In 2006, Parkinson's expert William Langston published his article, Parkinsonism: Tip of the Iceberg, framing PD as a divese brain (and autonomic) disorder.
Three new articles expand on today's "somatocognitive" framework.
Medscape here. Lead article in Nature here. News article in Nature here.
###
AI CORNER
###
Here is a standalone blog-style article.
The Somato-Cognitive Theory of Parkinson’s Disease
For most of the modern era, Parkinson’s disease (PD) has been framed primarily as a movement disorder: tremor, rigidity, bradykinesia, gait difficulty, and impaired movement initiation. That description is not wrong, but it has long been incomplete. PD also includes constipation, blood pressure dysregulation, REM sleep behavior disorder, apathy, executive slowing, and other non-motor symptoms, many of which can appear years before the classic motor syndrome. That mismatch has always created a conceptual problem: why should a “movement disorder” so consistently involve autonomic, sleep, motivational, and cognitive domains?
A new answer has emerged from recent network neuroimaging. In this view, PD is not best understood as damage to isolated limb-movement regions, but as dysfunction of a broader integrative circuit called the somato-cognitive action network, or SCAN. The core claim is that PD is a disorder of the neural machinery that links intention, arousal, autonomic state, and whole-body action. In this framing, the motor syndrome remains central, but it is no longer the whole story; it becomes the most visible expression of a wider circuit disorder.
What is SCAN?
The SCAN concept grew out of a revision of the classic Penfield picture of motor cortex. For decades, the cortical motor strip was taught as a fairly orderly map of body-part “effectors”: foot, hand, mouth, and so on. More recent precision functional imaging suggested that this account was incomplete. Between the classic effector zones are inter-effector regions that do not seem to represent a single body part. Instead, they connect with systems involved in goal-directed behavior, arousal, autonomic control, and whole-body coordination. These inter-effector regions, taken together, form the SCAN.
That matters because PD is conspicuously not an effector-specific disease. It does not merely weaken one limb program or one focal motor output. It alters gait, posture, initiation, vigor, coordination, autonomic tone, and the tempo of voluntary behavior. The 2026 Nature paper explicitly argues that the SCAN is well positioned to explain this broader phenotype because it is thought to coordinate arousal, organ physiology, whole-body motor plans, and behavioral motivation.
The central mechanistic claim: hyperconnectivity
The strongest empirical claim in the new model is not merely anatomical but physiological. In a large multimodal dataset of 863 individuals across 11 cohorts, investigators found that the major subcortical nodes implicated in PD — including the substantia nigra, subthalamic nucleus, globus pallidus, ventral intermediate/centromedian thalamus, GPe, and putamen — are functionally connected more strongly to the SCAN than to the classic effector-specific motor regions. In PD, the connectivity between SCAN and these subcortical nodes was abnormally elevated, a pattern described as hyperconnectivity.
The argument is therefore not simply that the SCAN exists and PD patients have symptoms consistent with it. It is that PD shows a specific network abnormality: excessive SCAN–subcortical coupling. Importantly, the study reports that this hyperconnectivity was specific to the SCAN rather than to effector motor cortex, and that it correlated with symptom severity. The Nature research briefing also notes that this abnormal coupling was not seen in the same way in comparator disorders such as dystonia and essential tremor.
This is the pivot from description to theory. The somato-cognitive view says that PD symptoms arise because the circuit that should flexibly translate motivation and bodily state into action becomes pathologically over-coupled with subcortical nodes, effectively jamming normal voluntary control. That formulation is still a model, not a settled fact, but it is more specific than the older broad statement that PD “involves more than dopamine.”
Why this theory feels different from older “PD is more than motor” arguments
A key point is that the general idea that PD extends beyond movement is not new. Langston’s 2006 essay, The Parkinson’s Complex: Parkinsonism Is Just the Tip of the Iceberg, already argued that parkinsonism represented only the clinically obvious portion of a broader disease process. He emphasized early and non-motor features such as constipation, hyposmia, REM sleep behavior disorder, daytime sleepiness, and autonomic pathology, and argued that these were not merely “risk markers” or “preclinical” hints but parts of the disease itself. He drew on neuropathology showing Lewy-related involvement in lower brainstem, olfactory structures, sympathetic ganglia, and myenteric plexus, and he criticized an overly nigrostriatal view of PD.
In that sense, the somato-cognitive theory is partly a modern network realization of an older clinical intuition. Langston’s article was already saying, in effect, that the textbook motor syndrome badly under-described the real disease. What is new in 2026 is that this older “iceberg” idea is no longer only phenomenological or pathological. It is being reformulated as a specific systems-neuroscience model with candidate nodes, measurable connectivity, and therapeutic implications.
Why the theory is attractive
The SCAN framework is attractive for several reasons.
First, it offers a single organizing principle for the coexistence of motor and non-motor symptoms. If one network participates in whole-body action, autonomic regulation, arousal, and behavioral motivation, then constipation, orthostatic symptoms, REM sleep behavior disorder, gait freezing, bradykinesia, and apathy need not look like unrelated add-ons. They can be read as different surfaces of one disturbed action-regulation system.
Second, it helps explain why such different treatments can all work. Levodopa, deep brain stimulation, focused ultrasound, and TMS are mechanistically quite different interventions. Yet the 2026 work argues that effective versions of all of them share a common network effect: they reduce SCAN–subcortical hyperconnectivity. That is a powerful unifying claim. It suggests that one may be seeing not four unrelated therapeutic stories, but several ways of reaching the same dysfunctional circuit.
Third, the model may improve neuromodulation targeting. The paper reports that approved DBS targets were all more strongly connected to the SCAN than to effector motor cortex, that lesion proximity to a thalamic SCAN “sweet spot” predicted better focused-ultrasound outcome, and that in a TMS study, targeting the SCAN rather than effector motor regions produced roughly double the improvement on a standard motor scale.
The deeper conceptual shift: from movement to action
The most interesting intellectual move in this theory is linguistic as much as anatomical. The investigators and commentators increasingly describe PD as an action disorder, not merely a movement disorder. That sounds semantic, but it is actually substantive. “Movement” implies output. “Action” implies the coupling of goal, internal state, autonomic readiness, cognitive set, and motor implementation. In ordinary clinical life, PD already behaves more like an action disorder than a mere movement disorder: performance worsens under cognitive load, improves with external cueing or music, and can sometimes transiently normalize in emergencies. Those phenomena fit awkwardly with a simple effector-deficit model but fit more naturally with a disorder of action selection and implementation in context.
This may also explain why the theory resonates with longstanding clinical observations without claiming that everything in PD is “cognitive.” The phrase somato-cognitive is important because it does not collapse the disease into a psychology-of-motivation narrative. It points instead to a circuit where bodily state and cognitive intent are integrated. In other words, the claim is not that PD is secretly a cognitive disorder masquerading as a motor one. The claim is that action itself is built from both somatic and cognitive ingredients, and the affected network sits precisely at that junction.
What, exactly, is new in the 2026 work?
Three things stand out.
One is scale. The Nature study did not rely on a single small imaging cohort; it combined multiple cohorts and multiple intervention modalities, creating a broader evidentiary base than many prior PD network papers.
Another is convergence across therapies. It is one thing to show a disease-associated imaging abnormality. It is more persuasive to show that when different effective treatments work, they all move that abnormality in the same direction. The SCAN hypothesis gains force because levodopa, DBS, TMS, and MR-guided focused ultrasound were all interpreted through the same circuit lens.
The third is translation. The paper does not stop at explanation; it proposes that functionally defined SCAN targets might improve current therapies and perhaps support new minimally invasive cortical stimulation strategies. It even raises the possibility of personalized cortical or epidural/subdural targeting after SCAN mapping.
How this fits with older pathology-based models
The somato-cognitive theory does not replace the classic neuropathology of PD. It sits on top of it. The 2026 paper still begins from the accepted hallmark of dopaminergic neuron degeneration in the substantia nigra and dysfunction of the broader cortico–basal ganglia–thalamic circuit. Its innovation is to say that the cortical component of that circuit is better described by SCAN than by generic “primary motor cortex” or by a foot-hand-mouth effector map.
Likewise, it does not negate Braak-style or Langston-style accounts of early involvement of olfactory, autonomic, and lower brainstem structures. Rather, it offers a new cortical-systems account of why the syndrome looks so distributed in life. Langston’s 2006 argument that PD was a far broader disorder than visible parkinsonism now finds a plausible network companion: the brain system coordinating body physiology with goal-directed action may be involved more centrally than older models assumed.
The theory’s current limits
For all its appeal, the somato-cognitive theory is not settled doctrine.
A first limitation is that the clinical non-motor story remains more inferred than demonstrated. The research briefing itself says that investigation of sleep, autonomic, cognitive, and affective symptoms remains limited and is an important next step. In other words, the theory is highly plausible for these domains, but the current evidence is strongest for the motor and neuromodulation findings.
A second limitation is reproducibility and generalizability. Outside experts quoted in Medscape called the work promising but early, emphasizing the need for replication across sites, scanners, acquisition pipelines, and academic centers.
A third limitation is biomarker specificity. The Medscape discussion notes skepticism that SCAN hyperconnectivity can yet cleanly distinguish PD from all competing disorders in real-world practice. Imaging biomarkers in movement disorders face a very high bar because diagnostic overlap is common, especially early in disease.
A fourth is that even if the imaging is robust, it is primarily a model of circuit dysfunction and symptomatic treatment, not yet a model of disease modification. Reducing hyperconnectivity may improve symptoms without necessarily arresting alpha-synuclein biology or neurodegeneration.
And a fifth is methodological: the TMS result, while intriguing, reportedly lacked a sham control arm, which weakens causal confidence.
So how should the theory be judged right now?
The most balanced reading is that the somato-cognitive theory is not a replacement for everything known about PD, but it may be the most coherent recent attempt to unify several stubborn facts:
PD is clinically broader than a motor syndrome.
The cortex involved in PD may not be best described by classic effector motor maps.
Different successful treatments may converge on a common dysfunctional network.
Future neuromodulation may improve if targeted to that network rather than to generic motor cortex.
That does not yet make SCAN the final answer to PD. But it does give the field a sharper hypothesis than the old fallback phrase that “Parkinson’s has motor and non-motor symptoms.” The new theory says more: those features may belong together because the disease disrupts a circuit whose normal job is to join somatic state, cognitive set, and whole-body action.
Bottom line
Twenty years ago, Langston argued that the field had been looking at PD through the “somewhat myopic lens” of visible parkinsonism and missing the much larger disease iceberg beneath it. The 2026 SCAN work can be read as a sophisticated update of that intuition. The older literature established that PD was broader. The new literature tries to show where and how that breadth is organized in brain circuitry.
Whether this theory becomes foundational will depend on replication, better linkage to non-motor syndromes, and proof that SCAN-guided interventions improve outcomes in larger controlled trials. But as a conceptual advance, it is already important. It shifts PD from being thought of mainly as a disorder of movement output to being considered, more fundamentally, a disorder of embodied action.
I can also turn this into a shorter DIHP-style blog version in your usual paragraph format with bold lead-ins.
