How does a legume learn?

Greetings phytophilosophy enthusiasts! Today marks the beginning of National Science Week here in Australia so I’ll be sharing a little plant science with you.

To begin with, I want to have a look at how plants learn. Learn? I hear you thinking, is that really something plants can do at all? If you’re anything like me, the concept of behavioural learning is something you might have wandered into while studying for introductory psychology exams, or perhaps stumbled upon after googling whether it’s possible to train your cat to open doors. At any rate, you might have thought, learning is surely something only animals are capable of – I mean, doesn’t learning require a brain? To the surprise of many though, recent research answers this question with a resounding no! While we might not generally think to associate learning with plants, we now know that plants are nonetheless out there associatively learning! However, the rigorous scientific study of plant learning is relatively new; plant physiologists have been interested in the topic for decades but robust demonstration of behavioural learning in plants has only occurred in the last few years. So, if you haven’t come across plant learning before, don’t worry – you aren’t too far behind the cutting-edge science!

Learning thyme*

Before we examine what learning looks like in plants, let’s take a step back: what is learning, anyway? At core, learning is a process whereby particular experiences lead to lasting changes in behaviour or knowledge. Some forms of learning, such as social learning, observational learning, and the kind of learning that comes about through activities such as play, require particularly complex kinds of nervous systems and social organisation. Other types of learning are more ubiquitous, and inform the basic ways that organisms interface with the world around them. For any organism born into an environment that varies from place to place or changes over time, the capacity to learn which things around them are dangerous, safe, nutritious, and so on – basically anything that relates to surviving and thriving – is an advantage. Given how this ability is likely to affect reproductive success, it is not surprising that many different sorts of organisms can learn – including plants!

gwrOne of the most basic kinds of learning, non-associative learning, was first studied in the giant sea slug Aplysia californica, work which later earned neuroscientist Eric Kandel his Nobel Prize in Physiology or Medicine. Aplysia has an involuntary reflex known as the gill and siphon withdrawal reflex (GSWR) – when stimulated by a curious neuroscientist or exposed to some other environmental disturbance, Aplysia’s siphon and gill retract defensively, as shown in the diagram below. If a harmless stimulus is repeatedly presented to Aplysia, there is a gradual decrease in GSWR for that stimulus, which is known as habituation. Habituation learning is useful so that organisms do not continuously respond to all the things they repeatedly encounter in their environments. Rather, they can ignore innocuous stimuli and put their time and energy to better use, foraging for food, looking for mates, or writing blogs.

Notwithstanding, if an organism suddenly encounters a different stimulus, it still wants to be responsive in case of danger. True habituation, then, needs to be stimulus-specific rather than just a general unresponsiveness. When a new stimulus is introduced and the organism starts responding to the novel stimulus as it did before habituation (so in the Aplysia case, responds with the GSWR), this is called dishabituation. Dishabituation distinguishes genuine learning from other kinds of changes in responsivity, such as that due to fatigue.

This sort of learning, then, basically occurs in environments where an organism must learn to ignore innocuous stimuli. In contrast, if our benign neuroscientist takes a sinister turn and repeatedly administers a harmful stimulus (such as an electric shock) to Aplysia, instead of a decrease we see a learned increase in responsiveness to the noxious and also neutral stimuli over time. This other type of behavioural learning is known as sensitisation.

So how does all this relate to plants? In 2014, biologist Monica Gagliano and colleagues put a little plant known as Mimosa pudica, the sensitive plant, to the test to see whether it could learn. The plant is so-named because, like our shy sea slug Aplysia, when stimulated, Mimosa’s little leaflets close up defensively (see picture).imageedit_2_3878689449

Gagliano’s research group was not the first to apply behavioural learning paradigms to plants or to Mimosa particularly, but earlier studies were not always well-controlled and findings were inconsistent. The rigour and comprehensiveness of this recent study make it the first to conclusively demonstrate true habituation (to dropping) and dishabituation (to shaking) in the plant. Mimosa plants learn that dropping is harmless and therefore do es not require a folding response, while still maintaining their capacity for responsiveness upon introduction of a novel, unexpected stimulus.

Moreover, these researchers found that not only did Mimosa plants display habituation learning, but that they did so in a context-sensitive way. Plants use their leaves to gather the sunlight they need to produce energy through photosynthesis. Therefore, in low-light environments, where this resource is scarce, there is a greater potential cost to a plant if it responds unnecessarily (leaf closure reduces the rate of photosynthesis by about 40%). By testing plants in both high and low light conditions, Gagliano’s team demonstrated that Mimosa plants learn faster in (low-light) environments where it the potential consequences are greater. Even when plants were initially germinated in high-light environments then transferred to a low-light context they displayed habituation much more rapidly than they had previously. It doesn’t take them long to learn and adapt!

Plants then, certainly seem capable of the simple sorts of non-associative learning that have previously been studied in animals like Aplysia. But what about more complex learning? Most animals are also capable of what is known as associative learning, which involves learning associations between different sorts of stimuli. No doubt many of you will at some stage have come across Pavlov’s famous experiments with dogs that demonstrate a kind of associative learning called classical conditioning. Meat powder naturally causes dogs to salivate. In Pavlov’s experiments, dogs that heard a bell paired with the presentation of meat powder eventually began to salivate when hearing the bell alone, even when meat powder was no longer present.

More generally then, in classical conditioning, organisms make associations that cause them to generalise their usual response to a stimulus onto a neutral stimulus with which it is paired. While the neutral stimulus would not usually elicit that sort of response, after an organism learns the association between the stimulus that would usually generate the response and the neutral one, the neutral stimulus alone becomes enough to elicit a response. Classical conditioning is important for organisms because it underlies the ability to make behavioural choices based on predicted outcomes. For instance, if a bell predicts when and where food will appear, it is adaptive to respond to the bell because it is a reliable predictor of reward.

In a study published late last year (2016), Gagliano and colleagues demonstrated that plants are capable of this sort of associative learning too. Moreover, they demonstrated this not with unusual, fast-moving plants like Mimosa, but this time with rather ordinary garden pea plants (Pisum sativum). One of the challenges of applying behavioural learning paradigms developed for animals to plants is in figuring out experimental designs that use environmental signals and plant responses that are ecologically relevant and can test the things we want them to. In this experiment the researchers used a Y-maze as shown in the picture. Plants learned that air movement in one arm of the Y-maze (caused by a fan) predicted subsequent light availability (in one experimental condition the fan was in the same arm as the light and in the other the opposite arm – plants were capable of learning either).


When seedlings in the test condition were examined for growth direction at the crucial junction of the Y-maze, they demonstrated a robust conditioned response to the fan. In contrast, plants in the control condition demonstrated only the innate blue light trophic response. This learning also turned out to be dependent upon signals relating to time of day – learning about light foraging cues occurred reliably only during subjective “day time” for the plant (which was manipulated experimentally). This makes sense as under usual conditions this would be when light might be available to plants and therefore the time to be looking out for cues that could predict its availability and location – attempting to do so at other times might incur too hefty an energetic cost.

Where to from here?

Plants, it seems, are smarter than we thought – they certainly display a pretty robust capacity to learn! What, then, does the current state of research into plant learning look like? The short answer is that it’s still very early days. This is an exciting time in the field as there are still so many questions! In the absence of neurons, what sorts of mechanisms enable this sort of learning in plants? How is information stored? Did the capacity for learning evolve several times in different lineages, or was it present before the point of evolutionary divergence? What sorts of directions will this research travel in? How can we improve, extend, and diversify our experiments? How can we foster cross-pollination between the fields of comparative psychology and plant biology? What sorts of things can plants learn? Does the capacity for learning vary across different members of the plant kingdom as it does with animals? To what extent do other non-neural organisms also display the capacity to learn?

Over the next few decades no doubt this research programme will expand and develop in unanticipated ways to answer some of these questions. In the meantime, it is uniquely exciting watching little pea seedlings in my garden emerging this season – as I spend my time learning about them, I wonder what they are learning about, too!

*Sorry, I couldn’t resist!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s