Can the mind create pain?

How do you know if your pain is psychosomatic? Can our minds create pain, out of nothing? Can we have pain for psychological reasons alone? Does such pure psychosomatic pain exist — pain with no injury at all, no issue in the tissue, no noxious stimuli of any kind?

Can we “hallucinate” pain?

Short answer: probably not, as far as I’ve been able to determine. But I do not know, and I am not done trying. The mind can harm us in ways that lead to pain in time, and it might be able to amplify pain (or ease it), but so far I have be unable to find any clear evidence that the mind can actually generate pain from nothing, with no injury or pathology, no nociception. And there seem to be many reasons to suspect that it either isn’t possible at all, or it’s rather rare and mostly related to substantial mental illness.

The long answer awaits you below — and it’s more of a work-in-progress than usual, because I’m still trying to learn enough to know what I think. But I’m not exactly a beginner with this topic either, and there are 3000 words of useful details ahead.

Update: This article was originally published in early 2023 as a members-only post. It was freed from the paywall in late 2025 when I published a closely related blog post, “Pain needs nociception: 3 new papers.”

Filippo Balbi, Testa Anatomica, 1854 — Source (Wellcome Collection)

Please err on the side of assuming pain is never psychosomatic

Unexplained chronic pain is often inappropriately, prematurely chalked up to psychology, and often obnoxiously, driven by terrible reasons (like sexism), with terrible outcomes. “Patients often find themselves trapped in a zone between the worlds of medicine and psychiatry, with neither community taking full responsibility.” (O’Sullivan) In a great many cases, there is a biological cause that simply hasn’t been diagnosed yet.

In other words, most pain patients need better diagnosis, not fixes for a non-existent psychological issue. So we should avoid a diagnosis of psychosomatic pain like the plague. It should be one of the last things we think about, if we ever do.

But it’s really important in principle — and interesting — to know if it’s possible for the mind to generate pain out of nothing.

Can’t stop believin': the psychosomatic paradox

Amazingly, no one actually knows for sure if psychogenic pain is a genuine phenomenon. Certainly no one wants to believe that their own chronic pain is psychogenic. An “all in your head” diagnosis — or obnoxious subtext — has become a cliché of medical gaslighting.

Ironically, despite that hated phrase, almost everyone does believe that it is possible for pain to be all in your head. The whole idea of mindbody medicine is smash cultural hit. This is a peculiar paradox!

So what is it? Do we believe or don’t we? Can pain be “all” in our own heads… or just other people’s heads? And if we don’t think pain can be psychosomatic at all, then what’s with all the bestselling self-help books about mindbody medicine, the endless torrent of fear-mongering rhetoric about the evils of stress?

It’s bizarre how obsessed we are with a theory of health that every chronic pain patient also seems to loathe.

The scope of this article

This article is focused on the “all” in all-in-your-head. Psychology is obviously a factor in chronic pain, especially anxiety, stress, depression. Many kinds of pain have their roots in long-term health issues that have complex interactions with mental health and personality. There is also (relatively) little doubt that pain can be amplified by the mind, or at least complicated in ways that are almost indistinguishable from amplification. There is also no doubt that suffering and emotional distress can make the experience of pain way worse (see Suffering, Disability & Pain).

But this is about evidence and arguments for existence of pure psychogenic pain, occurring in the total absence of any pathology or trauma of any kind. I will only consider pain amplification and other kinds of psychosomatic illness only insofar as they inform this central question.

The psychosomatic pain glossary

• Sensation — The conversion of a physical stimuli into nerve impulses and then some kind of conscious awareness of that stimulus.
• Nociception — The conversion of noxious stimuli into nerve impulses, which may or may not lead to pain.
• Sensitization — Increased nociception relative to stimulus.
• Perception — The organization, interpretation, and conscious experience of sensations.
• Amplification — Pain that is out of proportion to a noxious stimulus due to perception. This is distinct from “amplified pain syndrome.”1
• Suffering — A negative emotional experience.

Proving psychogenic pain requires proving a negative

There is no reasonably direct evidence of pure psychogenic pain as far as I know. All we seem to have is speculative arguments, mostly based on inference and extrapolation from other phenomena. “If x is possible, then maybe psychosomatic pain is also possible…”

To be sure that pain is psychosomatic, we must also be sure that it has no undiagnosed biological cause. But we can only ever be “quite” sure of that, at best — and we are rarely at our best in this way. Most professionals are ignorant of the scope of the possibilities, and tend to underestimate them.2

There are countless hard-to-diagnose causes of pain! Much about the physiology of pain remains mysterious. Some chronic painful conditions rarely have known causes. For instance, a staggering percentage of back pain is “non-specific,” not attributable to any specific cause, and there’s been a running battle for decades now over how many of those cases have a biological cause or trigger that simply can’t be easily identified … versus a psychogenic cause. Is it 60-40? 30-70? We can never really know.

Separating psychosomatic pain from suffering and alarm

And there are other basic limits on our ability to know if the mind can create pain. It’s nearly impossible to tease apart pain and suffering, for instance. Pain is entirely subjective — it is what people say it is. But the intensity of pain that people report can be a product of either:

1. a more intense sensation
2. and/or its emotional significance.

Does someone say that a pain is a 6 out of 10 on the pain scale because the sensation is that strong? Or because they are freaked out? Or both? Probably both, but in what proportion? It’s different for every person and every problem, and there is never any way to know. (And, no, technology is very unlikely to solve this.3)

The puzzle of pain’s subjectivity: how can we know if pain is “too loud”?

Again, pain is subjective, and cannot be “wrong.” No one can tell anyone they aren’t in pain, or “shouldn’t” be. Technically, we cannot ever know that any one example of pain is “too loud,” because there is never a knowably “right” (objective) amount of pain for any situation, or even a fully knowable situation (there could always be a sneaky cause we cannot detect). This puzzle cannot be solved any more than we can ever really know if someone is seeing reds that are “too red.” The pure subjectivity of pain places an absolute limit on our ability to judge whether it is “amplified.”4

But don’t be fooled: that is a technical limitation we have ways of working around. It’s no different than what psychiatry has always had to grapple with: pathologies that obviously exist, and yet have only subjective symptoms. How do we diagnose those? How can we know that a patient’s experience is “wrong”? By how much it is at odds with reality as we understand it from other people, and by how much distress and trouble it causes. If someone is extremely tormented by perceptions of things that no one else can perceive at all, we call that mental illness, and no one worries too much about the fact that we can’t actually prove that it is — not at the extremes anyway. Life is all about these good-enough guesses.

Consider the example of electrosensitivity, a clear example of a painful mental illness: people who report great pain and suffering caused by something that cannot plausibly harm them. In any one case, we cannot technically prove that their pain is wrong. And yet we can have very high confidence that the electrosensitives cannot all be right: most of them must be reporting pain that has no biological cause.

Psychosomatic pain versus pain-without-nociception

Some readers will notice that question of the existence of psychosomatic pain is just another version of the debate about whether we can have “pain without nociception” — a staple of dorky shop talk on social media for the last few years. And argument. The claim that pain can exist without any signal about potential tissue threats is controversial at the very least.5

The concepts do overlap greatly, but there is one key difference: “psychosomatic” strongly implies mental illness, a pattern of lasting and consequential symptoms. Examples of pain-without-nociception are often more like “brain farts” — little isolated glitches in perception, more like being fooled by an optical illusion than a “hallucination” of pain.

Arguments for pain without nociception

In the absence of good evidence that pain can exist without nociception, there are various arguments, suggestive examples, and lines of indirect evidence instead. They are all interesting in their own right, but none of seem to be up to the task. They all have issues. I’ll summarize them here, and elaborate on some of them below.

• Psychological stress and anxiety are implicated as major factors in almost every imaginable kind of health problem, and the weight of those example is so great that it seems like a no-brainer that stress and anxiety must be able to cause pain as well. However, it’s clear the there’s usually or always an intermediate causal mechanism, such as muscle tension or an ulcer. In other words, we might hurt because stress and anxiety cause or aggravate a painful illness.
• Boot-nail-guy is the best example of these kinds of examples: a case study of a fellow who experienced great pain because he thought a nail had gone through his foot, but in fact it had gone through his toes. But we simply don’t know enough about exactly what happened. I’ve written in detail about Boot Nail Guy.
• Phantom limb pain seems like an example of pain that exists entirely in the mind, since the hurting tissue no longer exists. Except that it does: it’s the severed nerve endings in the stump that generate the nociception, and the brain just can’t work out where the signal is coming from. It’s clearly just an example of misperception of the location of the arm, and not pain without any nociception at all.6
• Nocebo is the well-documented phenomenon of fear-powered symptoms that seems to show that psychosomatic pain is possible. It may be one of the best arguments for psychosomatic pain, but it’s also not quite the slam dunk it looks like. More on this one below.
• The phenomenon of sensitization can be used to make the case that pain can be generated from nothing, especially the phenomenon like allodynia — pain from non-noxious stimuli. But it’s an illusion. The stimulus is non-noxious, but it’s being amplified by dysfunctional, highly sensitized nociception, which occurs in the aftermath of tissue insult. While there's room for (highly technical) debate about the role of nociception in allodynia, it’s widely accepted that it remains a nociceptive phenomenon.
• Central post-stroke pain is a promising candidate for pain that unambiguously comes from a “bad brain,” and definitely not peripheral nociception. So it partially and vividly meets some of the criteria for psychosomatic pain, and it can be considered “hallucinatory.” But that’s the brain, not the mind: pain that arises from brain damage (central neuropathy), and not psychology. So … still not “psychosomatic.”7
• The functional neurological disorders — people who suffer blindness, paralysis, or seizures — are one of the most vivid demonstrations of the power of the mind, but it’s also not a direct analogy: they don’t involve novel perceptions. More below.
• Hallucination happens, and this has led some people to suggest that if you can perceive other things without stimulus, you can probably hallucinate pain too. This is quite compelling on its face, but the deal-breaker is that other kinds of hallucination always have a biological cause: either pathology or drugs. People perceive things that aren’t there when their brain is impaired or altered… not because of their psychology. Hallucinations are a reason to consider looking for the phenomenon of hallucinated pain, not evidence that it exists, and there is no clear evidence that people can "hallucinate" pain, even in conditions which have been highly optimized in various ways to achieve exactly that effect. Even if pain can be hallucinated like other kinds of sensation, it would likely be just as exotically rare, and probably not clinically relevant to the vast majority of pain patients.
• Electromagnetic sensitivity is a real illness in which people report pain in response to something that cannot actually harm them, which might mean that their pain is psychosomatic. However, it might also mean that they all just have an undiagnosed pathology (e.g. fibromyalgia), and the EMS thing is just a hypothesis they’ve embraced to explain it. Once that hypothesis is embraced, the obsessiveness and paranoia about EMF very naturally follows — but it gets them labelled as mentally ill.
• The rubber-hand illusion is the illusion of sensation, especially pain, from stimulating a rubber hand, rather than your real hand. This phenomenon supposedly demonstrates that we can essentially hallucinate pain, feeling pain that we expect to feel — and perhaps we can. However, the science that supposedly demonstrated phenomenon in the first place has significant flaws and is controversial — in other words, the illusion is itself may be an illusion! More specifically, it may be more of a research artifact than a true illusion (a neurologically “hard-wired” glitch in perception).8
• Dream pain suggests that it might be possible in principle for pain to arise entirely from a brain state. However, this is a weak argument insofar as dreaming is a really weird state, and even if dream pain truly is a form of psychogenic pain, that does not tell us much about the clinical reality of waking pain. There are quite a few reasons to question its relevance to the question. It’s worth considering, but it’s hardly a slam-dunk for psychogenic pain either.9

Is the believer in pure psychosomatic pain a straw man?

Another objection that is often raised on this topic is that no one actually believes in entirely psychosomatic pain, a classic “straw man.” Is there actually a meaningful faction of professionals and experts who think that the mind can generate pain in the absence of any nociception? Don’t most professionals just believe that psychosocial factors modulate pain, not that it can be created from nothing?

Yes, that faction absolutely exists. No, they do not limit their claims to “modulation,” or at the very least they believe in very potent modulation, which is also modifiable. But many of them clearly also go further: they believe that perception, expectation, and predictive processing are so powerful that we can truly “hallucinate” pain. We have even hard data on related concepts: at least 80% of healthcare professionals believe that pain can be classically conditioned,10 despite a clear lack of evidence.

But I am in a good position to know this myself, because I hear from them. And they raise all of the above objections to defend their position, and they do so with passion. They are precisely why I am familiar with all of those points, and have responses to them: because I’ve been getting plenty of practice responding to their arguments over the last decade!

Is nocebo the example we’re looking for?

Placebo is “relief from belief,” and nocebo is the opposite: “grief from belief.” Just as placebo is widely believed to be a fascinating example of the power of the mind over the body, so too is nocebo thought of as a vivid example of a psychosomatic phenomenon. If nocebo is a thing (and that is widely accepted), then psychosomatic pain is probably a thing.

Unfortunately, it’s not a clean win. It’s not that nocebo doesn’t exist, it’s just not what seems like. Some complications:

1. The Legend of Placebo is that it is “powerful,”11 but it probably isn’t powerful — not in the psychosomatic sense, anyway. “Expectation effects” appear to be limited to perception and are quite modest.12 Much of what is called placebo is not an expectation effect at all, but merely the appearance of it in a research context.13
2. Just because placebo might be able to reduce the perception of existing pain doesn’t mean that nocebo can generate the perception of pain from nothing. Even if nocebo is a legitimate phenomenon, it might be limited to amplification of existing nociception, as opposed to causing the perception of pain in the absence of nociception.
3. Nocebo may cause an increase in alarm and drama about pain, rather than pain itself.

There are impressive examples of placebo that are hard to dismiss. My favourite is “beet guy,” a story in the excellent book Pain: The science and culture of why we hurt, about a man who was rescued from agony by learning that his bright red poop was caused not by internal bleeding but… beets!14 So where did all the “agony” come from? Certainly not the beets! Is this an open-and-shut case? No, because it’s so hard to distinguish pain from panic. And because he might have actually had simultaneous indigestion.

Indirectly psychogenic pain — from mind to [insert something] to pain

Tension headache is a common, minor example of how mental state can directly drive pain with no clear intermediate mechanism … and this kind of pain is often rounded up to “psychosomatic.” But there probably is a mechanism for all headaches, it’s just murky, and we don’t even know if it’s “tension.” This is likely the case for many painful conditions.

Holding your hand over a candle isn’t an example of psychogenic pain — but why not? It was your foolish mind’s idea! But no one in their right mind would say that it was your mind that cause the pain. It was the candle, obviously!

But what if you replace “candle” with “cramp”? What if your mind causes a painful muscle tension or spasm? This is one hypothetical mechanism for some kinds of pain (and just as scientifically mysterious and uncertain).15 In this case, the links between “mind” and “pain” are more unconscious and subtle. You might have no awareness of the intermediate muscle causes between your mind and your pain, which makes it seem a lot more like mind-powered pain.

Such ideas might partly explain how the mind can be involved in causing pain without actually generating it from scratch, and may also help to convince people that psychogenic pain is a legit phenomenon.

But they are clearly not examples of psychogenic pain, and more than stabbing yourself in the eye causes psychosomatic pain just because there were (strange!) psychological reasons for your self-destructive behaviour. For the pain itself, a biological mechanism is still required, and that mechanism is still nociception. The problem may have begun with a thought and ended with a perception, but the flesh still got involved in the middle — even if you couldn’t tell.

If we can paralyze ourselves, we can probably make ourselves hurt

Functional neurological disorders definitely exist — and they seem closely related to psychosomatic pain.

The FNDs, formerly known as “conversion” disorders, make it abundantly clear that the mind does have some crazy powers. We can experience seizures, paralysis, blindness, and other neurological symptoms in the absence of neurological disease.1617 Strange but true!

The FNDs that we know do not generally involve wonky perception, but wonky function — it’s right in the name. Nevertheless, if we can paralyze ourselves with our minds, it does seems like we might be capable of invent pain too. In fact, some chronic pain might actually be one of the members of the FND family, just really difficult to confirm, because pain can have so many other causes. In contrast, seizures, paralysis, and blindness have relatively short lists of possible causes to eliminate, leaving only the power of the mind to explain the problem. And so, once again, no one really knows.

Notably "an FND that causes only pain" is actually indistinguishable from psychosomatic pain — different labels for the same concept.

However, inference from the known types of FNDs might be a compelling clue that pure psychosomatic pain is possible.

Notes

1. The label "Amplified Pain Syndrome" refers mainly to a controversial diagnosis of psychosomatic pain in children. In this article, I’ve used "amplification" in a more general sense, and one that isn't widely used in the scientific literature. The concept is common, but it has no widely accepted formal term that I know of.
2. My favourite personal anecdote about this is the time I consulted a neurologist because I was having “thunderclap headaches.” I had just read a scientific paper about well over one hundred distinct pathological types of thunderclap headaches. But because I didn’t have a brain bleed — just the most common serious cause — my neurologist declared authoritatively that I was fine, and that there were only a couple of other causes of thunderclap headache, and that they weren’t dangerous. He was egregiously wrong about that … and I don’t even think he was a bad doctor! He was an otherwise good doctor who just didn’t know much about thunderclap headaches.
3. Can we have objective measurement of pain that would filter out the messy feelings? Someone is always trying to invent a pain-o-meter, but the concept is deeply flawed, as pseudoscientific and ripe for abuse as a lie detector. If there is trust and empathy in a clinical relationship, a pain-o-meter isn’t needed! If there is not trust and empathy, a pain-o-meter is just going to make things worse: it will be a generator of bullshit “objective” reasons to be dismissive. And even a hypothetically accurate pain-measuring technology would still inevitably fail to capture all kinds of important subjective “colour” to the experience of pain. See There will never be a pain-o-meter — and maybe that’s a good thing.
4. Weisman A, Quintner J, Masharawi Y. Amplified Pain Syndrome-An Insupportable Assumption. JAMA Pediatr. 2021 Mar. PubMed 33683311 ❐
5. Weisman A, Quintner J, Cohen M. Adieu to an aphorism: why nociception is necessary for pain. Brain. 2025 Oct:awaf387. PubMed 41091638 ❐
6. Ilfeld BM, Khatibi B, Maheshwari K, et al. Ambulatory continuous peripheral nerve blocks to treat postamputation phantom limb pain: a multicenter, randomized, quadruple-masked, placebo-controlled clinical trial. Pain. 2021 03;162(3):938–955. PubMed 33021563 ❐ PainSci Bibliography 51431 ❐
7. Treister AK, Hatch MN, Cramer SC, Chang EY. Demystifying Poststroke Pain: From Etiology to Treatment. PM R. 2017 Jan;9(1):63–75. PubMed 27317916 ❐ PainSci Bibliography 51204 ❐
8. Tsuji N, Imaizumi S. {Order Effects on the Rubber Hand Illusion Expectancy: A Replication and Extension of Lush (2020)}. Collabra: Psychology. 2024 04;10(1):116190. PainSci Bibliography 51506 ❐

   The “rubber hand illusion” is the alleged illusion of sensation, specifically pain, from stimulating a rubber hand, rather than your real hand. The idea is that it supposedly demonstrates that we can essentially hallucinate pain. This paper is somewhat technical, but the gist is that the experiments that supposedly “proved” the RHI was a thing in the first place may have Very Serious Flaws. In other words, the authors conclude that the illusion is itself an illusion! More specifically, that it’s more of a research artifact than a classical illusion, a neurologically “hard-wired” glitch in perception:

   “The methods in the RHI paradigm insufficiently control demand characteristics and order effects; thus, findings based on these methods are confounded by these effects.”

   We follow the evidence as best we can, but sometimes it’s like following a drunken squirrel. And there’s always someone who is happy to explain to you that you followed it wrong. And sometimes they are right!

   This paper replicates and extends a 2020 paper, Lush. For Lush’s commentary, see “What’s up with the rubber hand illusion?”

9. I have dreamed of being in pain plenty in my life! And I think it’s just as uninterpretable as everything else about dreams. It does little to persuade me that pain can occur without nociception, although I do concede that it’s relevant and interesting.

   The only kind of pain I dream of seems to be bleeding over from my waking pain, which is common. I am in pain when I go to bed, I dream of that pain, I wake up still in pain. Much, most, or perhaps even all dream pain could originate with seeds of ongoing nociception. Dream pain could be entirely endongerous/central, but it’s not a given.

   Dream pain also might not be “pain” per se—it might be wholly affective/symbolic. We might just dream about the emotional response to pain. All the emotions and distress and frustration and anguish … but, perhaps, not the actual sensation.

   And dreams are weird, such that even debating their content and its significance is obviously fraught. A lot of dream stuff seems to be unique to dreaming. Even if dream pain is in fact “real pain” that is centrally generated, with no nociceptive seed, that still only gets us so far—it is just one of a handful of dubious sources of support for the weak claim that pain can arise without tissue damage. But it’s doing it in a properly weird neurological state, which obviously may not generalize to waking experience. And it says nothing at all about it being prediction-driven, psychologically mediated, or socially constructed.

   Dreams are a poor foundation. They are only knowable retrospectively, with all the distortion, confabulation, and narrative confabulation implied by that.

10. Madden VJ, Moseley GL. Do clinicians think that pain can be a classically conditioned response to a non-noxious stimulus? Man Ther. 2016 Apr;22:165–73. PubMed 26794284 ❐
11. Beecher HK. The powerful placebo. J Am Med Assoc. 1955 Dec;159(17):1602–6. PubMed 13271123 ❐

    This is the 1955 paper that launched the Legend of Placebo. Dr. Henry Beecher reported that 35% of 1000 patients were “satisfactorily” treated with a placebo alone. His conclusion catapulted placebo to lasting fame… and wasn't really questioned for a long time.

    Kienle and Kiene published a strong criticism of Beecher's findings in 1996, but no one took much notice. Much more prominently, Hróbjartsson and Gøtzsche reported in 2001 in the New England Journal of Medicine that they “found little evidence in general that placebos had powerful clinical effects” and concluded that there is “no justification for the use of placebos” outside clinical trials.

    The topic has been hotly debated ever since, but few experts still believe that placebo is “powerful.”

    For much more about Beecher and how this all started, see The Legend of the Wartime Placebo

12. Hróbjartsson A, Gøtzsche PC. Is the placebo powerless? Update of a systematic review with 52 new randomized trials comparing placebo with no treatment. J Intern Med. 2004 Aug;256(2):91–100. PubMed 15257721 ❐

    This paper was the first widely discussed challenge to the claim that placebo is “powerful” (starting with the first version in 2001, see Hróbjartsson, which followed a more obscure paper by Kienle in 1997). Hróbjartsson and Gøtzsche “found little evidence in general that placebos had powerful clinical effects” and concluded that there is “no justification for the use of placebos” outside clinical trials. The topic has been hotly debated ever since, but few experts still believe that placebo is “powerful.”

13. ScienceBasedMedicine.org [Internet]. Brissonnet J. Placebo, Are You There?; 2015 Mar 12 [cited 22 May 26]. PainSci Bibliography 54158 ❐
14. Jackson M. Pain: The science and culture of why we hurt. Trade paperback ed. Random House; 2003.
15. Cramps and twitches are the tip of the iceberg — there are many major types of unwanted muscle contractions. There are also many myths about them: dehydration and magnesium deficiency don’t cause cramps, “stiffness” isn’t caused by “tight” muscles, and “back spasm” is just a way of saying how back pain feels, not how it works. While many common aches and pains feel crampy, no one knows if there is any such thing as a contraction that can cause pain without being otherwise seen or felt — but it is possible. See Cramps, Spasms, Tremors & Twitches: The biology and treatment of unwanted muscle contractions.
16. Espay AJ, Aybek S, Carson A, et al. Current Concepts in Diagnosis and Treatment of Functional Neurological Disorders. JAMA Neurol. 2018 09;75(9):1132–1141. PubMed 29868890 ❐
17. O’Sullivan S. It's All in Your Head: True Stories of Imaginary Illness. Chatto & Windus; 2015.

    This book consists mainly of well-told stories of severe psychosomatic illness and functional neurological disorders (neurological symptoms without diagnosable disease). The key take-away is that psychologically powered illness is common and can be amazingly severe. Although Dr. O’Sullivan is clearly concerned about the risk of incorrect diagnosis, and she is cautious and compassionate enough that I think she mostly gets it right (with the notable exception of the chronic fatigue chapter). It’s well-written and fascinating and has plenty to offer. I do wish there were citations.