Is FND falsifiable?

23 Aug 2025 Christoff Trüter Spousal caregiver

0

Karl Popper's criterion of falsifiability, first presented in Logik der Forschung (1934) and later translated into English as The Logic of Scientific Discovery (1959), is one of the most influential frameworks for separating science from pseudoscience.

For example, imagine I claim that only white swans exist. The moment someone spots a black swan, the claim is proven false. Even if black swans or other colours did not exist, swans are at least real and observable, so the claim can be tested, making it falsifiable. In contrast, if I claim there is an invisible unicorn living in my garden, the only way to disprove me would be to see it, yet by definition you cannot, so the claim is unfalsifiable and rests entirely on trust.

When discussing the diagnosis of FND, given the claims that it is no longer considered to be medically unexplained (MUS), surely that means we can trust the diagnosis, we are dealing with something that is definitely falsifiable right?

However, the diagnosis relies on subjective signs such as Hoover's sign and non-specific fMRI findings that may simply reflect unidentified organic causes. In the absence of definitive biomarkers, it is less reliable than pursuing testable organic explanations. The attribution of such non-specific fMRI findings to a diagnosis of this nature is unclear, making confirmation bias the most likely explanation, and presenting them as evidence for FND is an overreach. It also remains uncertain when, or on what basis, the condition ceased to be regarded as a MUS diagnosis (I suspect from a constructivist perspective), as it clearly remains speculative.

From a realist perspective, FND is still medically unexplained (MUS). MUS, by definition, cannot be falsified because it is a statement of ignorance, not a testable claim. Its falsifiability is therefore inherently problematic. Yet FND is often presented as a concrete disorder. Since the diagnosis relies on indirect evidence and subjective clinical interpretation rather than objective markers, it cannot be considered a confirmed pathology and remains, effectively, in invisible-unicorn territory.

Even when organic causes are identified, the FND diagnosis is seldom overturned, as the DSM-5 itself notes. Such findings are typically dismissed as comorbid, and the diagnosis, being based on opinion, is rarely reconsidered. In practice, one has to rely on the clinician’s word, since it cannot be seen, measured, or tested, and caveats about signs such as Hoover’s are often ignored rather than prompting a revision of the diagnosis

For example, Hoover's sign has only moderate sensitivity (63% in the best-designed prospective study) despite being reported elsewhere as highly sensitive and specific, but these figures come from very small, biased samples. It can also be positive in other conditions like apraxia, cortical neglect, pain-related weakness, and multiple sclerosis, meaning it is not unique to FND. Bilateral weakness further reduces its utility, and the sign’s reliability is highly dependent on examiner technique and context. Despite these caveats, it is often treated as if it were definitive for FND. Stone J, et al. Hoover’s sign for the diagnosis of functional weakness: a prospective unblinded cohort study in patients with suspected stroke. J Psychosom Res. 2011;71(6):384–386. researchgate.net/publication/51835381 Espay AJ, et al. Assessment of functional neurological disorders. Lancet Neurol. 2018;17(3):195–210. pmc.ncbi.nlm.nih.gov/articles/PMC7328420/ Trial By Error: Questions About Hoover’s Sign. Virology Blog. 2023. virology.ws/2023/05/22/… Hoover’s Sign (Neurological). Physio-Pedia. physio-pedia.com/Hoovers_Sign_%28Neurological%29

Given all of these points, I do not believe that FND is strictly falsifiable according to Popper's criteria.

But that said, having a hypothesis does not automatically make something pseudoscientific. Even claiming that it is a philosophical constructivist invention, as discussed in a previous post, does not automatically make it quackery. Constructs can be useful, and theories help advance medicine.

However, as soon as we treat a hypothesis as a fact, we tread dangerously close to hocus-pocus. It becomes extremely risky when not knowing what is wrong is treated as knowing. In our ignorance, we act as if we are confident, almost clairvoyant, telling patients that they are safe and that their symptoms are temporary and will resolve with time.

Doctors must stop making baseless claims that FND is no longer MUS simply because of diagnostic authority bias and should keep diagnostic doors open by rigorously investigating the unknown.