Can Plants Learn to Associate Stimuli with Reward?

A group of pea plants has displayed a sensitivity to environmental cues that resembles associative learning in animals.

By Ben Andrew Henry | February 1, 2017


In 2007, plant biologists passionately argued the meaning of the word “neurobiology.” The year before, an article published in Trends in Plant Science had announced the debut of a new scientific field: plant neurobiology. The authors suggested that electrical potentials and hormone transport in plants bore similarities to animal neuronal signaling, an idea that raised the hackles of many a botanist. Thirty-six plant scientists signed a letter briskly dismissing the new field, calling the comparison between plant signaling—intricate though it is—and animal signaling intellectually reckless. “Plant neurobiology,” they wrote, was no more than a “catch-phrase.”

Upon close examination, the “neurobiology” debate did not center on very much scientific disagreement. Researchers in both camps agreed on the general facts: plants did not have neurons, nor did they have brains, but they did possess complicated, poorly understood means of responding to the environment that deserved rigorous study. The community was conflicted over how to talk about these abilities and whether the semantic umbrella of words such as “feel,” “choose,” and “intelligence” should extend to plants.

The rhetoric surrounding the argument has since cooled, but the debate was never entirely resolved. And in 2016, Monica Gagliano of the University of Western Australia and colleagues provided fresh fuel for the conceptual fire. The researchers conducted an experiment that they say shows plants performing associative learning (Sci Rep, 6:38427). This type of learning is the same process by which a dog can learn to associate the sound of a bell, as in Pavlov’s famous study, with a treat, and then salivate with anticipation every time a bell rings. The group’s experiment, modeled on Pavlov’s, was designed to subject pea seedlings to analogous stimuli and find out what the plants could learn.

The seedlings were grown in Y-shape tubes for about a week, receiving eight hours of light a day. Then, they were enrolled in a three-day training course. The grow lights were turned off, and three times each day, a small fan blew a light breeze down one arm of the Y-tube for an hour and a half. Beginning after the first hour of that period, the plants were given a one-hour dose of blue light (overlapping for one-half hour with the fan), their only sustenance during otherwise lightless days. For some, the breeze and the light came down the same arm of the Y. For others, the two stimuli came from opposite directions. In both groups, the stimuli were switched randomly from left to right between sessions, and a pilot experiment showed that the breeze had no influence over growth on its own. For all intents and purposes, the breeze was only meaningful to the plants insofar as it predicted where light would soon appear, the authors reasoned.

Unless we really explore the field experimentally, then we are just plant philosophers—
and there are already plenty of good philosophers around.—Monica Gagliano,
University of Western Australia

On the fourth day, seedlings had approached to within a centimeter or so of the bifurcation in the Y, and they were kept in the dark that day. A control group was left undisturbed, while a test group got the usual three courses of gentle breeze, but this time without the accompanying light. The fan breeze was applied in a direction that “predicted” light to appear opposite the side where it had last appeared.

The seedlings continued to grow, never bumping against the fork of the Y-tube but bending left or right. In doing so they made a choice, so to speak, to grow in the direction of their own survival. Of the 19 plants in the control group, 100 percent extended in whichever direction the light had last come from, exhibiting the well-known affinity of young plants for blue light. The 26 plants in the test group, however, had a decision to make. They could persist in the most recent direction of the blue light like the control group, or grow in the opposite direction, where the fan predicted light should appear—that is, show that they had “learned” something about the meaning of the breeze. Around 65 percent chose this latter option.

“This is exactly what Pavlov did,” Gagliano says. “If this were an animal of any kind,” rather than a pea plant, “this would be considered learning.” However, the experiment is notably limited in scope. A crucial feature of learning in animals is the flexibility to key in on virtually any stimulus. Rats, for example, can just as easily be trained to respond to a light breeze as to the sound of a bell or to vibrations in the ground, because the underlying neural mechanism is not stimulus-specific but all-purpose. Plants are known to respond to many aspects of their environment, from air temperature to soil moisture, but the extent to which a plant can build associations between stimuli other than blue light and a breeze remains to be seen.

The study by Gagliano and her colleagues was also short on controls, says Lincoln Taiz, professor emeritus at the University of California, Santa Cruz, who signed the 2007 letter criticizing the concept of plant neurobiology. Gagliano and her colleagues showed that the fan’s light breeze neither attracted nor repelled plants, but did not test whether the breeze interfered with the innate attraction toward light. Taiz points out that the slight rotating motion of a growing shoot’s tip, called circumnutation, influences growth patterns such as phototropism and could have been disrupted during the test.

Gagliano acknowledges the need for further work to explore how broadly the associative process can be generalized and to search for a mechanism that might account for it. “This is the beginning,” she says. “We needed to first show that learning by association was even possible.” Gagliano is no stranger to controversy, having weathered heavy criticism for her previously published evidence for habituation—a different type of learning—in plants, along with other findings.

As with earlier papers that sought to attribute animal-like behavior to plants, critics are likely to take issue with the authors’ use of words typically reserved for animals, including “learning” and “memory,” in much the same way researchers debated the moniker “plant neurobiology.” Those discussions are valuable and illuminating, says Gagliano, but should not obscure the underlying scientific exchange. “Unless we really explore [the field] experimentally, then we are just plant philosophers—and there are already plenty of good philosophers around.”

Add a Comment

Avatar of: You



Sign In with your LabX Media Group Passport to leave a comment

Not a member? Register Now!

LabX Media Group Passport Logo


Avatar of: Telekinetic


Posts: 5

February 9, 2017

Having seen my fair share of the "unexplainable", I've learned to resist the knee jerk reaction of dismissing anything out-of-hand. That's the mark of a true scientist, and BRAVA! to Ms. Gagliano. This continues the interesting work in the '60's when plants reactions to a range of stimuli like threats or music were measured.  

Avatar of: JonRichfield


Posts: 139

February 10, 2017

It is not work that I would have undertaken myself, but if stimulus-response is demonstrated, then the application of terms such as learning and memory are irrefutably semantically valid, neurons or no neurons.

If the term plant neurobiology OTOH  causes irritation, the semantic grounds are slim; it is a matter of taste. No one suggests that plants have neurons, but no one suggests that artificial neural networks have neurons either, and yet the term has neither caused collapse  of the field of technology, nor of its study.

If anyone or any group, really, really cannot tolerate such terms of convenience, it is up to them to propose something equally usable and suggestive, and establish its usage.

Such squabbles are routine in the advance of science and technology, and it is a mark of philological maturity not to take them too seriously. Consider for instance the harm done to the field of cosmology by the adoption of the term "black holes" which are not black, not holes, and in translation into some languages, and in translation even is obscene in some languages. Even in English, its overtones and subtexts caused some offence in the stage production "Oh Calcutta".

Avatar of: Mounthell


Posts: 53

February 14, 2017

@JonRichfield, thanks for the slice of sweet irony in science's soft underbelly of semantic indigestion! Shall we notch the discussion up a bit and dive into our labyrinthine collective ignorance about life?

1. We lack a sufficient and complete definition of "life."

2. We do not know fully what life is.

3. Yet life obviously involves ordinary matter doing something that, in polite educated circles, ordinary matter has not been given permission to do; what is it?

4. First, 'it' is not quantifiable or measurable but 'it' is associated with qualitative changes of which an irrelevant few are quantifiable.

5. Therefore, life's dynamics must not be a scientific topic because our mavens insist that their science is restricted to the measurable.

6. Yet that constraint casually dismisses from scientific consideration the most complex systems in the known universe ...

7. in which case, all science to date is little more than stamp collecting.

Avatar of: Old stick

Old stick

Posts: 5

February 15, 2017

Interesting work by ms. Gagliano. Maybe 'neurobiology' is not the best choice of a word for research into organisms which have no neurons. But that is a semantic issue. The question of 'plant learning' reminds me of research into 'intelligent' behaviour of Corvids. As I understand it, using a completely different brain structure than humans, Corvids are capable of reproducing intelligent responses to stimuli comparable to humans.

Avatar of: sanittar


Posts: 1

February 20, 2017

Plants have essentially utilized material (chemicals) for information transfer. Sir J.C.Bose opined that plants and animals share essentially similar fundamental physiological mechanisms. As do animals, plants co-ordinate their movements and responses to the world through electrical signaling1. He could demonstrate that oscillations in electric potential are linked to oscillations in cell turgor pressure. Thus electric charge could be used to control cellular turgidity.

Recently Dhir demonstrated that certain cells in bottle gourd plant tendril alter their turgor pressure in response to electric charge shortening2.

 It appears that plant kingdom evolved electric charge generation and utilized it for information transfer. However animal kingdom not only evolved electric charge generation but also evolved its transmission. Further neural tissue could analyze the information and institute appropriate response.

1.       J.C. Bose., The Nervous Mechanisms of Plants, Longmans, Green and Co. London (1926).

2.      Dhir S.P., International Journal of Scientific and Research Publications, Volume 6, Issue 8, August 2016.

Popular Now

  1. Estonia Offers Free Genetic Testing to Residents
  2. Human Brain Organoids Thrive in Mouse Brains
  3. New Ovarian Cancer Vaccine Shows Promise
  4. The Second March for Science a Smaller Affair