A plant with no brain, no neurons, and no nervous system appears to count. That is the central finding by William & Mary psychology professor Peter Vishton and his former student Paige Bartosh. Working with Mimosa pudica, commonly called the shy plant or touch-me-not, the researchers found that the plant’s leaf movements were shaped not simply by time, but by the number of light-and-dark cycles it had experienced.
The study, published in Cognitive Science in late 2025, presents what the authors describe as the first evidence that plants can enumerate, meaning they can distinguish and track discrete events in their environment. That capacity was previously assumed to require a nervous system.
Mimosa pudica folds its delicate, feathery leaves inward when touched or disturbed, then reopens them as conditions change. The plant also closes at night and reopens at dawn, a regular movement pattern called nyctinasty. It was this daily rhythm that Vishton and Bartosh used as their measurement tool.
The Shy Plant That Learned to Anticipate
Inside a windowless room at William & Mary’s Integrated Science Center, the researchers built a humidtent and exposed the plants to a repeating three-day cycle. On the first two days, the plants received 12 hours of light and 12 hours of darkness. On the third day, the lights stayed off entirely.
After roughly five repetitions of this cycle, the plants began moving more during the dark hours just before light was expected, but only on the days when light was actually coming. On the third day, the dark day, that pre-dawn activity dropped. The plants appeared to have learned the sequence.
The learning curve itself matched a well-established pattern from animal studies. “If you are teaching a rat to perform a series of actions in a certain order, you would expect to see a period of time when they’re figuring out the sequence and then a gradual increase in their ability to predict the pattern,” Vishton said. The plants followed the same arc, a rapid initial adjustment that leveled off into consistent anticipatory behavior.
Not the Clock. Something Else.
A straightforward explanation for the behavior was that the plants were simply following acircadian rhythm, the internal biological clock found in many organisms. Many plants naturally open and close on a roughly 24-hour cycle. If the Mimosa pudica plants were just tracking time, the experiment would need a different design to prove otherwise.
Vishton and Bartosh addressed this by shortening the day length from 24 hours to 20 hours. The plants adjusted almost immediately, reorganizing their movement around the new schedule. That flexibility argued against a fixed internal clock driving the behavior.
Then the researchers ran a final set of tests using random cycle lengths ranging from 10 to 32 hours per day. When cycles were shorter than 12 hours or longer than 24, the pattern broke down. The plants stopped anticipating correctly. But within that 12-to-24 hour window, they continued showing more movement ahead of expected light days and less before dark days.
“The simplest explanation for this result,” Vishton said, “is that these plants are tracking the number of events that take place, not simply responding to time.”
The failure at the extremes suggested something about the plant’s processing limits: a minimum window needed to register each light event, and a maximum duration before the pattern fades from whatever biological substrate holds it.
No Neurons Required
The broader implication of the findings, as Vishton framed it, is that cognitive-like functions may not be exclusive to organisms with neurons. “Every theory I’ve ever read on memory and decision making always involves neurons,” he said. “Big surprise, plants don’t have those. And yet it looks like they can perform cognitive-like functions. Just not cognitively, per se.”
Mimosa pudica movement is mediated by structures called pulvini, joint-like swellings at the base of each leaf and leaflet. These contain two layers of motor cells that regulate movement through rapid shifts in turgor pressure, driven by ionic exchanges involving potassium, chloride, and calcium. No neurons are involved. The information processing described in the study, if the results hold, would be occurring through biochemical and cellular mechanisms that science has not yet mapped.
Vishton said the mechanistic questions fall to chemists and biologists. His role was to characterize the behavior. “I’m hoping the chemists and biologists of this world can ask more mechanistic questions to understand how this is actually happening,” he said.
The study also raises the question of whether non-neuronal cells in animals and humans might be more capable than assumed. “There are lots of cells in animals and humans that aren’t neurons,” Vishton noted. “And we just assume they’re not involved in learning. But maybe they could be.”
What the Researchers Are Confident About
The authors are careful about the study’s scope. The paper acknowledges that higher variability was present in some results and calls for replication with additional controls before the findings can be considered settled.
Vishton and Bartosh identified potential applications including plant-based sensors, biologically derived computational devices, and research into how cellular-level learning might relate to habit formation and unlearning in humans. Those applications remain speculative for now.
Enjoyed this article? Subscribe to our free newsletter for engaging stories, exclusive content, and the latest news.