Kinesis

Abstract

Kinesis is a form of animal behavior—an undirected locomotor response in which the speed or the turning rate of movement changes with the intensity of a stimulus, with no steering toward or away from its source. It is the simplest way an organism reaches favorable conditions: because movement slows where conditions are good and quickens where they are poor, individuals accumulate statistically in the favorable zone without ever orienting to it. Research separates two kinds—orthokinesis, the modulation of speed, and klinokinesis, the modulation of turning—and the same logic reappears as the run-and-tumble biased random walk of a swimming bacterium and the reorientation bursts of a crawling worm or fly larva. A worked example and three interactive demonstrations accompany the text.

Keywords: kinesis, orthokinesis, klinokinesis, biased random walk, undirected orientation

A woodlouse placed on a surface that is dry at one end and damp at the other does not turn and march toward the damp. It wanders, apparently at random, yet after a while most of the animals are clustered in the damp region. Nothing in the woodlouse steered it there. It simply walks faster, and turns less, in dry air, and walks slower, turning more, in humid air—and that difference in pace is enough to concentrate a scattered population where the air is moist. This is kinesis: an orientation that arrives at a destination without ever aiming at it.

This article treats kinesis as the undirected limb of orientation behavior, the counterpart to directed taxis. The central distinction, fixed by Gottfried Fraenkel and Donald Gunn in the monograph that named the field, is that a taxis uses the direction of a stimulus to steer, whereas a kinesis uses only its intensity to set a rate—of locomotion or of turning (Fraenkel & Gunn, 1961). The questions that follow are mechanical. What exactly is modulated, and by what? How can a rule that contains no sense of direction nonetheless deliver an organism to the right place? And why does the same strategy, under different names, keep turning up from bacteria to insect larvae?

What Kinesis Is

Kinesis is locomotor behavior that is not a steering reaction, but in which movement may turn at random in direction, with the rate of movement or of turning tied to the intensity of stimulation. In MeSH it is classified under animal behavior, and its defining contrast is with taxis. A taxis is a directed response: a moth steering along the bearing of a light, a bacterium headed up the axis of a gradient, an animal that uses the direction from which a stimulus comes to orient its body. A kinesis discards that directional information entirely. The organism registers only how strong the stimulus is, here and now, and lets that reading govern a rate (Fraenkel & Gunn, 1961).

The consequence is counterintuitive: a rule with no sense of direction still produces a destination. If an animal moves quickly through unfavorable conditions and slowly through favorable ones, it spends more time in the favorable region simply because it dawdles there, and a population of such animals piles up where the going is slow. No individual ever decided to go there. The aggregation is a property of the statistics of movement, not of any goal held in the animal (Benhamou & Bovet, 1989). This is why kinesis is the natural baseline for the study of orientation: it is the least an organism can do and still end up somewhere useful.

Orthokinesis and Klinokinesis

Fraenkel and Gunn divided kinesis into two kinds by asking what the stimulus intensity controls (Fraenkel & Gunn, 1961). In orthokinesis, intensity sets the speed of locomotion: the organism moves faster at one stimulus level and slower at another, and may stop altogether where conditions are best. The woodlouse in humid air is the textbook case—its walking speed falls as humidity rises, so it stalls in the damp and hurries through the dry, and the colony drifts toward moisture. In klinokinesis, intensity sets the rate of turning: the organism changes direction more or less often depending on the stimulus, holding a straighter course under favorable change and tumbling into a new heading under unfavorable change.

The two are mechanically distinct and can be measured separately—speed against time, versus turns per unit distance—but they often act together, and both share the signature that disqualifies them as taxes: neither uses the direction of the stimulus. A klinokinetic animal that turns more in worsening conditions does not turn toward anything; it simply reorients at random more frequently, and the net effect of turning less when things are improving is to keep it, on average, moving the right way. Simon Benhamou and Pierre Bovet showed formally that this undirected rule, applied in a heterogeneous environment, produces a stable, predictable distribution of animals—a reframing of kinesis as a way of exploiting the environment rather than merely reacting to it (Benhamou & Bovet, 1989).

The distinction matters because the same word is used loosely in the wider literature for any undirected response, and the intensity-to-rate mapping is what keeps it precise. An orthokinetic response can be plotted as a speed curve against stimulus level; a klinokinetic one as a turning-rate curve. Where a measured response bends with intensity but shows no bias with direction, it is a kinesis, whatever the modality of the stimulus—humidity, temperature, light, or a dissolved chemical.

Kinesis as a Biased Random Walk

The most quantitatively understood kinesis is not in an insect but in a bacterium. Howard Berg and Douglas Brown tracked individual Escherichia coli cells in three dimensions and resolved their swimming into two states: smooth, roughly straight runs, punctuated by brief tumbles that randomize the next heading (Berg & Brown, 1972). In uniform conditions the alternation is a pure random walk that goes nowhere on average. In a chemical gradient the cell does one thing only: when it senses that conditions are improving along its current run—attractant concentration rising over the last second or so—it suppresses the next tumble and runs on; when conditions are flat or worsening, it tumbles at the baseline rate. Lengthening the favorable runs while leaving the unfavorable ones unchanged biases the random walk up the gradient.

This is klinokinesis in its purest form. The bacterium has no idea which way the gradient points—it is far too small to compare concentration across its own body—so it cannot perform a taxis in the strict sense. It compares concentration across time, as it moves, and converts that temporal derivative into a turning rate. The directionality of the outcome is entirely emergent: a longer run in a good direction simply happens to carry the cell further before the next random reorientation. Edward Codling and colleagues set this kind of rule inside the general theory of correlated and biased random walks, showing how a small per-step bias accumulates into reliable net transport—the mathematical backbone shared by microbial chemotaxis and animal search alike (Codling et al., 2008).

The temporal-comparison trick is the deep reason kinesis is so widespread. A directional sense is expensive: it requires either spatially separated detectors or sequential sampling with memory of direction. A kinesis needs neither. One sensor, read twice, yields a single number—getting better, or not—and that number is enough to climb a gradient. For an organism at the bottom end of the size or nervous-system scale, it is often the only navigation available, and it is remarkably good.

From Bacteria to Larvae: Kinesis Across Species

The same biased-random-walk logic reappears, elaborated, in small animals with real nervous systems. In the nematode Caenorhabditis elegans, Jonathan Pierce-Shimomura and Shawn Lockery showed that chemotaxis is driven by what they called pirouettes—bursts of sharp, frequent turns that the worm triggers when it detects that it is moving down a chemical gradient, and suppresses when it is moving up (Pierce-Shimomura et al., 1999). The worm, like the bacterium, is modulating its turning rate from a temporal reading of concentration; the pirouette is klinokinesis implemented in a 302-neuron circuit, and it was the first demonstration that this microbial strategy is also the core of orientation in an animal.

The fruit-fly larva adds a further layer. Alex Gomez-Marin, Greg Stephens, and Matthieu Louis found that Drosophila larvae combine klinokinesis with active sampling: between runs the larva sweeps its head from side to side, casting the olfactory organ across the local gradient, and uses that sampling to decide both when to turn and, increasingly, which way—a bridge from pure undirected kinesis toward a weakly directed strategy (Gomez-Marin et al., 2011). The same mixture governs walking adult flies, whose attraction to an odor Liangyu Tao, Siddharth Ozarkar, and Vikas Bhandawat decomposed into modulation of walking speed and turn rate—orthokinesis and klinokinesis operating together in a freely walking insect (Tao et al., 2020). Across this range, the comparative lesson drawn by Kate Baker and colleagues is that olfactory search in animals from worms to flies to mammals is built from a small, shared kit of movement rules, of which kinesis is the most basic (Baker et al., 2018).

The clearest way to see what kinesis is—and is not—is to set it beside its directed cousin. A taxis steers: it reads the gradient's direction and orients the body along it, so its path climbs almost straight. A kinesis cannot steer; it only modulates how fast or how often it turns, and its path wanders upward by biasing an otherwise random walk. Both reach the source; only one knows where the source is.

Worked Example

Orthokinesis makes a sharp, checkable prediction: at steady state the density of animals in a region is inversely proportional to their speed there. The argument is a flux balance. Consider a long arena divided into a dry zone and a humid zone, with woodlice crossing freely between them. Suppose a woodlouse walks at a speed of 16 mm/s in the dry zone and, slowed by the moisture, at 4 mm/s in the humid zone, with no preference for which way it turns at the boundary—a pure orthokinesis, no steering.

At equilibrium the number of animals crossing the boundary per unit time must be equal in both directions, or the distribution would still be changing. The rate of crossing from a zone is proportional to how many animals are in it multiplied by how fast they move, so:

Writing n for density and v for speed, flux balance at the boundary requires n_dry × v_dry = n_humid × v_humid. Rearranging, the density ratio is the inverse of the speed ratio: n_humid ÷ n_dry = v_dry ÷ v_humid = 16 ÷ 4 = 4. The humid zone ends up four times as crowded as the dry one. With 100 animals split across two equal-area zones, that is 80 in the humid zone and 20 in the dry—a four-to-one aggregation into the moisture, produced by animals that never once turned toward it. The entire effect is the slowdown: lingering where the stimulus is favorable is, statistically, the same as preferring it. The OrthokinesisAggregationDemo lets the reader set the two speeds and watch a scattered population settle to exactly this inverse-speed distribution.

Discussion

Kinesis occupies a specific place in the logic of orientation: it is what directed behavior is defined against. Fraenkel and Gunn's taxonomy was valuable precisely because it forced a mechanistic question—does this response use the direction of the stimulus, or only its intensity?—and that question still organizes the field (Fraenkel & Gunn, 1961). The answer is rarely all-or-nothing. Real organisms blend an undirected intensity response with whatever weak directional information they can extract, and much of modern navigation research is the work of separating the two contributions in a given behavior (Gomez-Marin et al., 2011; Tao et al., 2020).

The enduring surprise of kinesis is how much it achieves with how little. A single intensity reading, compared against its own recent past and converted into a speed or a turning rate, is enough to carry a bacterium up a gradient, hold a woodlouse in a damp refuge, and steer a worm to food. The strategy is slow and statistical rather than fast and certain, but it is robust, cheap, and available to organisms with no spatial sense whatsoever (Berg & Brown, 1972; Codling et al., 2008). That combination is why it recurs across such distant branches of life: it is less a quirk of particular animals than a near-inevitable solution for anything that must find better conditions while sensing only how good the current ones are.

Cognitive Implications

Kinesis is not a cognitive process, and that is exactly its interest to cognitive science: it marks the floor. It shows that behavior which looks goal-directed—an animal reliably ending up where it is best off—can be produced with no internal representation of the goal, no map, no comparison of directions, and no decision in any meaningful sense. The woodlouse in the damp has solved a navigation problem without having a navigation problem; the solution lives in the statistics of its movement, not in its head. Treating kinesis as the baseline keeps the attribution of richer cognition honest: a directed, representational account of some behavior earns its keep only where a kinetic, non-representational account provably fails (Benhamou & Bovet, 1989).

The boundary is not a wall, however, and the interesting cases sit on it. The Drosophila larva's head-casting is a kinesis that has begun to sample actively, acquiring a sliver of directional information and edging toward something more like a taxis (Gomez-Marin et al., 2011). Esteban Gorostiza has argued that even such innate, apparently reflexive orientation in the fly is plastic and context-dependent in ways that complicate the old reflex-versus-cognition dichotomy, so that the line between a fixed kinesis and a flexible, quasi-cognitive strategy is itself a research question rather than a given (Gorostiza, 2018). Kinesis thus serves cognition twice over: as the null model that representational claims must beat, and as the primitive from which graded, increasingly directed strategies demonstrably evolve.

Current Directions

The active work on kinesis is quantitative and comparative. Kate Baker and colleagues set out a shared framework for olfactory search across species, arguing that the movement algorithms of bacteria, worms, flies, and mammals can be described in a common language of run-and-turn rules, with kinesis as the shared substrate on which directional refinements are layered (Baker et al., 2018). Within that program, Liangyu Tao and colleagues dissected odor attraction in walking Drosophila into separately measurable orthokinetic and klinokinetic components, showing that an adult insect's approach to an odor source is built from the same intensity-driven speed and turn modulations first described in woodlice and bacteria (Tao et al., 2020).

A second front places kinesis inside movement ecology and search theory. Frederic Bartumeus and colleagues analyzed the trade-offs that govern foraging success under uncertainty, treating reorientation rules—how often and how sharply an animal turns—as strategic variables that an organism can tune to the structure of its environment, and placing undirected, intensity-dependent movement on a continuum with the directed and the purely random (Bartumeus et al., 2016). The open questions are shared across these efforts: how an organism sets its turning statistics to match a particular gradient landscape, how orthokinesis and klinokinesis are combined and weighted in a single behavior, and where exactly, in a given species, an undirected kinesis gives way to a genuinely directed taxis.

Common Misconceptions

Kinesis means the animal moves toward the favorable conditions.

It does not. A kinesis uses only stimulus intensity to set a rate of movement or turning; it contains no information about direction, so the animal never steers toward the stimulus. Accumulation in the favorable zone is a statistical byproduct of moving slowly there, not an act of approach (Fraenkel & Gunn, 1961).

Kinesis and taxis are two words for the same thing.

They are the two halves of a deliberate distinction. A taxis is directed—it steers using the direction of the stimulus—whereas a kinesis is undirected, modulating only a rate from the stimulus intensity. The direction-versus-intensity test is what separates them (Fraenkel & Gunn, 1961).

A random walk cannot reliably carry an organism to a goal.

A biased random walk can. By lengthening its runs when conditions improve and tumbling at the baseline rate otherwise, a swimming bacterium converts an undirected walk into reliable net transport up a chemical gradient (Berg & Brown, 1972).

Klinokinesis requires comparing the stimulus at two points across the body.

It requires only comparison across time. The organism samples the stimulus as it moves and adjusts its turning rate from the change over the last moment, which is why a bacterium far too small to detect a spatial gradient across itself can still perform it (Pierce-Shimomura et al., 1999).

Glossary

Active sampling.

The movement of a sense organ—such as a fly larva's side-to-side head casting—to read a stimulus across space, adding weak directional information to an otherwise undirected kinesis.

Animal behavior.

The broad study of what animals do and the mechanisms behind it; the MeSH category under which kinesis is classified.

Biased random walk.

A random walk in which a small, consistent per-step bias, such as longer runs in a favorable direction, accumulates into reliable net movement.

Chemotaxis.

Movement in response to a chemical gradient; in small organisms such as bacteria it is typically implemented as a klinokinesis rather than a true directed taxis.

Gradient.

A gradual change in the intensity of a stimulus across space, such as rising humidity or chemical concentration, that a kinesis exploits through its intensity response.

Kinesis.

An undirected locomotor response in which the speed or turning rate of movement varies with stimulus intensity, with no steering relative to the stimulus direction.

Klinokinesis.

A kinesis in which stimulus intensity sets the rate of turning or reorientation; straighter movement under favorable change, more frequent turning under unfavorable change.

Orientation.

The positioning or movement of an animal relative to a stimulus, encompassing both undirected kineses and directed taxes.

Orthokinesis.

A kinesis in which stimulus intensity sets the speed of locomotion; faster movement in unfavorable conditions, slower or arrested movement in favorable ones.

Pirouette.

In C. elegans chemotaxis, a burst of sharp turns triggered when the worm senses it is moving down a gradient; the nematode form of klinokinesis.

Run.

A smooth, roughly straight segment of movement between reorientations, as in the run-and-tumble swimming of a bacterium.

Taxis.

A directed orientation response that uses the direction of a stimulus to steer the body toward or away from its source, in contrast to the undirected kinesis.

Tumble.

A brief reorientation event that randomizes an organism's heading before the next run; its frequency, modulated by stimulus change, is what biases the walk.

Undirected response.

A behavior governed by the intensity of a stimulus rather than its direction, so that any resulting approach to a source is statistical rather than steered.

Key Researchers

Frederic Bartumeus

(CEAB-CSIC and CREAF, Catalonia). Movement ecologist who analyzes stochastic search strategies and the trade-offs between random reorientation and directed movement, placing kinesis on a continuum of optimal space-use strategies under uncertainty. ORCID - Google Scholar

Simon Benhamou

(CNRS, Centre d'Ecologie Fonctionnelle et Evolutive, Montpellier). Behavioral ecologist who reframed kinesis within movement ecology, modeling how undirected, intensity-dependent movement lets animals exploit heterogeneous environments. Google Scholar

Howard C. Berg

(1934–2021; Harvard University). Biophysicist who reconstructed bacterial swimming as a biased random walk of runs and tumbles, the microbial archetype of klinokinesis and the quantitative template for intensity-driven turning. Wikipedia - Wikidata

Gottfried S. Fraenkel

(1901–1984; University of Illinois). Insect physiologist who, with Donald Gunn, wrote the monograph that defined the taxis-and-kinesis framework, distinguishing undirected kineses from directed taxes as categories of animal orientation. Wikipedia - Wikidata

Shawn R. Lockery

(University of Oregon). Neuroscientist who quantified C. elegans chemotaxis as klinokinesis via pirouettes, giving a tractable nervous-system model of how intensity-driven turning produces gradient climbing. ORCID - Google Scholar

Matthieu Louis

(University of California, Santa Barbara). Systems neuroscientist who dissected odor-guided navigation in the Drosophila larva, showing how active sampling and run-length modulation implement a chemokinetic search strategy at single-neuron resolution. Google Scholar

Frequently Asked Questions

What is kinesis?

Kinesis is an undirected locomotor response in which an organism changes its speed or its rate of turning according to the intensity of a stimulus, without steering toward or away from the stimulus source. It is classified in MeSH as a form of animal behavior and contrasts with directed taxis (Fraenkel & Gunn, 1961).

What is the difference between kinesis and taxis?

A taxis is directed: it uses the direction of a stimulus to steer the body toward or away from it. A kinesis is undirected: it uses only the intensity of the stimulus to set a rate of movement or turning. The test is whether the response depends on the stimulus direction or only on its strength (Fraenkel & Gunn, 1961).

What are orthokinesis and klinokinesis?

They are the two kinds of kinesis. Orthokinesis is the modulation of locomotor speed by stimulus intensity; klinokinesis is the modulation of turning rate by stimulus intensity. Both are undirected, and they often operate together in the same animal (Fraenkel & Gunn, 1961; Tao et al., 2020).

How can undirected movement bring an animal to a favorable place?

Because the animal lingers where conditions are good and hurries where they are poor. Moving slowly in the favorable zone means spending more time there, so a scattered population accumulates in it as a statistical consequence of pace, with no individual ever heading toward it (Benhamou & Bovet, 1989).

Is bacterial chemotaxis a kinesis?

Yes. A swimming Escherichia coli cell alternates straight runs with random tumbles and simply lengthens its runs when it senses conditions improving over time. This biases an otherwise random walk up the gradient without the cell ever detecting which way the gradient points, which is klinokinesis (Berg & Brown, 1972).

How is kinesis studied in the laboratory?

By tracking individual trajectories and separating speed from turning rate as functions of the stimulus. Classic work tracked bacteria in three dimensions and nematodes and fly larvae on gradient plates, measuring run lengths, tumble or pirouette frequency, and head-casting to isolate the intensity response (Berg & Brown, 1972; Pierce-Shimomura et al., 1999).

How does kinesis relate to cognition?

Kinesis is a non-cognitive strategy, and that is its significance for cognitive science: it shows that goal-approaching behavior can arise with no internal representation of the goal, so it serves as the null model that any claim of genuine spatial cognition must outperform (Benhamou & Bovet, 1989; Gorostiza, 2018).

Why does kinesis matter to movement ecology and search theory?

Because it is the minimal, low-information search rule against which richer strategies are measured. Treating turning statistics as tunable strategic variables lets researchers model how organisms match their movement to the structure and uncertainty of their environment (Bartumeus et al., 2016; Codling et al., 2008).

References