Spinal muscular atrophy (SMA) primarily destroys motor neurons in the spinal cord, but the protein it disrupts is active throughout the brain. The SMN protein, which is deficient in SMA, is naturally produced in several brain regions, and growing evidence shows that some people with SMA, particularly those with the most severe forms, experience measurable effects on cognition. The short answer: SMA is not purely a motor neuron disease, though the brain involvement is far subtler than the muscle weakness that defines it.
The SMN Protein Is Active Across the Brain
The gene responsible for SMA produces a protein called SMN, and its expression isn’t limited to the spinal cord. During fetal development and into adulthood, high levels of SMN appear in the pyramidal neurons of the cerebral cortex, the Purkinje cells of the cerebellum, and neurons in the brainstem’s medulla oblongata (including the olive and hypoglossal nuclei). Sensory neurons in the spinal cord also express SMN, though at lower levels than motor neurons do.
This widespread expression suggests SMN plays a role in the development and maintenance of multiple types of nerve cells, not just the motor neurons that control muscle movement. When the body can’t produce enough of this protein, the motor neurons are hit hardest, but other parts of the nervous system aren’t entirely spared.
What Brain Imaging Shows
MRI studies have found structural differences in the brains of people with SMA, and the pattern depends on disease severity. In those with infantile-onset SMA (type 1), researchers have documented abnormalities in white matter, the basal ganglia, the thalamus, and the hippocampus, along with areas of high signal intensity around the lateral ventricles. In people with later-onset forms, the changes look different: reduced cerebellar volume and increased gray matter density in motor areas of the brain. These findings confirm that SMA leaves a footprint beyond the spinal cord, even if the clinical picture is dominated by muscle weakness.
Cognitive Effects by SMA Type
For most people with SMA types 2 and 3, overall intelligence falls within the normal range. Some studies have actually found that adolescents with SMA score higher than their peers on intelligence tests, likely because years of limited physical activity push them toward more intellectually demanding pursuits. A direct comparison across SMA types 1, 2, and 3 found no significant differences in general intelligence or cognitive function between groups.
That said, the picture isn’t entirely clean. Among adults with SMA type 2, about 33% scored below normal in verbal fluency and roughly 22% showed impairment in executive function (the ability to plan, organize, and shift between tasks). Working memory and perceptual reasoning scores were also lower in SMA type 2 compared to the general population. Children with SMA type 1 performed worse than healthy children on tasks requiring attention, spatial processing, and executive function, particularly as the tasks became more complex.
Verbal fluency, the ability to quickly generate words within a category, appears to be the cognitive domain most frequently affected across SMA populations. Language skills also differ between types: adults with SMA type 2 scored significantly lower on language measures than those with type 3.
SMN2 Copy Number Shapes Cognitive Outcomes
The severity of SMA depends largely on how many backup copies of a gene called SMN2 a person carries. More copies mean more functional protein and generally milder disease. This same relationship extends to the brain.
In a study of 40 children identified through newborn screening and treated early, the average cognitive score was 94.55 on the Bayley Scales (where 100 is the population average). But 35% of these children scored below average on cognitive testing, and 71% of those with below-average scores had only two SMN2 copies, the lowest number in the group. The fewer SMN2 copies a child carried, the lower their cognitive development score. This correlation between copy number and cognition was actually stronger than the correlation between copy number and motor function.
In a separate study of 20 treated SMA type 1 patients, 55% showed subnormal cognitive development. The results were highly variable from child to child, but the trend was clear: less SMN protein means more vulnerability in the brain, not just in spinal motor neurons.
Why Early Treatment Doesn’t Fully Protect the Brain
One of the more striking findings is that cognitive effects persist even in children who received treatment in the first weeks of life. The newborn screening study found that children with two SMN2 copies still showed impaired cognitive development despite being treated before symptoms appeared. This suggests that SMN protein plays a critical role in very early brain development, possibly before any treatment can be delivered, and that current therapies may not fully compensate for the protein deficit in brain tissue.
Current SMA treatments work in different ways and reach the brain to varying degrees. One treatment is delivered directly into the spinal fluid, targeting the central nervous system. Another, an oral medication called risdiplam, was specifically designed to cross the blood-brain barrier. It achieves this through high passive permeability and by avoiding a transport protein that normally blocks drugs from entering the brain. This means it reaches both peripheral tissues and the central nervous system. A third option, gene therapy, is delivered intravenously and can reach the brain through the bloodstream. Despite this access, none of these treatments fully eliminate the cognitive gap in the most severely affected patients.
How SMA Compares to ALS
Both SMA and ALS are motor neuron diseases, but their effects on the brain are quite different. ALS frequently involves cognitive and behavioral changes, with a significant proportion of patients developing symptoms that overlap with frontotemporal dementia. SMA does not carry this risk. When SMA and ALS patients were directly compared on cognitive testing, SMA patients performed better in memory, language, and executive function. The cognitive involvement in SMA, where it exists, tends to be mild and limited to specific domains like verbal fluency and working memory, rather than the progressive, broad cognitive decline seen in many ALS patients.
What Happens at the Cellular Level
Research in animal models has revealed a specific mechanism that may explain some of SMA’s effects on neural circuits. In healthy motor neurons, there’s a careful balance between excitatory signals (which tell the neuron to fire) and inhibitory signals (which tell it to stay quiet). In SMA, this balance breaks down in a counterintuitive way. First, the excitatory input to motor neurons drops because of impaired signaling from sensory neurons. You might expect the system to compensate by reducing inhibition. Instead, the opposite happens: inhibitory connections actually increase, piling on an excessive braking force that makes it even harder for motor neurons to fire and trigger muscle contractions. This maladaptive response, rather than a simple loss of neurons, contributes significantly to motor dysfunction in SMA and illustrates how the disease disrupts neural circuitry in complex ways beyond just killing cells.