the mental-health lexicon

“That was so traumatic.” “I'm processing my trauma.” A word that in clinical use named events like combat, disaster and assault now also covers breakups, difficult bosses and painful childhood dynamics. The usage is sincere: it names experiences that genuinely hurt.

The clinical anchor is PTSD's “Criterion A” — the DSM's definition of a qualifying traumatic event, whose own successive expansions the psychologist Richard McNally dubbed “bracket creep” [7]. The scholarly account of the wider cultural expansion is Nick Haslam's 2016 “concept creep” paper: psychology's concepts of harm and pathology — trauma among them — have steadily widened both horizontally (new kinds of events) and vertically (milder instances of the same kind) [4].

The descriptive claim — the word's reach has measurably widened — is unusually well-evidenced for a critical thesis: a validated scale measuring how broadly individuals apply harm concepts [5], and a corpus analysis of 825,628 psychology abstracts from 1970 to 2017 showing “trauma” increasingly used in less severe contexts inside the scientific literature itself [6]. This is not an old man complaining about language; the broadening is in the data. Grade: strong.

The evaluative claim is a different matter, in both directions. That the widening happened is settled; whether it is a problem is exactly what the field is arguing about:

The everyday gap is subtler than either camp: a person can be genuinely and lastingly hurt by an experience that does not meet the clinical threshold. The clinical word makes a specific promise — a defined construct, with specific evidence-based treatments attached. The colloquial word makes no such promise, and treating the two as interchangeable is where the confusion starts.

• “Does what happened to me meet the clinical definition you'd use — and does the answer change what treatment you'd recommend?”
• “If it isn't PTSD, what is the right name for what I'm carrying? Does it need treatment, support, or time?” — an experience that misses Criterion A is not thereby disqualified from mattering.

“The body keeps the score.” Trauma is said to live in your tissues — hips, shoulders, gut — and to be releasable through body-based work. Of the eighteen entries on this page, this is the one where a book title has most completely become folk doctrine.

A real scientific lineage: Bessel van der Kolk's 1994 paper “The body keeps the score: memory and the evolving psychobiology of posttraumatic stress” [10], and the 2014 book that carried the phrase into general culture [11].

The defensible reading — the body is involved in trauma — is well supported. A meta-analysis of PTSD psychophysiology finds altered heart rate, skin conductance and startle response [12], and interoceptive dysfunction after trauma is a live, legitimate research area [13]. If the sentence means “PTSD has measurable peripheral physiology and altered body perception,” it grades strong.

The popular reading — traumatic memories literally recorded in body tissue, retrievable and dischargeable through body-based therapy — has no mechanism in this literature. It is a metaphor, and the strong claims built on it fare badly under peer review. Under the house rule, this book's claims never appear here without the strongest published critiques beside them:

• A peer-reviewed appraisal of the book itself (Scheeringa 2025, BJPsych Bulletin) concludes its brain-damage claims and its claims of unique body-based-treatment efficacy lack support [14]; a further critical appraisal reaches similar conclusions [16].
• The repressed-memory framing the strong reading leans on is actively contradicted by the memory-science literature (Otgaar et al. 2019, “The Return of the Repressed”) [15].
• Flagship alternative therapies grade poorly on their distinctive claims: a meta-analysis found EMDR's eye movements add nothing over exposure therapy [17], and an EEG-neurofeedback meta-analysis rated the certainty of evidence “very low” for all outcomes [18]. (Both are findings about the therapies' distinctive mechanism claims — a separate question from whether people in treatment improve.)

This is the largest usage-evidence gap on this page: the settled part (physiological correlates) is real, and everything distinctive to the strong thesis — somatic storage as mechanism, recovered memories, unique body-based cures — is weak or contradicted.

• “When you say trauma is stored in my body, do you mean measurable physiology — or literal memory in tissue? Which does the treatment you're proposing assume?”
• “Which of the therapies you're suggesting has controlled-trial support for someone with my presentation — and what would we expect to change, by when?”

“My nervous system is dysregulated.” “You're not lazy, your nervous system is stuck in freeze.” A vocabulary of vagus nerves, ventral and dorsal states, and exercises promising to “regulate” you — usually traced, knowingly or not, to one theory.

Stephen Porges's polyvagal theory, introduced in 1995 and elaborated over a 2001 paper, a 2007 statement and a 2011 book [19]. The wellness idiom's entire vocabulary — vagal tone, ventral/dorsal states, safety cues — is polyvagal vocabulary.

A real autonomic research base exists, and it does not need polyvagal theory to stand: the neurovisceral-integration model and a meta-analysis linking heart-rate variability to brain regions involved in self-regulation [20], a meta-analytic review supporting HRV as a biomarker of top-down self-regulation [21], and meta-analyses showing HRV biofeedback reduces self-reported stress and anxiety [22]. If the phrase means “my autonomic nervous system responds to stress, and that response is measurable and somewhat trainable,” it grades moderate.

And the largest gap needs stating plainly: “nervous system dysregulation” is not a diagnosis anywhere. The phrase appears in no billable ICD-10-CM diagnosis name; neither does the bare word “dysregulation” — a negative verified three independent ways [26]. Content that treats it as an identifiable clinical state you can be screened into is naming something no diagnostic system recognizes.

• “When we say I'm dysregulated — what measurable thing do we mean, and how would we know it changed?”
• “Is there a recognized diagnosis that fits what I'm describing better — anxiety, panic, PTSD — with treatments that have trial evidence?”
• “If a program promises to regulate my nervous system, what outcome would show it worked — and what happens if we just treated the anxiety?”

Chronic stress is said to exhaust your adrenal glands until they can no longer produce enough cortisol — hence fatigue, brain fog, salt cravings — and supplements are sold to “support” the tired glands.

Not from the research literature. The term was coined in a 2001 trade book by James L. Wilson — described on his own company's site as “ND, DC, PhD,” a self-description we could verify only as his site's claim [29]. His website today operates as a storefront for adrenal-support supplements — the coiner sells the treatment for the condition he coined [29]. There is no founding paper, no validated test, and no diagnostic entity.

For the construct itself: nothing. The only systematic review of the question is titled “Adrenal fatigue does not exist” and concludes, verbatim: “This systematic review proves that there is no substantiation that ‘adrenal fatigue’ is an actual medical condition. Therefore, adrenal fatigue is still a myth” [27]. The Endocrine Society's patient guidance is equally direct: “No scientific proof exists to support adrenal fatigue as a true medical condition” [28] — and a second society on another continent says the same: the Endocrine Society of Australia's patient guidance states “Adrenal fatigue is not a recognised medical condition” [118]. What is real nearby: adrenal insufficiency (Addison's disease) — an actual, testable, treatable failure of the adrenal glands [28] — and the genuine HPA-axis stress literature, whose own nuanced findings (cortisol output varies by stressor timing and type, in both directions) undercut the worn-out-gland folk model [30].

Entirely. The fatigue is real; “tired adrenals” as its cause has zero substantiation, the cortisol-testing methodology sold around it is, in the systematic review's words, not endorsed by endocrinologists [27], and the endocrinologists' stated concern is concrete: accepting an unproven diagnosis delays finding the real cause of symptoms that deserve an actual workup [28].

• “My fatigue is real — what should actually be ruled out?” (There is a standard workup for persistent fatigue. Ask for it.)
• “Is there any reason to test my adrenal function — and if so, with the tests endocrinologists actually use?”

Abstain from stimulation — phones, sugar, music, sometimes eye contact — to let your “dopamine levels reset,” on the theory that pleasure spends dopamine down and abstinence refills it.

A 2019 Medium post by Cameron Sepah, “Dopamine Fasting 2.0 — The Hot Silicon Valley Trend” [31], amplified by the New York Times (“How to Feel Nothing Now, in Order to Feel More Later,” November 2019) and the Guardian [32][33]. The remarkable part: the originator disowned the literal reading immediately. Sepah told the Times that dopamine “makes for a catchy title… The title's not to be taken literally,” and the Times itself noted “the name — dopamine fasting — is a bit of a misnomer. It's more of a stimulation fast” [32].

The dopamine science the fad garbles is itself strong: dopamine neurons encode reward prediction error [35], and the wanting/liking literature shows dopamine tracks incentive salience — “wanting” — rather than pleasure itself [36]. None of it describes a currency depleted by enjoyment and replenished by abstinence. Harvard Health's assessment of the fad is blunt: “dopamine doesn't actually decrease when you avoid overstimulating activities, so a dopamine ‘fast’ doesn't actually lower your dopamine levels” [34]. What the practice actually is — removing cues and access for a compulsive behavior — is ordinary stimulus control from CBT, which is real, useful and unremarkable [31][34].

The neurochemical claim is wrong by the originator's own admission. One honesty note this page owes you: there is no peer-reviewed “dopamine fasting debunk” literature either — the only indexed paper on the phenomenon is a descriptive piece in a low-barrier venue [37]. The debunk above rests on institutional expert commentary [34] and the real dopamine literature [35][36], and is labeled accordingly. Anyone citing a “peer-reviewed dopamine-detox study” in either direction is citing something that does not exist.

• “What I'm calling a dopamine problem is really a behavior loop — can we treat it like one?” Stimulus control, behavioral activation and CBT for compulsive use patterns are askable, evidence-based things.
• If the urge to “detox” is driven by feeling unable to stop — that's worth saying out loud to a clinician in exactly those words.

“I'm anxious-attached; he's avoidant.” Online quizzes sort you into one of four types, usually presented as set in infancy and driving your relationships from the wings.

One of the most complete lineages in psychology, and this entry exists partly to show what a term with a real pedigree looks like: Bowlby's Attachment and Loss trilogy [38], Ainsworth's Strange Situation [39], the extension to adult romantic attachment [40], the four-category model the quizzes descend from [41], and modern validated instruments (ECR, ECR-R) [42].

The construct is real, measurable and predictive — grade strong. Attachment security is moderately stable from infancy through early adulthood, with “prototype” dynamics — early representations persist but keep being revised [43]. Insecure attachment predicts lower relationship satisfaction meta-analytically, and the field has an authoritative book-length review [44].

Two specific planks of the quiz version fail against the same literature that validates the construct:

• “Four types” — taxometric analyses find attachment is dimensional: degrees of anxiety and avoidance, not categories. In the researchers' words, attachment is “best measured and conceptualized in terms of dimensions, not as a categorical variable,” a finding replicated with modern methods [45].
• “Stamped in infancy” — meta-analysis finds infancy-to-adulthood continuity weak to moderate [46]. Early experience matters; it is not destiny, and adult attachment measurably moves.

(One label required by our method: no peer-reviewed critique of the popular book Attached could be located as of 2026-08-26 — recorded as “none located,” not “none exists.”)

• “Rather than which type I am — where do I sit on attachment anxiety and attachment avoidance, and what tends to move those dimensions?”
• “If my pattern is hurting my relationships, which therapies target this directly?” — a question with real answers, which is what a strong construct buys you.

Originally: a sustained campaign to make someone doubt their own perception of reality. Increasingly: any disagreement, denial, or mismatch of memory — “he said the meeting was at 3, that's gaslighting.”

The 1944 film Gaslight supplied the name (cultural context, not science). The first clinical use is a 1969 Lancet paper, “The gas-light phenomenon” [47], elaborated psychoanalytically in 1981 [48]. The modern scholarly life is real and recent: a 2019 American Sociological Review treatment of gaslighting as a sociological phenomenon [49], sustained philosophical work [50], and a 2025/2026 Personality and Social Psychology Review theoretical framework [51].

As a scholarly construct: moderate — a legitimate object of study in top venues (PubMed indexed 79 papers with “gaslighting” in the title as of a 2026-08-26 census), but young and definitionally unsettled; the PSPR framework paper itself says psychology “lacks clear scientific explanations” for it [51]. As a clinical diagnosis: none — no DSM-5-TR or ICD-11 entity, verified two independent ways. Calling someone “a gaslighter” names a behavior pattern, not a diagnosis that exists.

The dilution everyone senses — from “sustained manipulation of reality-testing within a power relation” to “you remembered this differently than me” — is complained about by the scholars themselves and visible in usage markers (Merriam-Webster made it the 2022 Word of the Year [52]), but here is the honest label: no corpus study has measured the drift, the way Baes et al. measured “trauma” [6]. The dilution claim is commentary-supported, not measured — this page carries it as exactly that.

• The useful clinical question is not “is this gaslighting?” but: “here is the pattern — I say what I saw, it gets denied, over time I trust myself less. What is happening to me, and what should I do?”
• A relationship in which your perception is persistently overwritten is worth bringing to a clinician whatever it gets labeled — the label is not the admission ticket.

The high achiever who is quietly frantic: outwardly successful, inwardly over-preparing, people-pleasing, unable to rest. Presented online as a hidden diagnosis the doctors miss.

Not from science. The negative is verified two independent ways: a PubMed exact-phrase search returns two papers total, neither treating it as a defined construct — one of them studies the term's media life and notes the concept “has managed to escape” a research framework [53] — and no foundational or validation paper exists in Google Scholar. It is not in DSM-5-TR or ICD-11 [53]. Even the health-media coverage concedes it: “Doctors do not recognize high functioning anxiety as a distinct anxiety condition” [55]. Its specific first popularizer could not be pinned down; the origin is honestly stated as: popularized in 2010s health media [55].

As a construct: none. But the experience it names sits directly next to something real and well studied: subthreshold generalized anxiety — anxiety that misses full diagnostic criteria yet carries real burden — which a systematic review (18 studies, N=48,214) found roughly twice as prevalent as full-syndrome GAD [54]. And full GAD itself requires distress or impairment — you can meet criteria for an anxiety disorder while performing well at work. Functioning is not evidence against diagnosis.

Listicles of “signs of high-functioning anxiety” describe an undefined thing — there are no criteria, so anything can be a sign. The cost runs in both directions: people over-identify with an unfalsifiable label, or use “high-functioning” to talk themselves out of care they would qualify for. The evidence-backed version of this entry's insight is simpler: anxiety that doesn't stop you functioning can still be clinically significant, common, and treatable [54].

• “My anxiety doesn't stop me functioning — could it still meet criteria for GAD or subthreshold anxiety, and would treatment help?”
• “What does my anxiety cost me that I've stopped counting — sleep, rest, relationships — even if work is fine?”

“TikTok gave teenagers Tourette's.” During the pandemic, clinics worldwide reported a surge of sudden-onset tic-like movements and vocalizations, mostly in adolescent girls, temporally linked to viral tic-content creators.

This one comes from the clinic, not the feed. Between 2021 and 2023 the movement-disorder literature documented the surge in detail: Hannover's clinic series (published in Brain under the deliberately provocative title “Stop that! It's not Tourette's but a new type of mass sociogenic illness”) identified a “virtual index case” whose specific vocalizations and movements patients reproduced [56]; North American groups described rapid-onset functional tic-like behaviors in young females [57]; a second group quantified the phenomenology (“TikTok Tics: A Pandemic Within a Pandemic”) — arm-dominant movements, around 29 tics a minute, near-universal coprolalia and self-injurious behavior, all rare in Tourette's [58]; and another group documented six patients matching a specific influencer's repertoire [59]. European consensus criteria for distinguishing functional tic-like behaviors from Tourette syndrome followed in 2023 [60].

That the phenomenon exists, is phenomenologically distinct from Tourette syndrome, has published diagnostic criteria, and has a good prognosis — grade moderate-to-high, effectively settled. Follow-up cohorts show substantial improvement: in the largest, tic severity dropped 8.9 points by six months and a further 6.4 by twelve [64]; in the longest (mean 26 months), 63% improved, 23% fully remitted, and none were worse [65]. These are real symptoms — functional does not mean fake, and nobody in this literature says the patients are faking.

And the popular version fails on its own terms: the best long-term follow-up found that changing daily social-media time had no influence on prognosis [65]. What the literature supports is narrower than the headline: a stress-precipitated functional presentation in which social media supplied the symptom model — not “an app gave kids Tourette's.”

• “Do these movements look like Tourette's-type tics or functional tic-like behaviors — and what in my presentation distinguishes them?” (There are published criteria; a movement-disorder specialist can apply them [60].)
• “If functional — what does treatment look like, and what is the honest prognosis?” The follow-up data is genuinely encouraging [64][65].

“I saw myself in the video.” A creator lists signs of ADHD or autism; the comments fill with people recognizing themselves; some arrive at a diagnosis without ever seeing a clinician — and some arrive at a clinician because the video was the first thing that ever fit.

The relevant science here is the content-accuracy literature — researchers taking the most-viewed diagnostic content and rating it against clinical criteria. Three independent teams, three conditions: the top 100 #ADHD TikToks (52% rated misleading, with misleading content coming predominantly from non-healthcare creators) [66]; a pre-registered pair of studies in which expert psychologists found fewer than 50% of symptom claims in top ADHD videos aligned with DSM-5 criteria [67]; and the top 133 informational #autism videos (27% accurate, 41% inaccurate, 32% overgeneralized) [68].

Convergent and replicated — grade moderate-to-high for three claims: roughly half or more of top-viewed diagnostic content fails expert review [66][67][68]; healthcare-professional creators are more accurate than others [66]; and engagement does not track accuracy — likes cannot tell you a video is right [68]. The viewer-side evidence adds a fourth: heavier #ADHD-TikTok consumption predicted recommending even the worst-rated videos and inflating prevalence estimates [67]. The genre's dominant failure mode, per the studies themselves, is not fabrication — it is overgeneralization from genuine personal experience [68].

In both directions. The panic version (“everyone is self-diagnosing off TikTok”) rests on a number that does not exist: no peer-reviewed population-prevalence estimate of social-media-driven self-diagnosis has been published as of 2026-08-26 — an explicitly labeled evidence gap, and the circulating industry statistics do not survive verification. And the limits of the accuracy studies are stated in the papers themselves: top-video sampling of one hashtag at one time point, rater-defined “misleading” with definitions that differ across studies (the headline percentages are not comparable), and content ratings that say nothing directly about viewer harm [66][67][68].

This page's own position, per its rules: self-diagnosis is treated with respect. Recognizing yourself in a description is often the first act of taking your own mind seriously — frequently by people the system has been slow to assess. The evidence above says the raw material is unreliable; it does not say the recognition is.

• “This content described me — can we do a proper evaluation?” Bring the specifics: which experiences fit, since when, in which settings. That is genuinely useful clinical information; the video citation is not required.
• “If it isn't what the video named, what else explains this pattern?” — the question a feed will never ask you.

“I'm completely burnt out” — offered, increasingly, as a medical self-description, and applied well beyond work: parental burnout, relationship burnout, life burnout. Its ancestor idiom, the “nervous breakdown,” did the same cultural job for most of the 20th century.

“Nervous breakdown” was never a formal diagnosis in any edition of any manual — a cultural idiom with a well-documented published history [76]. Burnout's lineage is scientific and continuous: Freudenberger's 1974 “Staff Burn-Out” [70], the Maslach Burnout Inventory [71], and a five-decade occupational research program [72].

Burnout as a measurable occupational phenomenon: strong. And its official status is one of the cleanest facts on this page, because the WHO wrote it down: burn-out “is included in the 11th Revision of the International Classification of Diseases (ICD-11) as an occupational phenomenon. It is not classified as a medical condition” — defined by three dimensions (exhaustion; mental distance or cynicism toward one's job; reduced professional efficacy), and, per WHO's own guidance, the term “refers specifically to phenomena in the occupational context and should not be applied to describe experiences in other areas of life” [69]. It is likewise absent from DSM-5-TR (verified via convergent secondary sources) [69].

So the gap is double. “I have burnout” as a medical claim asserts a diagnosis the WHO deliberately declined to create; and stretching it beyond work (“parental burnout”) exceeds the ICD-11 definition by the WHO's own words [69]. Neither makes the exhaustion less real — it makes the label less load-bearing than it sounds.

• “Is this burnout, depression, or both — and how would you tell them apart in my case?” (That question sits directly on top of the live academic dispute [73][74] — which is why it belongs to a clinician, not a quiz.)
• “If the job is the exposure, what changes — in the work or in me — have actually helped people in my position?”

“Let the intrusive thoughts win” — the meme version, where an “intrusive thought” is a whimsical impulse you secretly want to obey: dye the hair, buy the ticket, eat the cake.

A core clinical construct, central to OCD research and psychoeducation: unwanted, involuntary thoughts, images or urges that intrude against one's will and clash with one's values — ego-dystonic, in the clinical term. Often the thoughts are about exactly the things the person finds most abhorrent, which is why they stick.

This is the entry where the evidence is stronger — and kinder — than the meme. The key finding, from a 19-author study across six continents (“You can run but you can't hide: Intrusive thoughts on six continents”): unwanted intrusive thoughts are near-universal [77]. Having a sudden, unbidden, horrible thought is not a symptom of anything — it is a property of human minds. What distinguishes clinical from ordinary is not having the thoughts; it is what happens next — appraisal, distress, and compulsive response. Grade: strong, and inherently destigmatizing.

Uniquely on this page, the gap runs in reverse: popular usage is more trivializing than the evidence warrants, not more expansive. The meme's “intrusive thoughts” are ego-syntonic whims — things you half-want. Clinical intrusive thoughts are the opposite: unwanted precisely because they violate who you are, and “letting them win” is the opposite of their phenomenology — nobody with harm-themed intrusions wants them to win, which is the whole clinical picture. The meme is harmless as comedy; the cost is that someone tormented by real intrusions may not recognize their experience in the phrase — or may conclude, wrongly, that having dark thoughts says something dark about them. The six-continent finding says otherwise, about everyone [77].

• “My thoughts are unwanted, distressing, and I do things to neutralize them — is this OCD-spectrum, and is exposure-based treatment (ERP) right for me?”
• Worth knowing before you ask: the content of intrusive thoughts — harm, sexual, blasphemous — is not what clinicians find alarming; the near-universal finding is exactly why [77]. Clinicians who treat OCD have heard it, whatever it is.

“I'm acts of service; he's words of affirmation.” Every person is said to have one primary love language out of five, and relationships are said to succeed when partners learn to speak each other's. It is probably the most widely deployed relationship framework in the world.

A 1992 trade book by Gary Chapman, a Baptist pastor and counselor — not from a research program. Chapman has no indexed peer-reviewed publication on love languages (a negative verified two ways), and the framework's empirical literature is remarkably small — nine indexed papers total — and arrived decades later, from other researchers [78].

The instinct the book names — that partners differ in how they most feel loved, and that it pays to know — sits next to one of relationship science's best-supported constructs: perceived partner responsiveness, the experience of feeling understood, valued and cared for [84]. A dyadic love-languages study found exactly this shape: what predicted satisfaction was not typological matching but accurately understanding what your partner actually values [83]. That nearby construct grades strong; the book's specific machinery does not.

The framework's three central claims have now been tested, and a 2024 review in Current Directions in Psychological Science concludes the evidence does not provide strong support for any of them — a preferred language, five of them, or matching as the key to satisfaction [79]:

• The matching hypothesis — the largest direct test (n=696) states it plainly: “Chapman's hypothesis was not supported.” Satisfaction with a partner's primary-language behavior predicted outcomes no better than the lower-ranked languages — and words of affirmation and quality time predicted love and satisfaction better than whatever a person's designated primary language was [81]. An earlier couples study found only “limited evidence” for alignment, with self-regulation explaining more [80].
• “There are five” — a bottom-up factor analysis across three samples (with the 2024 review's lead author on the team) found solutions of 7 to 10 love languages fit better and predicted relationship quality better than the five-factor model [82].

The 2024 review's alternative metaphor is worth carrying whole: love is not a language you must learn to speak but a balanced diet — people need a full range of nutrients, not one designated dish [79]. The quiz's real risk is a couple optimizing a single channel while the diet goes thin.

• For a couples therapist: “rather than sorting us into types — what does each of us actually experience as care, and how accurately do we read each other?” That accuracy is measurable, and it moves [83].
• “If we're dissatisfied, which needs are going unmet across the board?” — the balanced-diet question, which no five-item quiz can answer for you.

“Talk therapy can't reach trauma — you have to release it from the body.” Somatic content promises to discharge stored survival energy through tremoring, pendulation and body awareness, usually naming Somatic Experiencing (SE), the method's flagship brand.

Peter Levine's 1997 trade book Waking the Tiger and its 2010 successor [85], built on the idea that trauma is interrupted defensive energy held in the body until discharged — the therapeutic arm of the wave-1 “stored in the body” story.

That the body is involved in trauma is well established (see entry 02: PTSD has measurable psychophysiology [12]) — that part grades strong and belongs to no brand. For SE specifically, the trial record is exactly two randomized controlled trials: a positive one — small (N=63), against a waitlist, with an SE trainer among the authors, self-described as the “first known randomized controlled study” and concluding SE “may be” effective [86] — and a larger null one (N=114, SE added to physiotherapy): “no significant group differences… on any of the outcomes at any timepoints” [87]. The field's own scoping review calls the evidence preliminary, mixed in quality, and in need of unbiased RCTs [88]; the broader body-oriented-therapy meta-analysis finds a moderate pooled effect with high heterogeneity and risk of bias in almost every study [89].

• “Proven” is not available. Citing Brom 2017 as showing SE works “as well as CBT or EMDR” is a miscitation — the trial compared SE to people waiting, not to any treatment [86].
• The mechanism has never been measured. No empirical test of the “discharge”/“pendulation” mechanism exists in the indexed literature — the mechanism papers are theoretical [90]. The grade is not “disproven”; it is unmeasured, which after nearly three decades is itself information.
• The peer-reviewed appraisal of the Body Keeps the Score claims names somatic experiencing explicitly among body-based treatments whose promotion as superior “is not currently justified by the weight of evidence” [14].

• “Before we try a body-based approach — am I a candidate for the treatments with the strongest trial evidence for PTSD, and if not, why not?”
• “If we do SE — what would we expect to change, by when, and how will we know it's the therapy and not the time?” The null add-on trial [87] is exactly why that question matters.

“My ex is a narcissist.” “Signs you were raised by a narcissist.” Online, the word functions as a diagnosis pronounced from the outside — usually of someone who hurt you — with “narcissistic abuse” as its companion term.

A genuine clinical lineage: psychoanalytic accounts by Kohut (1968) and Kernberg (1970) [91], the diagnosis's entry into the DSM in 1980 [91], and a parallel research tradition measuring narcissism as a trait in ordinary populations, anchored by the Narcissistic Personality Inventory [92]. Modern work distinguishes grandiose from vulnerable presentations and treats narcissism as a spectrum on which the disorder is one extreme region [96].

Narcissism as a studied trait: strong — decades of measurement and a live research field [92][96]. NPD as a diagnosis: real, and structurally contested inside the field — the standard review documents criterion problems, a grandiose-only bias, and the DSM-5 fight in which the diagnosis was nearly deleted [97]. And the prevalence numbers need their labels: the circulating 6.2% is a lifetime estimate from lay-administered survey interviews [93]; point-prevalence studies with structured clinical interviews find about 0.8% [94], and the systematic review across community studies puts the mean at 1.06% (range 0–6.2%) [95]. Roughly one in a hundred at a given time — not one in sixteen.

• The creep is measured. Uniquely on this page, this term's dilution has its own study: a 2026 vignette experiment in the concept-creep research program found about a quarter of participants endorsed over-inclusive NPD concepts extending to non-pathological features [98]. Selfishness, vanity and cruelty are real and can be worth leaving a relationship over — none of them is a personality disorder.
• “Narcissistic abuse” has essentially no research base: a title census finds four papers, none empirical, one of them a commentary conceding the gap [99]. The experiences people bring to the term — control, denigration, reality-distortion — are real and studied under other names (see gaslighting), and legitimate research exists on the burden of living alongside pathological narcissism [99]. The branded construct is ahead of its evidence.
• And the standing rule: a personality disorder is a clinical determination made from examination — not a label conferred by the diagnosis-from-a- distance content economy.

• About someone else: the useful version is not “is he a narcissist?” but “here is how I'm treated — what is this doing to me, and what should I do?” Your treatment does not depend on their diagnosis.
• About yourself, if the accusation landed on you: patterns of entitlement or empathy failure are assessable — and asking is itself evidence the worst version of the label probably doesn't fit.

Involuntary, obsessive romantic fixation — intrusive thoughts about a person, euphoria at any sign of reciprocation, despair at its absence. The internet rediscovered the word and made it an identity: “I'm limerent again.”

Coined by psychologist Dorothy Tennov in her 1979 book Love and Limerence, from interview research [100]. Then it slept for four decades: the entire indexed literature is a few dozen papers, and much of what circulates online rests on grey literature that was never peer-reviewed [105].

This entry is unusual: the science is younger than the meme, and catching up. The first validated measure — the LQ-11 — was published in late 2025; its own abstract notes no published measure existed before it [101]. The largest study to date (N=1,647, with real-time experience sampling) found limerence-prone people reported elevated adverse childhood experiences, insecure attachment and obsessive-compulsive cognitive traits, with high psychiatric comorbidity [102] — echoing, with data, what the scoping review could only note about an “extraordinarily limited literature” a year earlier [103]. Grade for the research object: emerging — real, correlational, and unreplicated.

Three specific gaps. Limerence is not a diagnosis — it appears in no manual, and describing yourself as limerent describes an experience, not a condition. The treatment literature is one case study [104] — anyone selling a limerence-recovery protocol is selling past the evidence. And the term's social-media resurgence itself is undocumented in the indexed literature — widely reported in media, unmeasured in research [105]. None of this diminishes the experience: the new data suggests that for some people it is genuinely disabling, which is precisely why the construct is finally being measured [102].

• You don't need the word to be official: “I have intrusive, obsessional preoccupation with a person; it's involuntary and it's disrupting my life” is a fully clinical sentence a clinician can work with today.
• Worth asking directly: “could this pattern be related to attachment anxiety or obsessive-compulsive tendencies — and would treatment aimed at those help?” That is where the new evidence points [102].

“Task paralysis”: sitting frozen in front of a task you fully intend to do — sometimes one you want to do — unable to start, while the deadline burns. Online it functions as ADHD's signature symptom and, often, as diagnostic evidence in itself.

Not from the clinical literature: the phrase appears in zero indexed papers as a construct — a negative verified four independent ways — and its nearest academic sighting is a conference abstract [106]. Even the Cleveland Clinic's own explainer, the term's most authoritative media treatment, opens by saying it “isn't a medical diagnosis” [107]. Its incubator is ADHD lifestyle media, and a 2025 study of ADHD content catalogued creators' “ADHD paralysis” explainers among videos classified as misleading [108].

What the term gestures at is one of the best-studied parts of ADHD: executive function deficits. The unifying theory is Barkley's 1997 account of behavioral inhibition and executive functions [109], and a meta-analysis of 83 studies finds reliable ADHD-associated deficits in response inhibition, vigilance, working memory and planning — exactly the machinery of starting tasks [110]. ADHD symptoms also correlate with procrastination in the (thinner, correlational) literature [111]. That nearby construct grades strong.

Two gaps, one in each direction. First, the meta-analysis's own caveat, verbatim: executive-function weaknesses are “neither necessary nor sufficient to cause all cases of ADHD” [110] — moderate effect sizes, not universal, so freezing at tasks is neither required for ADHD nor proof of it. Second, task paralysis is nobody's exclusive property: procrastination, depression, anxiety and plain overwhelm all produce it, which is why a symptom's vividness in a video is not a differential diagnosis. The experience is real either way — what's unsupported is the leap from “I freeze” to “therefore ADHD.”

• Bring the concrete pattern, not the label: “I freeze at task initiation — here's when it happens, since when, and in which settings.” That is assessable; “I have ADHD paralysis” is not.
• “If it is ADHD, what actually helps task initiation — and if it isn't, what else explains this pattern?” Both halves of that question have real answers.

“I notice everything — exits, moods, the shift in someone's tone. Hypervigilance from growing up on eggshells.” Online the word drifts toward an identity, sometimes a superpower: the trauma-honed empath who reads every room.

Unusually for this page: from the clinic, and it never left. Hypervigilance is a named symptom in the DSM-5 criteria for PTSD — the arousal and reactivity cluster — with over a thousand indexed papers [112]. This is the wave's entry where the word's credentials are in order.

The construct grades strong: a criterion symptom embedded in a large empirical literature, including the meta-analysis of threat-related attentional bias in anxiety [113] and a systematic eye-tracking review in PTSD [114]. But the best measurement adds a twist worth knowing: the eye-tracking evidence finds little support for enhanced threat detection — the “always scanning, notices everything” picture — and consistent evidence for sustained attention on threat: once danger is noticed, disengaging is what's hard [114]. Not a sharper radar; a stickier one.

Two honest labels. The lab's workhorse measure of attentional bias — the dot-probe task — has documented reliability problems, and whether the attentional-bias literature is in “crisis or opportunity” is the field's own open question [116]; the therapies built on retraining that bias have a mixed trial record [115]. And the superpower framing has no literature at all — no peer-reviewed work documents “hypervigilant” as the identity term social media has made of it [117]. Clinically, hypervigilance is described as a costly symptom — exhausting, sleep-degrading, anchored to a threat system that will not stand down — and romanticizing it is the one usage the literature gives no support at all.

• “My scanning never switches off — could this be part of a trauma- or anxiety-related condition, and what actually turns the volume down?” Hypervigilance responds to the same evidence-based treatments as the conditions it belongs to.
• If the vigilance feels like your best skill: “what would it mean to keep the perceptiveness and lose the cost?” — a real treatment goal, stated in one sentence.