Nocebo information increases fatigue and urge to stop
People with chronic illness live in a dense information environment. They receive guidance from clinicians, read accounts in online communities, watch videos, and absorb a continuous stream of communications about their conditions — what activities will worsen symptoms, what triggers to avoid, what the prognosis looks like. Most of this communication is treated as neutral information transfer: facts going in, understanding coming out. A 2021 paper by Lenaert and colleagues in Brain, Behavior, and Immunity provides experimental evidence that it is not neutral. The way information about symptoms is framed — specifically whether it primes negative expectations — changes the physiological experience of the symptoms themselves. Language about symptoms is not metadata about experience. It is an input to the system that produces the experience.
The Experiment: One Variable, One Difference
Lenaert and colleagues designed a straightforward experiment with important implications. Participants performed a standardized physical task. Before doing so, one group received instructions that described the task as fatigue-inducing — framing the expected experience in negative terms. The other group received neutral instructions describing the same task without expectation-setting language about fatigue. Everything else was identical: the task, the physical demands, the duration, the participants. The only variable was what the participants were told to expect.
The group primed with fatigue-expectation instructions reported higher fatigue during the task and showed a stronger urge to stop compared to the neutral instruction group. They were performing the same physical activity as the neutral group. Their muscles were doing the same work. Their cardiovascular systems faced the same load. The task was identical. What was different was the brain's model of what the experience should feel like, and that model shaped how the experience was actually felt. The information changed the expectation. The expectation changed the physiology.
The Predictive Processing Mechanism
The finding makes mechanistic sense within the predictive processing framework of brain function. The brain does not passively receive signals from the body and translate them into experience. It actively generates predictions about what internal body state should be given current behavioral context, and compares incoming signals to those predictions. Perception is the product of this comparison: a blend of what the brain predicted and what the body is actually signaling, weighted by the brain's confidence in each. This predictive architecture is central to understanding why autonomic conditions involve the brain's modeling of body state, not just peripheral hardware.
In the Lenaert experiment, the fatigue-priming instruction shifted the brain's prediction about what the task should feel like. The brain entered the activity with a model that said: this will be fatiguing. Sensory signals arriving from the body during the activity were then interpreted against that model. Ambiguous signals that might have been read as manageable exertion in the neutral condition were read as significant fatigue in the primed condition, because the model expected fatigue and provided that interpretation for incoming data. The prediction shaped the perception.
How Effort Regulation Translates Information Into Behavior
Beyond perception, the Lenaert findings also documented a behavioral effect: the stronger urge to stop in the fatigue-primed group. This connects to a second mechanism — effort regulation. The decision to continue or discontinue a physical or cognitive task is not made by the muscles or the heart. It is made by the brain, based on a comparison between the perceived current cost of continuing and the perceived value of the goal. When the brain's model says the cost of continuing is high, the motivation to stop increases. When it says the cost is manageable, persistence is easier.
The fatigue-priming instruction elevated the perceived cost of the task before it began. When actual effort signals arrived from the body, they were added to an already-elevated predicted cost, producing a higher total perceived effort signal. The urge to stop is the behavioral output of a brain that has calculated the cost as too high to continue. Changing the information — and therefore the prior — changed the calculation. The physical cost of the task was identical in both groups. The perceived cost was not.
What This Means for Health Communications in Chronic Illness
The Lenaert mechanism operates anywhere that information about symptoms precedes the experience of symptoms. Clinical consultations where activity is described as dangerous. Online community posts that frame specific tasks as guaranteed to trigger crashes. Management guidelines that lead with warnings about what will worsen the condition. All of these communications provide prior expectations that the brain incorporates into its predictive model before the experience occurs. They are not neutral background information. They are inputs to the system that generates the symptom experience.
This applies with particular force in conditions where post-exertional malaise is a real physiological phenomenon that genuinely warrants careful management. The challenge is not between taking symptoms seriously and recognizing that language shapes experience. Both are simultaneously true. PEM is a real consequence of oxygen delivery failure during exertion in ME/CFS. It is also the case that how any given bout of exertion is framed before, during, and after it occurs shapes the brain's interpretation of the sensory signals from that exertion. The physiology and the nocebo effect coexist. They are not mutually exclusive.
Saunders and colleagues' 2024 meta-analysis established that observing other people's negative symptom experiences produces real nocebo effects in observers — the socially learned expectation is as capable of generating physiological changes as directly received information. Online illness communities where severe negative symptom narratives circulate continuously are generating the exact conditions documented in the Lenaert experiment at scale, across every member who reads and views those narratives.
Precision in Language Is Not Toxic Positivity
The implication of the Lenaert findings is sometimes misread as arguing for minimizing, denying, or positively reframing serious illness. That is not what the evidence supports. Being disbelieved and dismissed has real physiological consequences in ME/CFS — the nocebo effect operates in both directions, and clinical communications that minimize real pathology generate their own form of harm. The argument is not for positivity as a denial mechanism.
The argument is for precision. There is a difference between accurate clinical communication about real risks and fatigue-priming framing that generates nocebo effects by pre-activating the expected experience. Accurate information about what activities have caused worsening at what doses, delivered with specificity and without catastrophizing language, serves management. Sweeping framing that characterizes broad categories of activity as dangerous, without distinguishing dose and context, generates expectation effects beyond what the evidence warrants. The brain takes both kinds of communication as inputs to its predictive model. Precision about which information is being provided — and at what level of certainty — is not a soft concern. It is a clinical one.