Touch

Mimosa pudica, the sensitive plant, folds its leaflets shut within a fraction of a second of being touched, brushed by wind, or even warmed. The trigger is an actual electrical signal — an action potential, the same basic kind of spike a nerve fires — racing along the leaf's rachis from wherever it was set off. Where it arrives, a small hinge called a pulvinus at the base of each leaflet pumps potassium ions out of its motor cells; water follows the ions osmotically, the cells lose the turgor pressure that was holding the leaflet open, and it collapses. It's the same ion-and-water mechanism a stomatal guard cell uses to open and close a pore — run in reverse, and about a thousand times faster. A real leaf has more than one of these hinges: a small one at the base of every leaflet, the kind touched below, and one large one — the main pulvinus — where the whole leaf meets its stem. Touch a leaflet for the small response, or the base for the large one.

The fold you just watched spread is a real, measured thing, not a stylized one: a review of the plant's electrical signalling (Shimmen, 2006) puts the actual "m-wave" that drives this response at 1.5 to 4 centimeters per second. This room uses 2 cm/s, and draws the rachis to stand for about twelve real centimeters — so touching one end and watching the fold reach the other takes a few real seconds, the same order of magnitude a real leaf's own signal does. The fold itself, once the signal arrives at a given pulvinus, is close to instant: Sibaoka's classic recordings put the actual bending at 10 to 20 milliseconds after the local action potential fires — faster than this page's own 0.22-second transition, which is slowed down on purpose so you can see it happen at all.

Reopening is not drawn to any real scale — a real leaflet takes three to ten minutes to unfold again, which is a long time to hold a browser tab open on purpose, so this room compresses that to a few seconds. What it does not compress away is the direction of a real finding: touch the leaf again before it has fully reopened, and the next hiding time is measurably longer, not shorter. Reed-Guy et al. (PeerJ, 2017) found real Mimosa pudica plants hid about 13.6 seconds longer per successive touch in short-term trials — sensitization, the opposite of getting used to something. The same study found individual plants explained 41 to 75% of the variation in hiding time across their experiments, and that light-deprived, energy-stressed plants hid for measurably shorter stretches — staying folded costs a plant the photosynthesis it can't spare when it's already running low. None of that individual variation is modeled here; every visitor's leaf starts identical and reopens on the same fixed schedule once you hit Reset.

Honest gap: a real compound leaf has three separate hinges, not two. Touch the whole petiole hard enough and the entire leaf droops from its base in one motion — the base target above models this one, closing every leaflet at once rather than a wave crossing them, since Sibaoka's 1966 recordings of the main pulvinus itself describe a bend that follows its own action potential almost immediately, with no distance to spread across; touch one side-branch and just that branch's leaflets fold — this room still has no side-branches to model, only the one row; touch a single leaflet and, as above, the fold can stay local or spread to its neighbors depending on how hard the stimulus was. Neither target here models more than one stimulus strength, and both draw on the same touch counter for sensitization — a real plant's base and leaflet pulvini don't necessarily fatigue in lockstep, and this room doesn't distinguish them. It also doesn't model the actual reason the reopening compression matters: a real Reset never truly resets a real plant. Hiding time doesn't just grow with recent touches, it also depends on the plant's whole condition — hunger, light, individual identity — the way this page's does not.

There's a more famous, and more contested, claim about this plant that this room deliberately doesn't build: a 2014 study (Gagliano et al., Oecologia) reported that Mimosa pudica can learn to stop folding in response to a harmless, repeated stimulus (being dropped, not touched) and remember that for weeks — called, at the time, the first evidence of learning in a plant with no nervous system. A 2018 response in the same journal (Biegler) argued the data are equally explained by simple motor fatigue, not learning: the study never ran the one test — showing the response comes back after a different, novel stimulus (dishabituation) — that would tell the two apart. No independent lab has since confirmed it does. The room above shows something smaller and better established instead: a real signal, spreading at a real measured speed, getting more cautious with repetition rather than less. Whether that adds up to anything worth calling memory is a live argument, not a settled one. New field note: The fold is real. The memory isn't.

A newer, better-controlled 2025 study found a different kind of memory in the same species — not whether it learns to ignore a stimulus, but whether it can anticipate one: Stir runs Mimosa pudica through a light–dark cycle and shows the dark-hour movement that shows up hours before an expected dawn, and only when a dawn is actually coming.

Sources: Shimmen, Electrophysiology in Mechanosensing and Wounding Response, in Plant Electrophysiology — Theory & Methods (Volkov, ed.), Springer, 2006; Sibaoka, Action Potential and Rapid Movement in the Main Pulvinus of Mimosa pudica, Journal of Plant Research, 1966; Reed-Guy, Gehris, Shi & Blumstein, Sensitive Plant (Mimosa pudica) Hiding Time Depends on Individual and State, PeerJ 5:e3598, 2017; Gagliano, Renton, Depczynski & Mancuso, Experience Teaches Plants to Learn Faster and Forget Slower in Environments Where It Matters, Oecologia 175(1), 2014; Biegler, Insufficient Evidence for Habituation in Mimosa pudica, Oecologia 186(1), 2018.