Every claim here is checked against its original research — graded, sourced, and measured against what the top of Google still gets wrong, with our own numbers from reading those ranking pages. Titles link to the full audit; filter by category or search below.
Going deeper on one topic: Procrastination · Attention · Habits · Energy.
CConfirmed
PPartial
HUntested / hypothesis
RRefuted
—Queued, not yet graded
Nothing matches. Try a shorter word.
| Claim | Grade | Where it comes from | Evidence |
|---|---|---|---|
| Energy, Sleep & Body ClockCold Showers and Cold Plunges: Every Claim, RankedCold showers and cold plunges boost your immune system. Blood markers move over 90 days, the only large trial found fewer days off work without fewer illness days, and pooled across trials there was no significant effect on immune function. Still on Google26 of 110 pages returned by fourteen queries state that cold exposure boosts immunity as established, while 24 name a limit and 12 say the evidence does not support itChecked Aug 20, 2026 | P | Winter swimming traditions and Wim Hof's public demonstrations, turned into a product category once plunge tubs, barrels and chillers became consumer goods. | Why it spreadCold water produces an immediate, unmistakable physiological jolt, which feels like evidence. Eleven distinct claims with very different evidence bases are sold as one practice with one list of benefits. What actually holdsPooled across 11 trials and 3,177 people, inflammation rose rather than fell after immersion, stress fell only at 12 hours, and immune function did not move. The 29% figure comes from one trial of 3,018 adults and refers to self-reported sickness absence, with illness days unchanged. After resistance training, immersion cost strength and hypertrophy over 12 weeks and blunted the testosterone response. Five fixed PubMed queries found no controlled trial of any cosmetic claim, against a control query returning 312 records. |
| Work & MoneyDoes Emotional Intelligence Predict Success?Emotional intelligence predicts success and matters more than IQ. Partly, and much less than the phrase implies: the associations are real and small, and the strongest of them goes to nothing once cognitive ability and personality are controlled. Still on Google88 of 123 top-ranking pages state that emotional intelligence predicts success without naming the overlap with personality and ability; 73 of the 88 sell an assessment, certification or training on the same pageChecked Aug 20, 2026 | P | The term covers two different measurement traditions: ability tests scored against correct answers, and self-report questionnaires that also ask about motivation, wellbeing and social confidence. Popular usage inherited the enthusiasm attached to the second and the scientific credibility of the first. | Why it spreadSelf-report emotional intelligence questionnaires contain items that measure conscientiousness, emotional stability, extraversion, self-efficacy and general mental ability. Those things do predict performance, so the questionnaires do too, and the correlation looks like evidence for a new construct. What actually holdsAcross 42,529 students and 1,246 effect sizes, emotional intelligence correlated rho = .20 with academic performance and added 1.7% (ability), 0.7% (self-rated) and 2.3% (mixed) of variance beyond intelligence and the Big Five, ranking third in relative importance behind intelligence and conscientiousness. In the workplace, an updated meta-analysis put the correlation between self-reported emotional intelligence and supervisor-rated job performance at rho = .29, down from an earlier .47, and the relationship became nil at beta = -.02 after controlling for ability EI, self-efficacy, self-rated performance, conscientiousness, emotional stability, extraversion and general mental ability, a covariate set that matched the questionnaire content at multiple R = .79. The ability version changes sign by occupation: positive in high emotional labour jobs, negative in low emotional labour ones. In medical students the correlation was r = .13. |
| Work & MoneyAre You the Average of the Five People You Spend the Most Time With?You are the average of the five people you spend the most time with. The line has no author who can be pinned down, no unit of measurement and no study. Still on Google7 of 23 top-ranking pages state the five-people averaging claim as true; 14 of 23 name a source for the line, 7 of 23 cite a studyChecked Aug 19, 2026 | R | Earliest documented instance in print is a 2005 epigraph in Canfield and Switzer's The Success Principles, crediting Jim Rohn; the trail back to Rohn runs through a caller who left the line on a teenager's answering machine in 1989. Rohn's own site defends the saying in two posts published in 2026, seventeen years after his death, citing nothing. | Why it stuckIt turns a four-century-old proverb about company into arithmetic, and arithmetic sounds like a finding — with an instruction attached: replace the people dragging your mean down. How it fell apartThe sentence never says what is being averaged, so as stated it has never been measured, and none of the 23 top-ranking pages read on 19 August 2026 produces a study of the claim itself. Where peers are assigned rather than chosen, one standard deviation of roommate grades moves own grades by about .05 across 1,589 students and by 1.26% of a standard deviation across 7,672; binge drinking crosses between assigned students at 0.086 while smoking sits at 0.002; and the one deliberate attempt to place weak students among strong ones left them 0.061 grade points worse. |
| Self-Control & WillpowerDoes Eating Sugar Restore Your Willpower?Eating something sugary restores your willpower because self-control burns through blood glucose. Still on Google10 of 49 unique reachable top-ranking pages still present “eat sugar for willpower” as settled fact, no caveatChecked Aug 8, 2026 | R | Gailliot and Baumeister, 2007: people who'd just exercised self-control had lower blood glucose, and a glass of real sugar-sweetened lemonade — not a sugar-free stand-in — reversed the effect. The paper's subtitle called willpower “more than a metaphor.” | Why it stuckA sugar crash already felt real to everyone, so a metabolic story for willpower loss made instant intuitive sense — and it came with an easy fix: eat something. How it fell apartRinsing a sugary drink and spitting it out beat a sweet placebo rinse — nothing swallowed, nothing absorbed, so digestion cannot be the mechanism. When blood glucose was measured directly instead of inferred from test scores, self-control tasks didn't lower it. A 2016 p-curve found the published literature indistinguishable from noise, and the 23-lab replication led by the same researcher behind the rinse studies found the depletion effect they were meant to offset was itself near zero. |
| Energy, Sleep & Body ClockIs Polyvagal Theory Evidence-Based?Calming techniques work because they stimulate the vagus nerve — Polyvagal Theory's specific story of two evolutionarily distinct vagal branches, arranged in a strict hierarchy and tracked by heart-rate variability, is disputed by 39 physiologists as of 2026. Still on Google30 of 57 top-ranking pages present Polyvagal Theory as settled neuroscience with zero caveatChecked Aug 16, 2026 | R | Porges, 1994 conference talk / 1995 Psychophysiology paper: proposed the vagus nerve has two evolutionarily distinct branches explaining fight, flight, and freeze. | Why it stuckGave trauma therapy a physiological explanation for freezing that older fight-or-flight models missed — mechanistic, hopeful, and easy to teach. How it fell apart39 physiologists and evolutionary biologists (Grossman et al., 2026) concluded the evolutionary story, the vagal-branch anatomy, and the heart-rate measure used to track it don't hold up on any of five tested premises. |
| Mindset & Self-BeliefDoes Growth Mindset Actually Work?Believing you can improve your intelligence boosts grades — the effect is real, but overstated by roughly a factor of ten, and it mostly shows up in students who are already struggling. Still on Google8 of 61 top-ranking pages present growth mindset as settled fact with zero caveat — and 0 of 61 cite the newest (Nov 2025) global studyChecked Aug 6, 2026 | P | Mueller & Dweck, 1998: praised fifth-graders for being smart vs. for trying hard, then watched who gave up first on a hard puzzle. Built into a theory that intelligence itself can be trained like a muscle. | Why it spreadA hopeful, simple idea — anyone can get smarter by believing they can — that matched how teachers and parents already wanted to talk about effort. What actually holdsA real but small effect (d≈0.08–0.14), concentrated almost entirely in students already struggling, and only when belief was paired with an actual strategy — not on its own. |
| Self-Control & WillpowerDoes Willpower Run Out Like a Muscle?Willpower isn't a battery. Resisting temptation doesn't drain some shared mental fuel tank the way a muscle tires out after a hard set. Still on Google34 of 70 top-ranking pages present the muscle model as settled fact, no caveatChecked Aug 4, 2026 | R | Baumeister, 1998: students resisted fresh cookies for radishes, then gave up on an unsolvable puzzle faster than a control group. He called it "ego depletion" and built a whole theory — the "strength model of self-control" — on that one afternoon. | Why it stuckA felt experience — running low on self-discipline — got a scientific-sounding metaphor attached to it, then decades of confident retelling with the doubt filed off. How it fell apart23 independent labs replicated it on 2,141 people in 2016 — d=0.04, essentially zero. |
| Attention & FocusThe Goldfish Myth: Your Attention Span Is Not 8 SecondsHuman attention fell from 12 seconds in 2000 to 8 seconds by 2013, shorter than a goldfish's 9 seconds. All three numbers trace back to a website that never produced any data. Still on Google7 of 20 top-ranking pages present the 8-second goldfish comparison as settled fact; 11 of 20 state the figure is wrong or unsourcedChecked Aug 19, 2026 | R | Microsoft Canada, spring 2015: a report called Attention Spans carried the 12, 8 and 9-second figures in a graphic credited to a commercial statistics site, Statistic Brain, which listed the National Center for Biotechnology Information and the Associated Press as its verification. Microsoft's own work in that report — a survey of 2,000 Canadians and EEG on 112 people — produced none of those numbers. | Why it stuckThree exact second-counts and a Microsoft name made the figure sound measured, and a goldfish is easier to repeat than a citation. How it fell apartIn March 2017 the BBC programme More or Less went to both named sources; neither could find any record of research behind the figures, and Statistic Brain never replied. No published measurement of a goldfish attention span exists. A measured trend does sit underneath it: mean time on a foreground window fell from 2 min 53 s in 2004 to 47 s in 2016, though across two different instruments. |
| Attention & FocusHow Long Does It Take to Get Your Attention Back?Cut your phone use for a couple of weeks and your attention comes back. No trial has established how long recovery takes or whether it lasts. Still on Google6 of 22 top-ranking pages name a fixed attention-recovery timeline as established; 6 more name a timeline but hedge itChecked Aug 19, 2026 | H | No study stands behind the timeline. On 27 July 2026, of the thirteen pages ranking for four attention-span queries, two named a recovery time — "a matter of weeks" and "most people notice improvements in weeks" — neither citing a source; the 14-to-28-day window quoted elsewhere comes from a review in a journal indexed in neither PubMed, Scopus nor Web of Science. | Why it spreadAnyone who has just given something up wants a date when it pays off, and a number in weeks sounds like something somebody measured. What is still untestedvan Wezel et al., 2021 cut social media by 50% or 10% for one week in 76 people and found better sustained attention at the 50% target — two doses over one week rather than a curve. Hall et al., 2021 randomised 130 people to one, two, three or four weeks of abstinence and measured loneliness and well-being instead of attention. Castelo et al., 2025 blocked mobile internet for two weeks and saw persistence only in self-reported outcomes. |
| Learning & MemoryAre Learning Styles Real? What the Evidence Actually SaysMatching how you teach to someone's "preferred style" doesn't improve learning. Even the people who built the most famous style test never claimed it would. Still on Google5 of 14 top-ranking pages overstate the debunk; 0 of 14 mention the 2024 meta-analysisChecked Jul 27, 2026 | R | Fleming & Mills, 1992: a school inspector's personal observations became the VARK questionnaire — meant to start conversations about teaching, not to diagnose anyone. The authors wrote that testing it for validity was "unnecessary and inappropriate." | Why it stuckIt feels intuitively true — everyone has a preference — and got taught to generations of teachers as settled fact (89% belief among educators, one 18-country survey found). How it fell apartA 2008 review found exactly one study that even approached testing the real claim, and explained why it didn't hold up either. |
| Habit FormationDoes Habit Stacking Actually Work?Habit stacking works because bolting a new action onto an existing routine is backed by decades of behavioural science. Anchoring to a routine does build automaticity — no better than anchoring to a time of day. Still on Google16 of 17 top-ranking pages present habit stacking as backed by behavioural science, no caveat; 9 of 17 cite a named studyChecked Aug 19, 2026 | P | Habit stacking repackages Peter Gollwitzer's if-then implementation intentions, formalised in the 1990s, and borrows their headline number: d = 0.65, from Gollwitzer and Sheeran's 2006 review of 94 studies. | Why it spreadOne fill-in-the-blank sentence — after I [current habit], I will [new habit] — hands the remembering to something you already do every day. What actually holdsKeller and colleagues preregistered the head-to-head in 2021: 192 adults, 84 days, a routine anchor against a 9 a.m. clock anchor, and neither condition beat the other on any measure — how many days a person repeated the plan predicted automaticity instead. Sheeran, Listrom and Gollwitzer pooled 642 tests in 2025 and reported d = 0.36 raw, 0.15 after correction for publication bias. |
| Motivation & ProductivityDoes the Pomodoro Technique Actually Work?The 25-minute interval isn't backed by research — it's a number one person picked, that everyone since has repeated as if it were tested. Still on Google11 of 16 top-ranking pages assert scientific backing for the exact intervalChecked Jul 27, 2026 | R | Cirillo built the method as a student in the late 1980s around a tomato-shaped kitchen timer. The claim that 25 minutes is "scientifically optimal" traces to his own self-published 2007 technical report — no data, no peer review. | Why it stuckA memorable timer plus studies on completely different things (radar-watch fatigue, task-switching) got cited as if they were about the number 25 specifically. How it fell apartThe one randomized trial to test fixed 25-on/5-off breaks directly, in 2025, found no advantage over self-chosen breaks — and faster fatigue. |
| ProcrastinationWhy We Really Procrastinate: The Neurobiology of Avoidance and a Step-by-Step Action ProtocolProcrastination is your nervous system dropping into freeze — a dorsal-vagal shutdown — so you have to calm the body before you can start. The paper this rests on never tested that pathway on an ordinary task. Still on Google13 of 21 top-ranking pages present procrastination as an autonomic freeze state as settled; 0 of 21 note polyvagal theory is contestedChecked Aug 19, 2026 | H | Polyvagal Theory, set out in a 1995 Psychophysiology paper as a theoretical account of how the autonomic nervous system is organised, describes a three-state response sequence; popular writing then read procrastination as its shutdown state. | Why it spreadFeeling frozen, foggy or heavy in front of a task is a familiar sensation, and a nervous-system label turns it into biology rather than character. What is still untestedThe 1995 paper proposes a three-state sequence but runs no test in which an unstructured task produces it, so the step from task to shutdown has no study behind it. Steel's meta-analysis of 691 correlations instead put task aversiveness, task delay, self-efficacy and impulsiveness among the stronger predictors. |
| Habit FormationYour Brain Never Deletes a Bad HabitYour brain never deletes a bad habit, so an old routine stays wired in for life and waits to resurface. A learned response can stay easy to retrieve, which is a weaker statement than permanence. Still on Google13 of 22 top-ranking pages present old habit wiring as permanent and never erased, no caveatChecked Aug 19, 2026 | P | Moore and colleagues, 2026: male mice in an auditory go/no-go task, tracked across thousands of trials with outcome devaluation, dorsolateral-striatum lesions and neural recordings, showed an abrupt switch from goal-directed to habitual control. The permanent-wiring half of the story leans on Lee and colleagues' rat work on incubated cocaine craving and silent synapses. | Why it spreadIt excuses every relapse in advance — if the wiring cannot be undone, failing to quit is anatomy rather than a choice. What actually holdsThe mouse study placed the final transition at roughly three trials in one task, and the rat craving work measured silent synapses at day 1, day 10 and day 45 of withdrawal — neither licenses a human timetable or a universal day-30 craving wave. What survives is narrower: a cue can make an old response easier to retrieve, especially where it was learned. |
| Habit FormationHow Long Does It Take to Form a Habit? The 21-Day Myth vs. the DataIt takes 21 days to form a habit. Measured directly, the median lands closer to 66 days, and gym habits take four to seven months. Still on Google0 of 17 top-ranking pages present 21 days as settled fact; 6 of 17 name Maxwell Maltz as the source of the numberChecked Aug 19, 2026 | R | Maxwell Maltz, a plastic surgeon, wrote in Psycho-Cybernetics (1960) that his patients needed “a minimum of about 21 days” to get used to a new face after surgery. Retellings dropped “minimum” first and “about” second, turning a note on adjusting to a changed body into a law of behaviour change. | Why it stuckTwenty-one days is short enough for motivation to hold out and long enough to sound biological, and it arrived with no study attached for anyone to argue with. How it fell apartPhillippa Lally and colleagues at University College London followed 96 people for 84 days in 2010: among the 39 whose automaticity curves fit well, the median time to 95% of peak was 66 days, with a modeled range of 18–254 days. A 2023 PNAS study of 12 million gym check-ins and 40 million hand-washing scans found hygiene habits settling within weeks and gym attendance taking four to seven months. |
| Energy, Sleep & Body ClockStubborn Fat: What the Science Really SaysTrain the body part you want to slim and you will burn the fat sitting on top of it — crunches for the lower belly, arm work for the arms. One trial has now measured a small local difference, under a workload no ordinary routine matches. Still on Google0 of 18 top-ranking pages present spot reduction as settled fact; 17 of 18 state that training a body part does not remove fat from that areaChecked Aug 19, 2026 | P | Older than its first test: Gwinup, Chelvam and Steinberg, 1971 checked it on tennis players, comparing fat over the dominant arm, worked for years, with fat over the other arm. | Why it spreadAb machines, wraps, creams and problem-zone routines all sell a link between where you work and where you lose, and that zone is where people are most desperate for something to do. What actually holdsThe tennis players carried the same fat on both arms despite years of one-sided work, and six weeks of abdominal training in Vispute et al., 2011 moved ab endurance and no measure of abdominal fat. Brobakken et al., 2023 did record a local difference of about 697 g, roughly 3% of trunk fat, but after 10 weeks and 40 demanding sessions in overweight men — a single finding awaiting replication. |
| Energy, Sleep & Body ClockTraining to Failure: What the Science Actually SaysTo build muscle you have to push every set to failure, because growth lives in the last grinding rep. Training closer to failure may add some size, while strength improves without it. Still on Google0 of 14 top-ranking pages present training to failure as required for growth; 13 of 14 state that stopping short of failure builds muscle tooChecked Aug 19, 2026 | P | Robinson et al., 2024: a meta-regression in Sports Medicine pooled training studies and found muscle growth rising as sets ended closer to failure, a graded dose-response that gym culture compressed into a universal 0-2 RIR target. | Why it spreadThe last rep is the hardest moment of a session, so crediting it with the growth matches what the effort felt like, and it turns lifting into a test of will anyone can choose to pass. What actually holdsRefalo et al. (2023) found a trivial overall advantage for failure and none once the comparison narrowed to true momentary failure; Grgic et al. (2022), across 15 studies, found similar strength whether sets ended at failure or short of it. An eight-week randomised trial in trained lifters (Refalo et al., 2024) produced similar quadriceps growth at 1-2 RIR and at failure. |
| OtherTopic queued — article not yet publishedA row appears here once an article in this category is published and graded. | — | queued | queued |