up that old news and drop in some buzzwords. I can title it "Trees form Beowulf Clusters to incentivize the diversification and downsizing of foliage."

Re:Um, ok

[frobber]

I agree - sounds like the obvious dressed up in trendy jargon.

From the article it looks like they're trying to understand how a plant knows how many total pores to have open for breathing, given that each pore only has local information - there's no global sensor telling the pores what to do. They're also interested in why open pores are found in clusters.

A simple answer might involve the following:
(1) the pores are simple oscillators locally linked causing local synchrony. Groups of pores tend to be in sync, and the neighbors of the open group are induced to open because they're near the group. Therefore the pores tend to open near the open group, and close in the middle of the group. The result is travelling waves like a rock thrown in a pond. This type of idea was investigated I think first by Turing (of all people) in 1952 (for example)
(2) either the intrinsic period of each pore's oscillation or it's duty cycle (how long open compared to how long off) is modulated by the pore's detection of how much the plant needs to breathe locally. The result is oscillations all over the plant of openning and closing pores whose open times are modulated to solve the plant's total breathing needs.

Anyway, I don't see what's interesting about calling this computation. Air transmits sounds by local interactions of gas particles and the speed of transmission is modulated by density. But I don't see what is gained by claiming that the air is solving a computation to transmit sound at the right speed!

Re:Um, ok

[Urkki]

• Anyway, I don't see what's interesting about calling this computation. Air transmits sounds by local interactions of gas particles and the speed of transmission is modulated by density. But I don't see what is gained by claiming that the air is solving a computation to transmit sound at the right speed!

Well, there is no right speed of sound, and speed of sound is determined very locally, by the interaction of immediatly neighbouring molecules, and stays about the same regardless of conditions a mere hundred molecules away.

If I understand correctly, the point is that a plant is able to optimize it's gas intake/output without any actual nervous system or central controlling unit, and that for this type of a problem, this might actually be the optimal way of solving the problem.

And problem it is, for plants. Do it wrong and die (either directly or by being suffocated on more successful plant neighbours). So unlike with sound propagation in air, there is an optimal way to do it, perhaps even several almost optimal ways, so a choice is involved. The plant that is able to choose best wins. (Note: I'm not implying concious choice here, any more than a "choice" made by neural network software is concious).

Re:Um, ok

[quinkin]

Heh.

I thought it should have been title "Cellular automatons successfully model yet another cellular matrix feedback system."

Q.

Re:Distributed computing plants?

[uncoveror]

Plants feel pain, communicate, and may even be sentient, yet vegetarians claim it isn't cruel to eat them. Yes it is! Here is an exerpt from an informative article:

"Maybe Disney will make a cartoon about a happy little vegetable. He will be called Buddy the Carrot. He'll lose his mother to the farmer when he picks her, and eats her. That could do to vegetables what Bambi did to meat! Carrots may in fact be more intelligent than deer. Who knows for sure?"
Read more.

Say what?

[Anonymous Coward]

Are they really sure it should be called a "distributed computing network" and not just a multi-element feedback network?

While we may be able to identify the various metabolic pathways and processes in all cells, to call them "computers" implies a certain amount of discreteness either in process pathways or elements making it up. Sure, at some level there is quantization (i.e., cellular), but one cannot identify one part and say, "this is the atmospheric pressure sensor", and "this is the hygrometric sensor".

Is the feedback system in our bodies that regulates heartrate, blood pressure and respiration a discrete computing process easily identified into its component parts? No, it's a bunch of feedback loops at various levels with a few simple inputs that produces a complex state that manifests in a few simple responses.

Re:Distributed computing plants?

[Aardpig]

Which is why, if you notice, a lot of Hindus do not eat things like Garlic, Onions, Carrots, Potatos, and the like -- anything where a plant is killed.

I'm skeptical about this assertion. Just a couple of common items of Indian cuisine would seem to disprove your case: onion bhajee and sag aloo (potato). Perhaps you are thinking of Jainism?