Why Do Babies Move Their Arms and Legs So Much?

Babies move their arms and legs constantly because those seemingly random kicks, waves, and flails are how their nervous systems wire themselves, how their brains learn where their bodies are, and how their bones and joints take shape. Far from being meaningless thrashing, infant movement is a form of self-education that begins months before birth and accelerates through the first year. The story behind all that wiggling involves spinal cord circuits, sleep, gravity, and a surprisingly tight connection between moving and thinking.

It Starts in the Womb

A baby’s habit of constant movement is not new at birth. Fetuses begin moving as early as the first trimester, and by the third trimester those movements are frequent and forceful enough for the mother to feel them clearly. Research has found continuity between fetal and newborn movement: all movements observed in fetal life are present in neonatal life, meaning babies are born already practicing a repertoire they have been rehearsing for months.¹ Even specific patterns like leg-kicking rates at 37 weeks of gestation carry over into the newborn period, with one study finding continuity between fetal and neonatal leg movements per minute, particularly during sleep states at birth.²

This prenatal movement is not just practice for life outside. Fetal kicking generates mechanical forces on developing cartilage and bone. Those forces help sculpt joint shape, and the hip joint is a prime example. Computational modeling studies have shown that a full range of symmetric fetal movements helps maintain the depth of the hip socket and the roundness of the femoral head, while reduced or absent movements can lead to flattened, deformed joints and may contribute to conditions like developmental dysplasia of the hip.³ The biomechanical stimulation from kicking is genuinely important for prenatal musculoskeletal development, particularly for shaping joints properly.⁴ In other words, a fetus that does not move enough may end up with skeletal problems that last well beyond birth.

Built-In Movement Circuits

One reason babies move so much is that they come equipped with neural hardware specifically designed to produce rhythmic movement. Circuits in the spinal cord, known as central pattern generators, can produce coordinated, repetitive leg movements without requiring instructions from the brain for every kick. These spinal circuits are the same ones that eventually underlie walking, and they are active long before a baby can stand. The brainstem controls and modulates these circuits, but the basic pattern of alternating flexion and extension in the legs is generated locally in the spinal cord itself.⁵

On top of those spinal circuits sits a collection of primitive reflexes. Newborns arrive with reflexes like the Moro (startle), grasp, and stepping reflexes, which trigger limb movements in response to specific stimuli. These reflexes are among the oldest and simplest tools clinicians use to check that a baby’s central nervous system is intact.⁶ They produce a lot of the arm and leg activity you see in a newborn, and they are supposed to fade over the first several months as higher brain centers mature and voluntary control takes over. A baby waving their arms after a loud noise is not choosing to do that; the Moro reflex fires automatically. But even involuntary movements serve a purpose in training the sensory and motor pathways they travel through.

Building a Body Map

Perhaps the most fascinating reason babies move so much is that movement is how they figure out they have a body and learn to control it. When a baby swings an arm and that arm brushes against their own torso, or when a kicking foot lands on the mattress, the brain receives two streams of information at once: it knows the motor command it sent, and it receives sensory feedback from the skin and joints. Matching those signals is the foundation of body mapping, the process by which infants connect local skin sensations with specific body locations and learn which limbs are moving, in which direction, how far, and how fast.⁷

These early sensorimotor experiences are not optional extras. They constitute a foundational step in developing motor control and set the stage for goal-directed actions like reaching for a toy or bringing food to the mouth.⁷ A baby who is not moving is a baby whose brain is getting less of the raw data it needs to build accurate internal models of the body. That is part of why restricting movement for extended periods can be a concern, and why occupational therapists pay close attention to how much and how variably a baby moves.

Research on spontaneous touches underscores this point. When scientists compared very preterm infants to healthy full-term infants, the preterm group made fewer spontaneous touches to their own body and to surfaces, roughly nine touches per minute compared to about thirteen in full-term babies.⁸ Those touches matter because each one is a tiny data point the brain uses to refine its map of the body.

Why Babies Move in Their Sleep

Parents often notice that their sleeping baby is far from still. Limbs twitch, fingers curl, legs jerk. This is not accidental. During active sleep, which is the infant equivalent of REM sleep, isolated limb movements send signals back to the developing brain’s sensory and motor areas. In neonatal animal models, these sleep movements are necessary for the normal development of the somatomotor cortex, and they are encoded in the brain’s electrical activity in a way that maps specific body parts to specific brain regions.⁹

Research in very preterm human infants, around 30 weeks corrected gestational age, has found similar neural patterns: limb movements during active sleep elicit brain oscillations that mirror what is seen in animal models, suggesting an analogous developmental role in humans.⁹ The implication is striking. When your sleeping baby’s arm twitches, their brain is quite literally using that twitch to calibrate its wiring. Sleep is not downtime for the infant motor system; it is a construction shift.

From Flailing to Reaching

If you watch a baby closely over the first few months, you can see the shift from chaotic arm waves to purposeful reaching. That transition does not happen overnight, and it does not follow the same path for every baby. A longitudinal study tracking four infants weekly found that first arm-extended reaches appeared anywhere from 12 to 22 weeks of age, and the babies used different strategies to achieve them. Two infants who had naturally large, vigorous spontaneous movements had to learn to dampen their force to control a reach. Two quieter infants had to generate faster, more energetic movements to lift their arms toward a toy.¹⁰

This tells us something important: the constant movement you see in early infancy is not just generically “good for development.” It provides each individual baby’s nervous system with the specific data it needs to calibrate its own unique body. A baby with strong, fast-twitch tendencies needs practice in inhibiting force. A quieter baby needs practice in generating it. Both accomplish this through high volumes of spontaneous movement. The early chaos is the raw material from which purposeful action gets carved.

Modern computational research has even drawn on this insight. One recent project analyzed the power spectral densities of babies’ limb velocities and found that the noise structure of infant movement changes with age, becoming more temporally organized over time. When researchers used that infant-inspired noise profile as an exploration strategy in artificial intelligence systems, it improved learning efficiency compared to conventional approaches.¹¹ Babies, it turns out, have stumbled onto an exploration strategy that even machine-learning engineers can learn from.

How Gravity Changes Everything

One thing that catches many parents off guard is how dramatically a baby’s movement changes depending on their position. A baby kicking vigorously on their back may seem almost motionless when held upright. This is not fatigue or disinterest; it is physics. When a three-month-old is placed in a more upright posture, gravitational resistance at the hip increases by four to ten times compared to lying flat on their back, which requires much larger muscle forces to produce the same flexion.¹²

Babies respond to this challenge by adjusting their movement patterns. In vertical postures, kicks show a smaller range of motion at the hip and a tighter synchronization between hip and knee movements, as the baby recruits joints together rather than independently to cope with the increased load.¹² What looks like a decrease in activity is actually the motor system adapting in real time to a new gravitational context. Even at three months, the infant nervous system is remarkably sensitive to the mechanical demands of a situation and adjusts its output accordingly. That is why tummy time, side-lying, and being held upright all elicit different movement patterns and all contribute different kinds of sensorimotor experience.

How Preterm Babies Move Differently

Babies born early offer a natural experiment for understanding why movement patterns matter. Preterm infants often show different movement characteristics compared to full-term babies, and those differences tend to resolve over a predictable timeline. One key finding is that preterm infants initially kick at much higher rates than full-term infants, followed by a steep decline that brings them in line with full-term levels by a corrected age of about 12 weeks.¹³

More telling is the coordination between joints. Full-term newborns show a tight coupling between the hip, knee, and ankle when they kick, meaning the joints flex and extend in a coordinated pattern. Preterm infants, especially those born before 32 weeks, initially show much weaker coordination between the knee and ankle, and between the hip and ankle. These differences appear to reflect neurological maturity rather than differences in leg size or postural control, and they typically resolve by around 12 weeks corrected age, coinciding with a known transformation in neural function around that period.¹³

There is also a quality difference. When researchers looked at center-of-pressure movements (essentially, how the body’s weight shifts during activity), preterm infants produced larger but more repetitive and stereotypic patterns, while full-term infants showed smaller but more complex and unpredictable movements.¹⁴ In motor development, variability and complexity are signs of a healthy, flexible system. A preterm baby’s more stereotypic patterns are not a sign that something is wrong per se, but they do reflect a nervous system that is still catching up in sophistication.

When Movement Patterns Raise Concerns

Because infant movement is so tightly linked to nervous system health, clinicians can learn a great deal just by observing how a baby moves. One widely used method is called General Movement Assessment, which looks at the quality, variability, and fluency of a baby’s spontaneous movements rather than checking for specific milestones. Normally, at around two to five months of corrected age, babies produce what are called “fidgety movements,” small, elegant, continuous movements of the neck, trunk, and limbs in all directions. When these fidgety movements are absent, it can be a warning sign.

A prospective study of high-risk infants found that absent or sporadic fidgety movements predicted cerebral palsy with a sensitivity of about 76% and a specificity of about 82%. When absent fidgety movements were combined with abnormal neonatal brain imaging, the accuracy of the prediction climbed to over 95%.¹⁵ Conversely, among infants who had only sporadic (not absent) fidgety movements, only about 8% went on to develop cerebral palsy, which is reassuring for parents whose baby might show an uneven pattern but still has fidgety movements present.¹⁵

Persistence of primitive reflexes beyond their expected window is another clinical flag. Most primitive reflexes fade during the first year as voluntary motor control improves. When they persist or when normal postural reactions fail to emerge on schedule, it can suggest that the central nervous system is not maturing as expected, and this is one of the earliest indicators of conditions like cerebral palsy.⁶ So the answer to “should I be worried about my baby’s constant moving?” is almost always no: vigorous, varied movement is exactly what healthy development looks like. It is a reduction in movement variety, or the absence of those small fidgety motions, that clinicians pay attention to.

The Energy Cost of All That Activity

All of this movement takes fuel. Parents sometimes wonder whether their baby’s constant wiggling is burning significant calories, and the answer is yes, though perhaps not as many as you would guess. Direct measurements in preterm infants found that the energy cost of physical activity accounted for about 5 to 17% of total energy expenditure. In a group of preterm infants studied with force plates and indirect calorimetry, total energy expenditure averaged about 69 calories per kilogram per day, and the portion attributable to physical activity was about 2.4 calories per kilogram per day.¹⁶

The energy cost varied dramatically by activity state. Quiet sleep burned about 0.5 calories per kilogram per day from movement, while active wakefulness burned about 7.5 and crying shot up to over 15.¹⁶ For parents of preterm babies in particular, this is a practical consideration: nursing strategies that minimize unnecessary crying and agitation can help conserve calories that these infants need for growth. For healthy full-term babies, the caloric cost of movement is well within their normal intake, so there is no need to worry that a wiggly baby is “wasting” energy by moving too much.

Swaddling and What Happens When You Restrict Movement

Given how important spontaneous movement is, it is worth asking what happens when you reduce it. Swaddling is the most common way parents constrain infant movement, and the research picture is mixed. A systematic review found that preterm infants who were swaddled showed improved neuromuscular development, less physiological distress, better motor organization, and more self-regulatory ability.¹⁷ For fragile preterm babies in a noisy, brightly lit NICU, swaddling may reduce the kind of disorganized, high-frequency movement that costs energy without providing much developmental benefit, while still allowing some movement within the wrap.

For healthy full-term infants, moderate swaddling for sleep is a different question. It can reduce startle-induced wake-ups and help babies sleep longer, but tight swaddling that immobilizes the hips is a recognized risk factor for hip dysplasia, especially given what we know about how important movement is for joint development. The general guidance from pediatric organizations is that if you swaddle, the legs and hips should have room to flex and move naturally. Swaddling should also stop once a baby can roll, which is typically around two to four months, because of the suffocation risk.

The broader point is that movement restriction is not inherently harmful, but the dose matters. Brief, gentle swaddling for calming is different from prolonged immobilization. A baby who spends most of their awake time free to kick and wave is getting the movement input their developing brain, muscles, and joints need, even if they spend some sleeping hours snugly wrapped.

Quantifying What “Normal” Movement Looks Like

One challenge in this field is defining normal infant movement precisely enough that clinicians and researchers can compare babies meaningfully. Traditional assessment methods like the General Movement Assessment rely on trained observers watching video and judging the quality of movement holistically. Newer approaches use computer-aided video analysis to quantify velocity, acceleration, range of motion, and the spatial shape of limb movements. Recent work on newborns in their second or third day of life found that limb movement parameters differed between left and right sides, and that infants whose movements were judged qualitatively as having “writhing” patterns showed greater range and more dynamic, circular movement shapes, while those judged as having “poor repertoire” movements showed less variation in velocity, acceleration, and range of motion.

This kind of quantitative analysis is still mostly a research tool rather than a routine clinical one, but it points toward a future where movement quality might be assessed more objectively and earlier. For parents, the practical takeaway is the same as it has been for decades: a baby who moves a lot, in varied ways, with smooth and flowing patterns, is showing signs of a well-functioning nervous system. The quantity of movement matters, but the quality and variety matter even more.

1
Frontiers in Neurorobotics. Fetal Origin of Sensorimotor Behavior
2
PubMed. Human fetal and neonatal movement patterns: Gender differences and fetal-to-neonatal continuity
3
PubMed Central. Effects of normal and abnormal loading conditions on morphogenesis of the prenatal hip joint: application to hip dysplasia
4
PubMed Central. Altered biomechanical stimulation of the developing hip joint in presence of hip dysplasia risk factors
5
PubMed Central. Spinal cord pattern generators for locomotion
6
PubMed Central. Primitive reflexes and postural reactions in the neurodevelopmental examination
7
ScienceDirect. Progress in Motor Control
8
PubMed Central. Quantity of spontaneous touches to body and surface in very preterm and healthy term infants
9
UCL. Developmental trajectory of movement-related cortical oscillations during active sleep in a cross-sectional cohort of pre-term and full-term human infants
10
Europe PMC. The transition to reaching: mapping intention and intrinsic dynamics
11
arXiv. Infant Spontaneous Movement Noise Improves Exploration in Deep RL
12
PubMed. Adaptive Dynamics of the Leg Movement Patterns of Human Infants: I. The Effects of Posture on Spontaneous Kicking
13
Developmental Psychobiology. The organization of leg movements in preterm and full-term infants after term age
14
PubMed Central. Infants born preterm exhibit different patterns of center-of-pressure movement than infants born at full term
15
PubMed Central. The Predictive Accuracy of the General Movement Assessment for Cerebral Palsy: A Prospective, Observational Study of High-Risk Infants in a Clinical Follow-Up Setting
16
PubMed. Direct measurement of the energy expenditure of physical activity in preterm infants
17
PubMed. Swaddling: a systematic review