📝 Functional Neurological Disorder (FND) — Research Briefing

1. What FND is — the modern consensus

Definition (Edwards & Stone, accepted by the FND Society since 2019): Motor and/or sensory symptoms arising from the voluntary nervous system but experienced as involuntary. The hallmark is NOT "no lesion, therefore psychogenic" — it's a breakdown in how the brain predicts, attributes agency to, and integrates sensorimotor signals.

The paradigm shift: FND moved from diagnosis of exclusion ("nothing else fits, so it's psychogenic") to rule-in diagnosis based on 60+ positive clinical signs across subtypes:

• Hoover's sign for functional leg weakness — weakness of hip extension that normalises with contralateral hip flexion
• Tremor entrainment for functional tremor — tremor re-paces when the patient voluntarily moves another limb at a set rhythm
• Variable responsiveness during functional (non-epileptic) seizures
• Paradoxical memory performance in functional cognitive disorder

This is arguably the most important change in the field in the last 20 years. The presence of multiple convergent signs (e.g. entrainment + distractibility + coactivation in tremor) increases diagnostic confidence; the diagnosis should be made by a neurologist on positive evidence, not by ruling out other conditions.

Mechanism (current best theory): Predictive processing / Bayesian active inference. The brain works as a statistical organ making predictions about self and world; in FND, priors dominate sensory evidence. Erroneous predictions about movement, sensation, and agency get encoded as "real" because that's how the hierarchy works — top-down expectations override bottom-up input. This framework converges with neuroimaging findings of disrupted networks linking arousal, interoception, salience attribution, affective-motor integration, and self-agency (insula, cingulate, amygdala, TPJ, SMA, stria terminalis).

2. Epidemiology — what's solid vs. oversold

Solid:

• Sex ratio ~3:1 female (~75% of cases), onset typically 30s–50s. One of the most gender-imbalanced neuropsychiatric disorders.
• ~16% of general neurology clinic referrals — Scottish Neurological Symptoms Study.
• ~20–30% of FND patients also have a structural neurological diagnosis (epilepsy, MS, migraine). FND comorbidity is ~20–30% in epilepsy; ~5.8% of MS patients have comorbid FND.
• Misdiagnosis rate is ~0.4% over 18 months in Stone's 1,030-patient cohort. ~4% in older systematic reviews going back to the 1950s — stable since 1970.
• Pediatric FND is 10–20% of children's-hospital neurology presentations and up to 20% of adolescent intractable-seizure admissions.
• Disability comparable to MS or Parkinson's in quality-of-life and long-term function studies.

Oversold:

• "#2 after headaches" — widely repeated, but the SNSS data that's usually cited placed functional/conversion symptoms much lower. FND experts have publicly agreed to correct the inflated version. FND is very common but not #2.

3. Mechanism deep-dive

Three interacting layers:

1. HPA-axis dysregulation — chronic early unpredictability wires the stress-response system toward hyper- or hypo-reactivity. This also explains the high overlap between FND and autoimmune conditions.

2. Network-level brain changes — subtle but reproducible at the group level:
   - Altered white matter in stria terminalis/fornix (the principal output pathway of the amygdala and hippocampus)
   - Cortical thickness changes in emotion-processing regions
   - TPJ / SMA / insula functional connectivity abnormalities
   - Important caveat: group-level only — individual MRIs still read as normal. The approach uses research methods not validated for clinical use.

3. Predictive-processing breakdown — aberrant priors + attenuated sensory prediction errors → symptoms experienced as involuntary despite arising from the voluntary motor system. This is the bridge theory connecting trauma, attention, and the lived symptom phenomenology. It also predicts the most useful therapeutic finding in the field: attention makes it worse.

4. Triggers — the 3P model

• Predisposing factors: Early adverse experiences (attachment disruption especially), childhood exposure to medical illness, female sex, alexithymia / emotion-regulation deficits.
• Precipitating factors: Psychological trauma, physical injury, infection (including COVID), migraine, surgery, panic attack — anything that acutely shifts the prediction machinery.
• Perpetuating factors: Iatrogenic harm (disbelief, "nothing wrong with you"), fear-driven avoidance, protective deconditioning, social withdrawal, continued attention to the symptomatic limb.

All three operate largely outside conscious awareness — which is why a patient may genuinely not recall a precipitating event, and why "trauma" can't be treated as the necessary cause.

5. Long COVID — the live controversy

Viral illness including COVID is a documented FND precipitant. An international EHR cohort showed post-COVID FND incidence at ~1.5x that of other respiratory infections (0.033% vs 0.021% in the early pandemic; 0.038% vs 0.027% subsequently).

However, leading FND researchers have published strongly that Long COVID ≠ FND and that mis-labelling Long COVID as FND delays correct diagnosis and management. The motor and balance symptoms of Long COVID do problematically overlap with FND diagnostic criteria, but the pathophysiologic mechanisms are argued to be distinct.

The clinical takeaway: Hold both facts simultaneously. COVID can trigger FND in a susceptible minority, AND Long COVID as a population-level condition has its own pathophysiology that is not FND. Getting this wrong in either direction harms patients.

6. Treatment — what the evidence actually says

Gold standard: multidisciplinary. Specialist physiotherapy + psychotherapy, anchored by a well-delivered diagnosis.

Intervention	Evidence strength	Key finding
Explanation / diagnosis done well	Multiple studies	Most impactful, lowest-cost intervention. A substantial fraction of functional seizure patients improve from the diagnostic conversation alone.
Specialist physiotherapy (Nielsen/Stone protocol)	Phase-3 RCT (Physio4FMD, n=355)	Effective for selected patients with functional motor disorder. Exploits the "worse with attention" phenomenon — e.g. distract during gait → normal gait emerges.
CBT (CODES trial, n=368 — largest FND trial ever)	RCT	At 12 months CBT did not reduce seizure frequency vs. standard care. BUT: improved QoL, psychosocial functioning, psychological distress, somatic symptoms, and dissociative-seizure-free periods. A nuanced positive result — and an important lesson that seizure frequency may be the wrong primary endpoint.
Combined PT + CBT	RCT (JAMA Neurology)	Growing evidence; interdisciplinary outperforms single-modality.
TMS (transcranial magnetic stimulation)	Small trials	In a 33-patient FMD sample, ~66% showed >50% improvement. Mechanism likely cognitive/behavioural ("the machine is treating my brain, so I can move") rather than direct cortical neuromodulation. Still off-label.
Medication	No FDA approvals for FND	Useful for comorbid depression/anxiety/PTSD but not for FND symptoms directly.
Nocebo-hypothesis CBT	2025 pilot RCT	Emerging — explicitly reframes symptoms as "overlearned predictions" rather than real weakness.

No Class I evidence for any treatment of functional motor symptoms specifically. The field is young; trials are hard to blind; outcome definitions are contested.

The single most important clinical insight: how the diagnosis is delivered is itself a therapeutic intervention. Standards of care now specify naming FND, showing the positive signs that led to the diagnosis, and partnering with the patient. The old "good news, nothing's wrong with you" reassurance perpetuates the disorder.

7. Prognosis — realistic numbers

Adults (honest picture): Prognosis isn't great in aggregate, but it's better than the old "conversion disorder is untreatable" lore.

• At 1-year follow-up with standard neurology care: ~40% report improvement, often without specific treatment.
• At 14 years post-diagnosis: ~80% still have limb symptoms.
• With multidisciplinary rehab: ~70% good long-term recovery at median 13-month follow-up.

Pediatric (much more favorable): Remission 43–81%, symptom improvement 71–100%, full resolution 75–95% in treated cohorts. Younger age is a consistent good-prognosis predictor.

Good-prognosis predictors:

• Early diagnosis
• Younger age
• Absence of comorbid psychiatric disorder
• Absence of comorbid epilepsy or chronic illness
• Symptom duration < 2 years at diagnostic conversation

Poor-prognosis predictors:

• Chronic symptoms at presentation
• Fear-avoidance behaviours
• Unresolved mental-health comorbidities
• Multiple functional somatic symptoms

Implication: the gap between symptom onset and correct diagnosis is itself a modifiable prognostic variable. Early-diagnosis pathways, especially in paediatrics, are the highest-leverage systems-level intervention.

8. Where the field is going

• Biomarkers — machine-learning classification from structural MRI is now published but not clinic-ready.
• Subtype-specific trials — moving away from lumping all FND together. Functional tremor, functional gait, and functional seizures may need different protocols.
• Remote CBT delivery — pandemic-era informal work is being formalised.
• Neurostimulation — TMS, tDCS under active investigation, mechanism debated.
• Pediatric early-intervention models — Mass General, UCD, Stanford programmes; given the far better paediatric outcomes, early pathways are a priority.
• Untangling from Long COVID and ME/CFS — politically charged and clinically consequential.

9. What matters most for a practitioner

1. How you deliver the diagnosis is treatment. Naming FND, showing the positive signs, partnering with the patient — this alone produces measurable symptom improvement in a meaningful fraction of cases.
2. "Worse with attention" is the core therapeutic lever. Distraction-based PT exploits it; predictive-processing theory explains it. Every effective rehab protocol in the field uses this in some form.
3. Iatrogenic harm is a first-order problem, not a soft-skills concern. "Good news, there's nothing wrong with you" feedback is a perpetuating factor and arguably the single biggest correctable harm in the FND care pathway.
4. Duration predicts difficulty — so the gap between symptom onset and correct diagnosis is itself a modifiable target.
5. Kids do much better than adults. Early-pathway design is the highest-leverage place to intervene at a system level.

10. Sources

The three source videos (each with their own fact-check table in the Franki Research Library):

• Case: FND and POTS — Dr. Nathan Keiser — chiropractic neurologist case study
• What Causes FND? — Dr. Karen Sullivan — neuropsychologist overview
• What is FND? — Zara Beth — patient perspective

Plus the peer-reviewed and authoritative sources listed in the Related Resources table above.