Gait Apraxia

Abstract

Gait apraxia is a form of apraxia — a loss of skilled, learned movement — in which a person cannot organize the legs into normal walking despite intact strength, sensation, and coordination. First systematized by Meyer and Barron in 1960, the disorder is defined by a striking dissociation: leg movements performed lying down may be near-normal, yet upright walking becomes hesitant, shuffling, and “magnetic,” as though the feet were stuck to the floor. This article traces the concept from its clinical origins through the influential argument that it is better understood as a higher-level gait disorder than as a true apraxia, examines its basis in the frontal lobes and supplementary motor area, and reviews its assessment and its reversible causes such as normal-pressure hydrocephalus. Three interactive demonstrations explore the disorder.

Keywords: gait apraxia, higher-level gait disorder, frontal gait

Gait apraxia is the loss of the ability to walk normally that cannot be explained by weakness, sensory loss, or a disorder of the cerebellum or basic motor pathways — a breakdown, in the classical view, of the learned program for upright locomotion rather than of the machinery that carries it out. A patient with gait apraxia has adequate power in the legs and can often move them freely and even skillfully while lying down, yet, stood up and asked to walk, produces a hesitant, shuffling, foot-scraping gait, with difficulty starting, small dragging steps, and a tendency to seem glued to the floor (Meyer & Barron, 1960; Nutt, 2013). Because it is defined by the same logic as the limb apraxias — a skilled, learned act that fails when strength and sensation are intact — the National Library of Medicine files it in the Medical Subject Headings directly beneath apraxias, as “impaired ambulation not attributed to sensory impairment or motor weakness.”

The National Library of Medicine also files it, in its disease trees, among the neurologic manifestations and the gait disorders: gait apraxia is itself classified as a disorder — specifically a neurologic gait disorder, an acquired pathology of walking — and this article treats it as such, describing the lesions that produce it and the level of motor control they disrupt. Meyer and Barron introduced the modern concept in 1960 with a clinico-physiological study of patients who had lost the integration of walking while retaining the elementary leg movements, and the term has framed the field ever since, even as its accuracy has been vigorously contested (Meyer & Barron, 1960; Fisher, 1982).

The Defining Dissociation

The observation that made gait apraxia a coherent clinical entity is a dissociation between what the legs can do lying down and what they can do while walking. In the classical picture, a patient examined on the couch can move each leg with adequate strength, trace a figure in the air with a foot, or perform bicycling movements, sometimes almost normally; stood upright and asked to walk, the same patient hesitates to start, takes short shuffling steps with the feet barely clearing the ground, and may freeze in place, the feet appearing magnetically stuck to the floor (Meyer & Barron, 1960; Della Sala et al., 2002). The elementary movements survive; what fails is their assembly into the coordinated, postural, weight-shifting sequence that walking requires.

This dissociation is what licenses the word apraxia: as in the limb apraxias, a skilled and learned act breaks down while the motor apparatus that would execute it remains intact, so the deficit appears to lie in the program for the act rather than in the muscles (Leiguarda & Marsden, 2000; Rothi et al., 1991). The parallel is more than terminological. The cognitive model of limb praxis treats a skilled movement as the retrieval and execution of a stored spatiotemporal program, and gait apraxia can be read as the same failure applied to the most overlearned motor skill of all — upright bipedal locomotion — so that the representation of how to walk can no longer drive the legs even though the legs themselves are able (Rothi et al., 1991). The first demonstration makes this dissociation concrete, letting the reader place a patient supine or upright and compare the same leg movements in each posture.

The defining dissociation: leg movement vs. walking

Place the patient supine or standing, then choose a leg task. The elementary movements survive; only their assembly into walking fails.

LEG MOVEMENT SPARED

Preserved. The elementary, out-of-context leg movement is performed near-normally.

Higher-Level Gait Disorders

The single most influential development in the field was a challenge to the concept itself. Nutt, Marsden, and Thompson argued in 1993 that “gait apraxia” is at best a misleading label, because the tests meant to reveal leg apraxia — the supine tracing and bicycling movements — correlate poorly with the walking disorder and are often normal in patients whose gait is severely disturbed, while the walking deficit does not behave like a true apraxia of the classical limb type (Nutt et al., 1993). In its place they proposed a framework that organizes all gait disorders by the level of the nervous system at which control fails: a lowest level of peripheral musculoskeletal and sensory apparatus, a middle level of pyramidal, cerebellar, and extrapyramidal pathways, and a highest level of frontal cortical and cognitive systems that select, initiate, and adapt the walking program (Figure 1). What was called gait apraxia belongs to this highest level, and is more accurately named a higher-level gait disorder (Nutt et al., 1993; Thompson & Nutt, 2007).

Within that highest level, several recurring patterns are distinguished (Table 1). These subtypes are recognized by their phenomenology rather than by any confirmatory test, and a single patient may combine more than one; the classification is a clinical aid, not a set of separate diseases.

The reframing has organized decades of subsequent work. Liston and colleagues refined the highest-level category in patients with cerebral small-vessel disease, distinguishing subtypes such as gait ignition failure, frontal disequilibrium, and isolated gait initiation problems that had been lumped together under the older term (Liston et al., 2003). Snijders and colleagues folded the scheme into a practical clinical classification of gait disorders in the elderly, in which the highest-level disorders are diagnosed by their phenomenology — start hesitation, freezing, shuffling, and preserved lower-limb function out of context — rather than by a contested claim about apraxia (Snijders et al., 2007). Nutt later described the highest level as “an open frontier,” its disorders still the least understood and the hardest to classify of all gait problems (Nutt, 2013). The second demonstration builds this three-level model, letting the reader damage each level and see the resulting gait disorder.

The three levels of gait control

Damage a level and read the gait it produces. Only a highest-level lesion yields the higher-level gait disorder called gait apraxia, with preserved leg praxis on the couch.

Normal gait

Supine leg praxis: Normal

All three levels intact: the walking program is selected, transmitted, and executed normally.

Not everyone abandoned the stronger claim. Della Sala and colleagues reported patients with lesions of the bilateral supplementary motor area in whom the dissociation was as clean as the concept demands: leg praxis and strength were preserved, yet the organization of walking collapsed, which they took as evidence for a genuine gait apraxia arising from a specific cortical site rather than merely a diffuse higher-level disorder (Della Sala et al., 2002). The disagreement is not merely semantic: it turns on whether upright locomotion is served by a discrete, lesionable praxis representation or by a distributed control system whose failure looks apraxic without being apraxia in the limb sense (Zadikoff & Lang, 2005).

The Neuroanatomy of Gait Control

Whatever it is called, the disorder points to the highest, cortical tier of the locomotor system. The supplementary motor area and the adjacent medial frontal cortex are implicated most consistently: these regions program the postural adjustments and the sequencing that precede and accompany each step, and bilateral damage to them produces the cleanest cases of the syndrome (Della Sala et al., 2002). More diffuse disease of the frontal lobes and their subcortical connections — especially the periventricular white matter tracts linking frontal cortex to the basal ganglia and the brainstem locomotor centers — produces the common clinical picture, in which small-vessel ischemia or the stretched, distorted white matter of hydrocephalus interrupts the descending control of gait (Fisher, 1982; Liston et al., 2003).

Modern imaging has begun to give this account quantitative substance. Studies of Parkinson's disease, in which higher-level gait failure and freezing are prominent, link the microstructural integrity of the supplementary motor area to the severity of gait impairment, tying the clinical syndrome to a measurable property of the same medial frontal region that the lesion cases implicate (Wrobel et al., 2025). The convergence of focal-lesion, small-vessel, hydrocephalic, and degenerative causes on a frontal–subcortical network explains why so heterogeneous a set of diseases yields so recognizable a gait: they damage, by different routes, the same highest level of control (Leiguarda & Marsden, 2000; Nutt, 2013).

Assessment and Reversible Causes

Because the disorder is defined by exclusion, its assessment is a structured search for a walking deficit that survives after weakness, sensory loss, and cerebellar and extrapyramidal signs have been accounted for. The examiner tests leg strength and coordination with the patient seated or supine, looks for the classical dissociation between preserved out-of-context leg movement and disordered walking, and then observes the gait itself: the hesitation at initiation, the height and length of the steps, the width of the base, the turning, and the tendency to freeze (Snijders et al., 2007; Thompson & Nutt, 2007). The phenomenology is what classifies the case, since the highest-level disorders are recognized by their pattern rather than by any single confirmatory sign.

The most important reason to make the diagnosis is that some causes are reversible. Fisher's observation that gait disturbance in the elderly could be the presenting sign of hydrocephalus established that a shuffling, magnetic, higher-level gait, together with cognitive slowing and urinary urgency, may signal normal-pressure hydrocephalus — a condition in which removing cerebrospinal fluid can restore walking (Fisher, 1982). At the bedside this is tested directly: a large-volume lumbar puncture, the “tap test,” removes fluid and gait is timed before and after, an improvement predicting benefit from a shunt. The frontal-gait picture in hydrocephalus has itself been re-examined critically — whether it is as specific as once believed — but the principle stands that a reversible cause must be sought before a higher-level gait disorder is accepted as fixed (Morel et al., 2019). The third demonstration models this tap test, letting the reader remove cerebrospinal fluid and watch the gait parameters respond.

The tap test: probing reversibility

Choose a cause, then remove cerebrospinal fluid and watch the timed-walk parameters respond. A hydrocephalic gait improves; a fixed one does not.

BASELINE

Baseline: short strides and slow speed, before any fluid is removed.

Worked Example

The reasoning behind the diagnosis is best seen by working through a bedside examination, because the conclusion turns on how performance varies across conditions rather than on any single failure. Consider an older patient brought in for a gait that has declined over months. On the couch, leg strength is full, sensation is intact, there is no tremor or rigidity, and the patient can trace a circle in the air with each foot and perform bicycling movements with only mild slowing. Stood up, the same patient hesitates several seconds before the first step, then advances with short shuffling steps, feet barely clearing the floor, turning in many small movements rather than a single pivot, and pausing as if stuck in a doorway.

That contrast already localizes the failure. The preserved strength, sensation, and out-of-context leg movement place the deficit above the elementary motor and sensory levels; the intact tone and absence of tremor exclude the middle, extrapyramidal level; what remains is a failure of the highest level that selects and initiates the walking program — a higher-level gait disorder of the kind historically called gait apraxia. The diagnosis is then sharpened by asking what caused it. Cognitive slowing and urinary urgency alongside the gait raise normal-pressure hydrocephalus, and imaging shows enlarged ventricles out of proportion to cortical atrophy. The decisive test is reversibility: a large-volume lumbar puncture is performed, and the timed walk — stride length and speed — is compared before and after. A clear improvement points to a reversible hydrocephalic cause and predicts benefit from shunting; no change points instead to fixed small-vessel disease or a degenerative process. The same branching logic — exclude the lower levels, recognize the highest-level pattern, then test for reversibility — is what the three demonstrations animate, turning a vague “unsteadiness” into a localized and, sometimes, treatable diagnosis.

Discussion

Gait apraxia matters to cognitive psychology because it exposes upright walking — usually the most automatic of skills — as a cognitively controlled act that can fail at a high level while its motor components remain intact. The dissociation between preserved supine leg movement and disordered walking makes the same point for locomotion that the limb apraxias make for gesture: that skilled action depends on a program distinct from the apparatus that executes it, and that the two can be pulled apart by disease (Meyer & Barron, 1960; Rothi et al., 1991). The long dispute over whether the disorder is truly an apraxia is itself instructive, because it forces a precise question about what counts as a stored motor representation and whether walking has one in the sense that hammering does (Nutt et al., 1993; Della Sala et al., 2002).

The disorder also carries a practical lesson that pure theory can miss. Because higher-level gait failure is common in the elderly, disabling, and sometimes reversible, classifying it correctly is not an academic exercise: distinguishing a treatable hydrocephalus from fixed vascular disease changes what can be done for the patient (Fisher, 1982; Snijders et al., 2007). The migration from a single contested label to a graded, level-based classification is a case study in how a clinical concept sharpens: the phenomenon that Meyer and Barron named survived, but the theory around it was rebuilt to fit the evidence (Thompson & Nutt, 2007).

Current Directions

Contemporary work is pushing the field in two connected directions. The first is epidemiological and functional: recent population-based study is quantifying how common higher-level gait disorders are, and what they cost, showing associations with reduced quality of life, depression, and loss of confidence in walking that had been suspected clinically but not measured at scale (Larsson et al., 2025). This reframes the disorder from a diagnostic puzzle into a public-health problem of aging populations, and motivates the search for treatable subtypes.

The second is mechanistic. Advances in diffusion imaging are replacing the coarse lesion–syndrome correlations of the past with measurements of tract and cortical microstructure, so that the contribution of the supplementary motor area and its connections to gait impairment can be graded rather than merely localized (Wrobel et al., 2025). At the same time the specificity of the classical “frontal gait” of hydrocephalus is being reexamined, with careful comparison studies questioning whether the pattern is as distinctive as textbooks claim and therefore how much weight it should carry in selecting patients for shunting (Morel et al., 2019). Together these lines are moving the highest-level gait disorders — long the least understood tier of the locomotor system — from description toward measurement.

Common Misconceptions

Gait apraxia is a kind of leg weakness.

No. Strength is intact, and the same patient who cannot organize walking may move each leg competently while lying down. The deficit is in assembling the legs into gait, not in the muscles (Meyer & Barron, 1960).

The name means it is settled that walking has a stored “praxis” program.

The label is contested. An influential view holds that these patients have a higher-level gait disorder rather than a true apraxia, because leg praxis tested out of the walking context is often normal (Nutt et al., 1993).

A higher-level gait disorder is untreatable.

Some causes are reversible. A magnetic, shuffling gait with cognitive slowing and urinary urgency can signal normal-pressure hydrocephalus, in which removing cerebrospinal fluid may restore walking (Fisher, 1982).

Glossary

Bicycling movements.

Alternating leg movements a supine patient can be asked to perform; their preservation despite disordered walking is part of the defining dissociation of gait apraxia.

Freezing of gait.

A sudden, transient inability to step forward despite the intention to walk, often at initiation or in doorways; a hallmark of highest-level gait failure.

Frontal gait disorder.

A higher-level gait disorder arising from disease of the frontal lobes and their subcortical connections, marked by start hesitation, short shuffling steps, and freezing.

Gait apraxia.

The loss of the ability to walk normally in the absence of weakness, sensory loss, or cerebellar deficit, classically attributed to a failure of the learned program for locomotion.

Gait ignition failure.

A higher-level gait subtype in which the difficulty is concentrated at the start of walking, with prolonged hesitation before the first step.

Higher-level gait disorder.

The preferred term for a gait failure arising at the frontal, cognitive level of control, encompassing what was historically called gait apraxia.

Levels of gait control.

The division of the locomotor system into lowest (peripheral), middle (pyramidal, cerebellar, extrapyramidal), and highest (frontal, cognitive) levels, each producing a characteristic gait when it fails.

Magnetic gait.

A walking pattern in which the feet appear stuck to the floor, lifting only with difficulty; a classic descriptor of higher-level gait failure.

Normal-pressure hydrocephalus.

An enlargement of the cerebral ventricles at normal pressure, presenting with a higher-level gait disorder, cognitive slowing, and urinary urgency, and sometimes reversible by shunting.

Praxis.

The capacity for skilled, learned, purposive movement; apraxia, including gait apraxia, is its impairment.

Small-vessel disease.

Diffuse ischemic damage to the small penetrating arteries supplying subcortical white matter, a common cause of a fixed higher-level gait disorder in the elderly.

Supine leg praxis.

The performance of skilled leg movements while lying down, such as tracing a shape with the foot; its preservation contrasts with the disordered walking that defines the syndrome.

Supplementary motor area.

A medial frontal cortical region that programs postural adjustment and movement sequencing; bilateral damage to it produces the cleanest cases of gait apraxia.

Tap test.

A large-volume lumbar puncture that removes cerebrospinal fluid; gait timed before and after gauges reversibility and predicts benefit from shunting in normal-pressure hydrocephalus.

Key Researchers

Gilles Allali (contemporary). Neurologist at Lausanne University Hospital; studies higher-level gait disorders and the frontal-gait construct in normal-pressure hydrocephalus and aging. ORCID - Google Scholar

Bastiaan R. Bloem (contemporary). Neurologist at Radboud University Medical Center; a leading authority on gait, balance, and freezing in Parkinson's disease and their clinical classification. ORCID - Wikidata - Google Scholar

C. Miller Fisher (1913-2012). Canadian-American neurologist who identified hydrocephalus as a reversible cause of gait disturbance in the elderly, founding the modern approach to normal-pressure hydrocephalus. Wikipedia - Wikidata

C. David Marsden (1938-1998). British neurologist who co-authored the higher-level gait disorder framework and the influential account of limb apraxias as disorders of sensorimotor integration. Wikipedia - Wikidata

John G. Nutt (contemporary). Neurologist at Oregon Health & Science University; lead author of the three-level classification of gait disorders that reframed gait apraxia as a higher-level gait disorder. ORCID - Faculty

Sergio Della Sala (contemporary). Professor at the University of Edinburgh; reported gait apraxia after bilateral supplementary-motor-area lesion, defending a specific cortical basis for the disorder. ORCID - Wikipedia - Google Scholar

Philip D. Thompson (contemporary). Neurologist at the University of Adelaide; co-author of the higher-level gait disorder framework and of subsequent syntheses of the highest-level gait disorders. Faculty

Frequently Asked Questions

What is gait apraxia?
Gait apraxia is the loss of the ability to walk normally that cannot be explained by weakness, sensory loss, or cerebellar disease — the legs work, but the learned program that organizes them into walking fails (Meyer & Barron, 1960).

How is gait apraxia different from leg weakness?
Weakness is a failure of the motor apparatus; gait apraxia is a failure of the program that guides it. An affected patient has normal leg strength and may move each leg well lying down, yet cannot organize upright walking (Della Sala et al., 2002).

Why do some experts avoid the term “gait apraxia”?
Because tests of leg praxis performed out of the walking context are often normal in these patients, an influential view holds that they have a higher-level gait disorder rather than a true apraxia of the limb type (Nutt et al., 1993).

What does the gait look like?
It is typically hesitant at initiation, with short shuffling steps, feet that barely clear the floor, turning in many small movements, and a tendency to freeze, the feet seeming magnetically stuck (Snijders et al., 2007).

Which part of the brain is involved?
The frontal lobes, the supplementary motor area, and their subcortical connections; bilateral supplementary-motor-area damage produces the cleanest cases, and diffuse frontal white-matter disease the commonest (Della Sala et al., 2002; Liston et al., 2003).

Can gait apraxia be reversed?
Some causes can. A magnetic, shuffling gait with cognitive slowing and urinary urgency can signal normal-pressure hydrocephalus, in which removing cerebrospinal fluid may restore walking (Fisher, 1982).

What is the tap test?
A large-volume lumbar puncture that removes cerebrospinal fluid; gait is timed before and after, and an improvement predicts benefit from a shunt in normal-pressure hydrocephalus (Morel et al., 2019).

How common are higher-level gait disorders?
They are common in aging populations and carry a real burden; recent population-based study links them to reduced quality of life, depression, and lost confidence in walking (Larsson et al., 2025).

References