Trial By Error: So FND Is Not Only a Brain “Software” Issue After All…

By David Tuller, DrPH

For decades, experts in functional neurological disorder (FND) have categorically asserted that the condition is strictly a “brain network” disorder, representing an issue solely with the brain’s so-called “software,” or functioning, rather than its “hardware,” or structural elements. According to this framework, structural elements of the brain, or other organic or pathophysiological processes, played no role in causing the condition.

The well-known FND website, neurosymptoms.org, has long popularized this notion—and continues to do so. The site currently proclaims on its home page that FND is “caused by a PROBLEM with the FUNCTIONING of the nervous system” and that it is “a ‘software’ issue of the brain, not the hardware (as in stroke or MS).” From this perspective, the “hardware” is irrelevant.

Given this view, anyone suggesting that the etiology of FND might involve something other than “software” problems or malfunctioning brain networks has been routinely dismissed as engaging in “dualistic” thinking. As I understand it, the purported “dualism” involved the belief that somatic symptoms were likely caused by unidentified pathophysiological dysfunctions, not just aberrant brain “software.” This accusation of “dualism” always struck me as backwards, since it was the FND proponents who themselves advanced the rigid “software” vs “hardware” dichotomy. Isn’t it “dualistic” to argue for a complete split between “software” and “hardware”? Or am I the confused one here? (I assume the FND folks would say yes.)

Whatever. In the introduction to a recent collection of papers on FND in the journal NeuroImage: Clinical, the authors flatly acknowledge that the “software”-only hypothesis is dead. (I’m not they would express themselves that way—but the admission is obvious to those outside their bubble.) That paper, called “Brain imaging research is thriving in functional neurological disorder,” offers this conclusion:

“Observations continue to support that FND is in part a problem of alterations within and across multiple brain networks…The data for FND representing a “software” and “hardware” problem of the brain is also growing.” 

The evidence of “hardware” involvement has been growing for years alongside the “software”-only hypothesis. But you don’t need to be a neuroscientist, or even a native English speaker, to know that this new paper’s conclusion is not remotely compatible with the persistent claims over many years, on neurosymptoms.org and elsewhere, that FND can be completely attributed to “software” and brain network problems. To argue now that FND is merely “in part” a brain network issue–and how big a part is that, anyway?–and to note that “hardware” also likely plays a central role in the condition is a pretty sharp turnaround from the past.

I mean, either “hardware” is involved, or it’s not. Both can’t be true. And the FND field seems pretty confused about which one it is—despite the history of definitive claims regarding the exclusive role of “software” problems and dismissive attitudes towards those challenging that view. (As a sign of the confusion, neurosymptoms.org includes a separate section on the emerging evidence of “hardware” changes, but for unexplained reasons has not altered its conclusions on its home page about the exclusively “software” nature of FND.)

Let’s be clear. This “software” and “hardware” proposition marks a huge change in the framing of FND. For many years, we have been told ad nauseam that FND was solely “a problem of alterations within and across multiple brain networks.” There was no “disease.” The only issue was “software” that needed fine-tuning and retraining with treatments like cognitive behavior therapy and physical rehabilitation. Given that, FND experts have struggled to make sense of the emerging reports of structural changes as well, suggesting that they might be downstream consequences of the “software” issues.

Now we have a clear acknowledgement that the “software”-only hypothesis is no longer operative. That raises a critical question: If “hardware” or other pathophysiological factors are potentially at play in causing the devastating symptoms that characterize FND, what makes the condition “functional” in the first place?

In medicine, “functional” has become popular as a seemingly neutral replacement for words like “psychosomatic” and “psychogenic,” which patients perceive as insulting. “Functional” illnesses have also frequently been called “medically unexplained symptoms”—meaning that they resist ready biomedical explanations, whether because of their very nature or our current limited state of knowledge. If it now appears that biomedical or “hardware” factors are part of FND’s causal pathway, then “functional” is likely a misnomer.

The generally favored treatments for FND and other so-called “functional” conditions have been psychological and behavioral interventions. In FND, these treatments have been designed to target the purported brain network issues said to be causing it. Unfortunately, major randomized clinical trials, including for functional seizures and functional motor disorders, have yielded poor results for their primary outcomes, raising self-evident questions about the rationale for the interventions.

These disappointing findings do not appear to have shaken the confidence of FND experts in the correctness of their strategies or led to greater humility in their pronouncements. Instead, they have generally explained away these disappointing results by citing personal clinical experience and engaging in contorted reasoning. But perhaps these interventions failed because, as the most recent paper notes, FND is only “in part” a brain network disorder. 

One of the problems with the longstanding FND claims involves a simple fact: brain scan research is still in its infancy. Functional magnetic resonance imaging (fMRI) documents brain activity, but the findings represent  associations only; interpreting them as causal, as FND experts routinely have, is fraught with peril. After all, everything we do involves differential brain networks; the phenomenon is not somehow limited to people experiencing FND. Causal claims derived from these scans are speculative and theoretical and should not be presented as fact. Maybe all the “software” issues are being caused by the “hardware” ones. Seems like no one much knows at this point. The issue is up for grabs.

Anyone (like me) who has challenged the “software”-only construct has been harshly criticized by FND proponents for…well, I’m not quite sure. I have pointed out serious weaknesses and logical flaws in FND studies and have suggested multiple times that “hardware,” organic or other pathophysiological factors could play a causal role. I’m glad to see the FND field has–finally–reached the same conclusion. Too bad it’s taken so long.