Zkitszo - The Real
October 08, 2026

Why Lie? I Want Beer- Persuasion in a Split Second

San Francisco, autumn 2003. A Cambridge psychologist is in town for a conference, staying in a cheap hotel in a rough part of the city, and every morning and every night he walks past the same crowd outside... the placards, the soggy cardboard signs, the hard-luck stories. The first couple of days he hands over cash. After that- nothing.

Then a guy he's never seen before turns up holding a dog-eared bit of cardboard with five words on it:

WHY LIE? I WANT BEER!

His hand is in his coat before he's thought about it, past the loose change, going for the wallet. He called it being "legally mugged" afterwards.

Years before that, in London, he'd turned up late to a family dinner, got into a row with his dad the minute he walked in, and stormed out of the restaurant. No way on this earth was he going back. A friend chased him all the way to the tube station in the rain- a big man, close to 28 stone (nearly 400 pounds), who by his own admission had trouble walking. Buttons gone off his coat, his woolly hat lying in a puddle somewhere back down the street. He got ten words in:

"When was the last time you ever saw me run?"

He went looking for an answer and there wasn't one. They'd been friends a long time and he had never once seen the big man run... the longer he looked, the more the anger drained out of him. They walked back to the restaurant and sat down to dinner a second time.

Nobody got talked round there- the decision just flipped, in about the time it takes to read a sign. The psychologist, Kevin Dutton, went on to collect more than 150 of these sudden turnarounds over 18 months and built a book on them: Split-Second Persuasion in Canada in 2010 and the US in 2011, Flipnosis in the UK. "Flipnosis" is his own word for it. He lines the stories up against a big pile of psychology research.. a good chunk of which has had a rough ride since. Replications that failed, numbers that drifted a little with every retelling.

Persuasion Before Words

Ever been woken at five in the morning by a cat that wants breakfast? That isn't its ordinary purr. Researchers at the University of Sussex recorded cats purring while they pestered for food and found a cry tucked inside the low rumble- a peak somewhere between 220 and 520 Hz, close to the pitch of a human baby crying, riding on top of the usual 25 to 30 Hz purr. They played the recordings to 50 people. Even the ones who'd never owned a cat rated these "solicitation purrs" as more urgent and less pleasant than the normal kind, and when the team scrubbed the cry out digitally, the urgency ratings dropped.

Biologists call that kind of thing a key stimulus- a trigger that sets off a built-in response just by being there, no thinking involved. Herring gull chicks peck at the red spot on a parent's yellow beak, the pecking gets the parent to cough up food, the chick eats. In Niko Tinbergen's experiments around 1950 the chicks barely cared what the rest of the "parent" looked like. A thin red rod with three white bands near the tip got pecked harder than a realistic model of a gull's head. (The book has the colors the other way round, a brown stick with red stripes.) A fake that beats the original is a supernormal stimulus. We've got plenty of those aimed at us- the six-pack and the cinched waist, for starters.

Babies arrive with three triggers already loaded: the cry, the face and the eye contact. The cry slides about in pitch without warning, so there's no tuning it out. The face- round head, big eyes set low, small nose and chin- is what Konrad Lorenz called Kindchenschema, the "baby schema". A 2008 brain-scan study found adult brains flag a baby's face as special in about 130 milliseconds, an eighth of a second or so.

In 2008 a psychologist had 240 wallets dropped around the streets of Edinburgh, most of them with a photo showing through a clear window. 88% of the ones with a baby photo were posted back. A puppy got 53%, a family 48%, an elderly couple 28%, and wallets with no photo at all 15%. No journal paper on it turned up... it ran as a newspaper story, with roughly 40 wallets per photo going by the totals.

SPICE- Five Ingredients in One Line

Dutton treats the split-second line as a grown-up cousin of those triggers- an analogy, not something anyone has measured. He handed his collection of turnarounds to a group of volunteers and asked them to write down what made each one work. Five things kept coming back: simplicity, perceived self-interest, incongruity, confidence and empathy. S, P, I, C, E.. SPICE.

Simplicity means short enough to slap on the front of a T-shirt. "When was the last time you ever saw me run?" also happens to be a question, so the angry friend had to go off and find an answer- the anger cooled while he was busy looking. Our brains like things that are easy to process, to the point of rating a rhyme as more true: people judged "What sobriety conceals, alcohol reveals" more accurate than the same saying with the rhyme taken out (McGlone and Tofighbakhsh, 2000). The effect shrank when people were told to judge the meaning and ignore the poetry. Johnnie Cochran knew the trick anyway at the O.J. Simpson trial- "If it doesn't fit, you must acquit."

Perceived self-interest is what's in it for them, as they see it, which isn't always the same as what's true. Terry Wogan, grilled about an £800,000 salary for his radio show, worked it out at two pence per listener and called himself cheap at the price.

Incongruity is the script break: the beggar who owns up to wanting beer when every other sign on the street says hungry. Confidence is the flat, unhurried delivery with no "sort of" anywhere in it. Empathy makes it sound like it came from their side of the fence- same group, same language, same problem.

Dutton maps the five onto an older set, from a con man who called his method the "three As": attention (what the brain lets in), approach (how it weighs things up and decides) and affiliation (whose side it ends up on). In his version simplicity and incongruity grab the attention, confidence steers the approach, self-interest and empathy take care of affiliation. The con man, "Keith Barrett", is a composite of several real people- the author says so himself in a note at the front.

SPICE, and where Dutton says each ingredient lands grouped by which of a con man's three As they feed, so the acronym reads S, I, C, P, E S Simplicity short enough for a T-shirt I Incongruity breaks the script they expected C Confidence said flat, no hedging P Perceived self-interest what they get, as they see it E Empathy sounds like it came from their side Attention what the brain lets in Approach how it weighs and decides Affiliation whose side it ends up on A sorting of 150+ collected turnarounds, with the factors named by volunteers. Not a lab-tested model.

SPICE is a sorting of one person's story collection, with the factors named by volunteers. As far as could be found, nobody has tested it as a model in controlled experiments, and the only write-up outside the book is Dutton's own 2010 article in Scientific American Mind.

Incongruity Kicks the Door In

When something happens that the brain didn't predict, the surprise circuits fire- the amygdala, the brain's alarm bell, and the anterior cingulate cortex, a strip of brain that flags conflict and "huh?". For a beat everything else waits while it sorts out what just happened. The book's theory is that whatever arrives in that gap gets through with less checking than usual. It borrows a karate word for the beat, teishin, a "stopping mind".. a term that's hard to find in karate writing outside the book.

Join them instead of fighting them

Friday night on the London Underground, a packed Piccadilly line train held in a tunnel by a signal failure, and people are getting restless. A guy in a tracksuit sparks up a cigarette right under the No Smoking signs (smoking on Tube trains was banned back in 1984; the book dates the ban to the King's Cross fire of 1987). The whole carriage goes quiet. Nobody says a word.

Then a man in a pinstripe suit leans over with a cigarette of his own and asks the smoker if he's got a light. For another passenger that's the last straw- he snaps at the pair of them that you can't smoke in here. The suit "notices" the signs, says sorry, and turns to the smoker: maybe they'd both better put them out. The smoker puts his out.

Challenge someone like that head on and he has to win or lose face in front of the whole carriage. Ask him for a light and suddenly there are two offenders, a gang of two. The request to stop then comes from a fellow offender instead of a stranger telling him off.

Answer inside their frame

The book recalls a story from a local London paper: a drunk gets on a bus, can't find a seat, and shouts a racist insult at an elderly man to get him to move- an insult that also called him fat. The man's answer was four words, "You calling me fat?" The bus erupted laughing and the drunk got off.

Two builders on a train mocking a man in a suit for never doing a real day's work got the same kind of treatment, with one extra step. He asked which job they'd rather have, theirs or his. One of them said he wouldn't be him for a second... so what were they complaining about?

A big move to hide the real one

Ron Cooper, a cop 23 years in, is first on scene when a young man gets up on the roof of a multi-storey car park, 14 floors up, threatening to jump. Cooper asks if he minds him taking his jacket off- running up all those stairs has made him hot. The man couldn't care less. So off comes the coat, then the tie, then the shirt, each one folded up neatly, with the wind tearing at him and sleet coming down, until Cooper is down to a T-shirt reading PISS OFF- I'VE GOT ENOUGH FRIENDS! He gets up on the ledge, turns so the slogan is facing the man, and asks him if he wants to talk about it or what. The man saw the funny side.

Magicians work the same way, a big obvious action covering the small one that matters. The "hot from the stairs" excuse made the undressing look ordinary.

Two beats, then break the third

Jokes and music both lean on an "AAB" pattern- set something up twice, break it on the third. Paul Rozin and colleagues documented it across both in 2006. An Englishman, an Irishman and a Scotsman walk into a bar...

Let them show you

The story told about the British barrister F. E. Smith: defending a bus driver against a passenger who claimed his arm had been injured, Smith skipped the aggressive cross-examination and went all gentle. Would the gentleman show the court how high he could lift the arm now? Shoulder height, with a wince. And how high could he lift it before the accident? Straight up over his head.

Disrupt, then reframe

Door-to-door sellers in a 1999 study by Barbara Davis and Eric Knowles offered packs of Christmas cards for a charity. The normal pitch gave the price as $3. The odd pitch said the cards sold for 300 pennies.. a short pause.. "That's $3. It's a bargain!" Later papers report 65% buying with the odd pitch in the first study, against 35% with the plain one. The idea is that the weird unit trips the brain for a second while "It's a bargain!" slips in.

A French team ran it again in 2015, with scouts selling cards for their group, 50 people per version, prices said in cents with euros as the plain version- and they tested each half on its own too. Cents plus the bargain line sold to 62%, against 36% for the plain pitch. Cents with no bargain line got 42%, the bargain line with an ordinary price 34%, and the bargain line said before the cents 46%. Only the full version, in that order, clearly beat the plain pitch.

The book illustrates it with "36 Christmas cards for 2,844 cents", which is Dutton's own example- every account of the actual study gives the price as $3, quoted as 300 pennies. The book's other example, cupcakes sold as "half-cakes", couldn't be traced to any study.

Small Words, Bigger Yes

"Will you...?" and then stop talking

Gordon Sinclair ran a restaurant in Chicago with a no-show problem, around 30% of bookings. His receptionist used to say "Please call if you have to change your plans." He had her ask "Will you please call if you have to change your plans?" instead- and then wait, quietly, until the caller said yes. No-shows reportedly fell to about 10%. It gets passed around as a study, but it's one restaurant owner's experience as retold by the influence researcher Robert Cialdini, with no published data turning up behind it.

The yes is a small promise, said out loud, to a real person. Breaking it later means going back on our own word. Most of us really hate being inconsistent with ourselves... and on the phone, somebody has to fill a silence.

Chris Voss, the ex-FBI hostage negotiator, sees yes differently. In his book Never Split the Difference he splits it three ways: the counterfeit yes (said to get rid of you), the confirmation yes (a reflex answer to a black-and-white question) and the commitment yes, the real one that leads to action. In his view people have been walked into so many yes-ladders by salespeople that a yes on its own is worth very little, so he'd rather flip the yes to a no or go for "that's right". Sinclair's question is fishing for the third kind- a tiny, specific promise in the caller's own interest, asked once.

Foot in the door, door in the face

In 1966 Jonathan Freedman and Scott Fraser had householders in Palo Alto asked to put a huge, badly lettered DRIVE CAREFULLY sign up on their front lawn- shown a photo of one on somebody else's house. Asked cold, 4 out of 24 said yes, about 17%. Two weeks earlier a different researcher had asked other households to put a little three-inch BE A SAFE DRIVER sign in a window. Of those, 76% then said yes to the big ugly one. The small first yes was on the same theme as the big ask. The book says 73% turned down the cold request where the paper has 83%... and the 76% is the best of four groups, the other three landing around 48%. Reviews since have found the effect real but small.

In 1975 Cialdini's team stopped students on campus and asked them to chaperone a group of young offenders on a day trip to the zoo- 17% agreed, as later sources report the paper. Other students were first asked to volunteer as a counselor two hours a week for the next two years. They said no, of course. When the zoo trip came up next, 50% agreed. A concession from the asker seems to pull a concession out of the person being asked. A German team ran it again in 2021 with similar results, but a 2012 review of dozens of these studies found it gets people to say yes more often than it gets them to actually do the thing.

Agree first, hear the catch after

Cialdini's low-ball study came out in 1978. Students were asked to sign up for a psychology experiment. Told up front that it started at 7 a.m., 24% signed up and turned up. Asked first, and only told about the 7 a.m. start once they'd already said yes, 53% turned up- 18 of 34 students, at seven in the morning! Car showrooms run the same play.. the great price gets the decision, then the extras appear. Across later studies the effect holds on average, though a few small ones came out flat or even backwards. When the terms change after a yes, it's worth deciding again from scratch, as if that first yes never happened.

Turn back and give a little more

Restaurant chocolates, 2002. In a small study at an Italian restaurant in New Jersey- one server, about 20 tables per condition- the server either brought nothing with the bill or offered a basket of little chocolates. With no chocolate, tips averaged 18.95% of the bill. One chocolate each nudged that to 19.59%, two each to 21.62%. The trick version was one chocolate each, then the server heading off, turning back as if she'd had a second thought, and offering everyone another pick- 22.99%. The book reports that last one as tips going up 23%.. 23% was the tip itself, the rise was about 21%.

The con man in the book swears by a cousin of this one: ask while walking away, so the other person has to call you back to give, though there's no data on it beyond his word.

The word "because"

Ellen Langer's 1978 photocopier study ran in a university library. Someone asks to cut in at the copier: "Excuse me, I have 5 pages. May I use the xerox machine?" That got 60% to step aside. Adding "because I'm in a rush" got 94%. Adding "because I have to make copies"- which says nothing at all, everyone at a copier has to make copies- still got 93%, as good as a real reason.

Bump it to 20 pages, though, and with no reason only 24% let them in. The empty "because I have to make copies" got exactly the same 24%. Only a real reason helped, and only up to 42%. The book tells the first half and skips the second... and a 1985 attempt to repeat the study didn't get the same results either.

Plain ask vs the same ask with the technique % who agreed, bought or showed up. Solid dots: the original study. Hollow dots: a later re-run. plain ask with the technique 0% 25% 50% 75% 100% Foot in the door: 17% with the plain ask, 76% with the technique (Freedman & Fraser 1966, best of 4 groups) Foot in the door Freedman & Fraser 1966 · best of 4 groups 17% 76% Low-ball: 24% with the plain ask, 53% with the technique (Cialdini et al. 1978, showed up at 7 a.m.) Low-ball Cialdini et al. 1978 · showed up at 7 a.m. 24% 53% Door in the face: 17% with the plain ask, 50% with the technique (Cialdini et al. 1975, zoo trip) Door in the face Cialdini et al. 1975 · zoo trip 17% 50% Disrupt, then reframe: 35% with the plain ask, 65% with the technique (Davis & Knowles 1999, 300 pennies) Disrupt, then reframe Davis & Knowles 1999 · 300 pennies 35% 65% Disrupt, then reframe: 36% with the plain ask, 62% with the technique (Saint-Bauzel et al. 2015, French re-run) Disrupt, then reframe Saint-Bauzel et al. 2015 · French re-run 36% 62% Empty "because", 5 pages: 60% with the plain ask, 93% with the technique (Langer et al. 1978, copier queue) Empty "because", 5 pages Langer et al. 1978 · copier queue 60% 93% Empty "because", 20 pages: 24% with the plain ask, 24% with the technique (Langer et al. 1978, copier queue) Empty "because", 20 pages Langer et al. 1978 · copier queue 24% both ways Towel card, US hotel: 37% with the plain ask, 49% with the technique (Goldstein et al. 2008, "this room") Towel card, US hotel Goldstein et al. 2008 · "this room" 37% 49% Towel card, German hotel: 93% with the plain ask, 77% with the technique (Bohner & Schlüter 2014, re-run) Towel card, German hotel Bohner & Schlüter 2014 · re-run 77% 93% Most rows rest on 15 to 50 people per group; the towel studies had hundreds of guests. The 1999 and 1975 numbers come from later papers quoting the originals. The German re-run went backwards.

Numbers and Words That Steer

Put a number in the room first and every estimate after it drifts towards it- that's anchoring. Ever noticed how the "was" price on a sale tag is always so high? In a 2006 German study, 23 judges and 19 prosecutors, about ten years on the job on average, read a rape case file and took a phone call from a "journalist" asking whether the sentence would be higher or lower than three years. They went on to recommend 33 months on average. The ones asked "higher or lower than one year?" recommended 25.

In a third version, 52 young lawyers in judicial training set the prosecutor's demand in a shoplifting case by rolling a pair of dice themselves. The dice were loaded to come up 3 or 9 months... and their sentences still followed the dice, 5.3 months on probation against 7.8!

Precise numbers seem to anchor harder than round ones. According to a 2008 paper, as Wikipedia summarizes it, people adjust less away from a price like $4,988 than from $5,000, and in 12,581 Florida home sales, houses listed at precise prices sold closer to their asking price than round-number ones. The main effect has reportedly held up in two re-runs since, though some of the paper's finer claims haven't.

Elizabeth Loftus and John Palmer showed students films of car crashes in 1974 and asked how fast the cars were going when they "smashed" into each other: 40.8 mph on average. Asked with "contacted", it was 31.8. A week later 32% of the "smashed" group remembered broken glass at the scene, against 14% asked with "hit" and 12% who hadn't been asked about speed at all. There was no broken glass in the film. That one is less anchoring than a leading question, a question worded to push the answer one way.

Political pollster Frank Luntz built a career on picking words. A 2002 memo of his told Republicans to stop saying "global warming" and say "climate change", which he called less frightening, and to keep making the lack of scientific certainty a primary issue in the debate. In 2019 he told a panel of Senate Democrats on climate that he'd been wrong.

During the O.J. Simpson trial, defense lawyer Alan Dershowitz argued that fewer than 1 in 2,500 men who beat their partners go on to murder them. Roughly true.. and beside the point, since Nicole Brown Simpson had already been murdered. The number that matters is how often the abuser was the killer, given a battered woman who was murdered. Working it per 100,000 battered women, about 40 a year are killed by their abuser and about 5 by somebody else, so the abuser did it roughly 90% of the time. That's still not the chance O.J. did it- it's only where the reasoning starts, before any of the actual evidence in the case gets weighed. Mixing up "the chance of A given B" with "the chance of B given A" is the same trap that turns up with Bayes and base rates (how common something is to begin with) in Think Clearly to a Super You.

The Crowd Gets a Vote

Hotel towel cards! Cialdini's group has run these more than once. In their first hotel test, written up in 2007, a card asking guests to reuse towels to help the hotel save energy got under 16% of guests doing it, while the other cards averaged about 30%. In the 2008 paper, the usual "help save the environment" card got 35% reusing their towels and a card saying most guests reuse theirs got 44%. In a bigger follow-up with 1,595 guests, the standard card got 37% and "most guests who stayed in this room" got 49%. Copying what people like us do in the same spot is social proof. It pulls hardest when we're unsure how to act.

Two German hotels tried it in 2014 and it fell over. In the first, guests reused at around 82% whatever the card said, and the "this room" version did worse than the general one. In the second, the plain environmental card beat the guests-like-you card, 93% to 77%- though that plain-card group was only 30 people. One pooled re-analysis put the norm card's average boost at about six percentage points; a critique of it called the evidence behind even that weak.

Solomon Asch's line studies in the 1950s put one real volunteer in a room of actors, everyone calling out which of three lines matched a fourth, the actors all giving the same wrong answer on purpose. About three in four volunteers went along with the wrong answer at least once, and overall they went along on roughly a third of the rigged rounds. Three actors were enough- adding more barely changed anything. One actor breaking ranks and giving the right answer cut the conforming to about a quarter of what it had been.

In 1968 Stanley Milgram had people stand on a busy sidewalk and stare up at a sixth-floor window. One person staring got 42% of passers-by to look up and 4% to stop. Fifteen people staring got 86% looking up and 40% stopping- the book's version, 40% rising to 80%, doesn't match either measure.

In a 1975 study on labels, fifth graders in one class were told again and again that they were neat and tidy people. Another class got told they should be neat and tidy. The "you are" class ended up littering less than the "you should" class, with the gap reportedly still there at a two-week check- though with one class per condition, the teacher and the class itself could be part of that.

Put a group of like-minded people in a room to talk something over and they tend to come out further in the direction they already leaned- more cautious if they started cautious, riskier if they started risky. That's group polarization. Irving Janis's groupthink is the bigger version: a tight team that feels invulnerable, waves off warnings, leans on whoever dissents, keeps its own doubts quiet and reads the silence as everyone agreeing. Janis built it mostly from case studies of policy disasters, and lab support for the full package is patchy.

A small, steady minority can move a group too, more slowly, from the inside. Serge Moscovici's famous 1980 version had people looking at blue slides while a stooge kept calling them green, then measured the color of their afterimage- the ghost color left on a white wall after staring at something. That afterimage result didn't hold up cleanly in a 1998 series of re-runs. The broader idea- a steady minority changing minds privately while a majority mostly gets public lip service- is still being studied.

Confidence, Cold Empathy and the Lab Coat

Groups tend to go with their most confident member, right or wrong. Ratings of 30-second silent video clips of college teachers- just how they looked and moved, no sound- correlated with the end-of-term ratings from their real students.

Stanley Milgram's obedience experiments usually get told with a professor in a white lab coat at Yale. The man giving the orders was a 31-year-old high school biology teacher in a gray technician's coat. At Yale, 26 of 40 volunteers (65%) kept going to the highest shock on the board. Moved to a shabby rented office in Bridgeport, Connecticut, with no Yale name on the door, 19 of 40 (47.5%) still did. Pulling Milgram's archive together in 2014, Alex Haslam and colleagues found obedience swung a long way depending on the set-up- with an ordinary man giving the orders instead of the experimenter, it was 20%. The book has a technician in jeans and sneakers taking over, and a downtown office dropping obedience to 25%. Neither turns up in Milgram's variations; the nearest are that 20% and the Bridgeport 47.5%.

A 2008 study reported that a picture of a brain scan made a news article about neuroscience more convincing. Ten re-runs with nearly 2,000 people in 2013 found almost nothing: 0.07 of a point on a four-point scale.

In a 1977 Stanford study, pairs of students drew lots- one wrote hard trivia questions from their own knowledge, the other tried to answer them. People watching rated the question-writer at 82 on general knowledge and the contestant at 49, where 50 is the average student. The contestants rated the question-writer 67 and themselves 41, even though the roles came out of a draw. Psychologists call that the fundamental attribution error: putting what people do down to who they are instead of the situation they're in.

The book files the 1973 Good Samaritan study under self-interest. Seminary students on their way to give a talk, some of them about the Good Samaritan, passed a man slumped and coughing in a doorway. With time to spare, 63% offered help. Told they were already late, 10% did (Darley and Batson, 40 students). The study itself was about hurry, which puts it closer to the situation-beats-character side.

The empathy in SPICE comes in two kinds. One is similarity- shrink the distance. Faces digitally morphed a little towards a person's own face got trusted more in a money-sharing game (Lisa DeBruine, 2002). A shared birthday, a shared hometown, a name spelled the same odd way as your mom's.. Cialdini added a seventh principle to his famous six in his 2016 book Pre-Suasion and called it unity: once people see us as one of their own, everything else in persuasion gets easier. The other kind is salience- saying it in their language, with their examples.

Psychopaths turn out to be fine at the reading part, on one test anyway. Shown photos of just the eye region of faces and asked what the person was feeling, prisoners scored as psychopaths did no worse than other prisoners (Richell and colleagues, 2003). A 2012 meta-analysis found them weaker at recognizing emotions across the board, though, fear and sadness included. The usual name for reading feelings without sharing them is cold empathy.

Believe First, Check Later

Descartes thought we understand an idea first and then decide whether to believe it. Spinoza thought understanding is believing, with doubt coming after as extra work. Daniel Gilbert tested it in 1993. People read crime reports in which some statements were marked as false, shown in a different color, then recommended a prison sentence. Some readers also had to hunt for the digit 5 in a stream of numbers at the same time, to keep their brains busy. Busy readers whose false statements made the crime sound worse- statements they'd been told were false- gave 11.15 years on average. When the false statements made it sound better, 5.83. Readers with nothing else to do gave 7.03 and 6.03.

Dutton's line for it is that a persuader's job is to "stop them unbelieving". A 2005 study narrowed that down, though: distraction only made people forget a statement was false when the false version told them nothing useful. When knowing it was false carried real information, they kept track just fine.

Leon Festinger and James Carlsmith paid students $1 or $20 to tell the next person that a mind-numbingly boring task had been fun. Afterwards the $1 group rated the task as more enjoyable (+1.35 on a scale from −5 to +5) than the $20 group (−0.05) or students who hadn't lied at all (−0.45). Twenty dollars was reason enough to lie. One dollar wasn't, so the task must have been sort of OK.. that's cognitive dissonance, the discomfort of holding two clashing ideas, settled by changing whichever is cheaper to change. Festinger's team had also gone undercover in a 1950s UFO group whose prophesied end of the world didn't come. By their account, the core members went out recruiting harder afterwards.

What tends to work is pitching inside what someone can already accept and leaving them a way to change their mind without losing face. The book takes it as far as a man who was sure the face in his mirror belonged to a stranger. The neuropsychologist who studied him told Dutton the mistake people make is thinking the answer lies in logic.... no argument moved him.

When It's Pointed at Us

Social engineering runs on the same parts. The US cybersecurity agency CISA describes it as using human interaction to get information or access out of an organization. Chris Hadnagy's social engineering framework lists eight influence tactics built straight on Cialdini's work- reciprocity, obligation, concession, scarcity, authority, commitment and consistency, liking and social proof. The UK's NCSC points out that people click on phishing (fake messages dressed up as someone trustworthy) for ordinary reasons, like being busy and stressed at the time.

A uniform, a piece of paper flashed at the right moment, "it's all been arranged", plus some other demand on attention at the exact same time.. the book's distraction-theft stories run on that combination. Every piece eats some of the attention that would otherwise go on checking the story.

A 1975 study had students sentence a woman for either a burglary or a swindle. With an attractive photo attached she got off lighter for the burglary and harder for the swindle, the crime where her looks were the tool. Later studies haven't always found the swindle half, so treat that part as a maybe.

The book also describes a young con man's dating routine: turn up at a stranger's door on a Friday evening, dressed up, with wine and a restaurant booking, ask for a woman who doesn't live there, act crushed, make a joke at his own expense... and since he's here anyway, would she like to join him? Every ingredient of SPICE is in there- simple, a free dinner in it for her, unexpected, confident, and all on her side.

Try These on Friends

Lay four cards out on the table showing 3, 8, RED and BROWN, and tell your friends every card has a number on one side and a color on the other. The rule to test is: if a card has an even number on one side, it's red on the other. Ask them which cards they'd need to turn over to check it. Lots of people go for RED, which can't break the rule whatever's on the back. The only two that can are the 8 (anything but red behind it breaks the rule) and BROWN (an even number behind it breaks the rule). In Peter Wason's original 1966 version fewer than one person in ten got it right.. most of us look for cards that could confirm the rule rather than the ones that could break it. Turn it into a bar rule- "if you're drinking alcohol, you must be over 18"- and most people get it straight away.

For the next one, have everyone read this once and count the letter F:

FINISHED FILES ARE THE RESULT OF YEARS OF SCIENTIFIC STUDY COMBINED WITH THE EXPERIENCE OF YEARS.

There are six. Plenty of people stop at three... the eye skates right over every "of".

Then read this list out once at a steady pace and get everyone to write down every word they can remember:

sour, candy, sugar, bitter, good, taste, tooth, nice, honey, soda, chocolate, heart, cake, tart, pie

A lot of people will write down "sweet". It was never on the list! Psychologists call it the Deese-Roediger-McDermott effect, a false memory built out of all the words around the one that's missing.

If there's a coin handy, auction off a £1 coin- or a dollar- to the room, bids going up a penny at a time. Highest bid wins the coin. The second-highest bidder pays what they bid and gets nothing. Bidding tends to sail right past the coin's value, since the runner-up is always better off topping the leader than eating their own bid. The economist Martin Shubik came up with that one in 1971.

Asch's warm/cold test from 1946 needs two groups who don't talk to each other. Read one group this description of a person: intelligent, skillful, industrious, warm, determined, practical, cautious. Read the second group the same list with "cold" instead of "warm". Then ask everyone whether the person sounds generous or ungenerous, happy or unhappy, good-natured or irritable. In Asch's original, 91% of the "warm" group said generous, against 8% of the "cold" group.

For a look at our own suggestibility, the Multidimensional Iowa Suggestibility Scale is free to use from its authors with no permission needed, questions and scoring included in their brief manual. The Reading the Mind in the Eyes test comes from the Autism Research Centre at Cambridge, which offers it as downloadable PDFs for research use, and a free interactive version lives on a third-party site, psytests.org- not run by the center itself.

What Didn't Hold Up

Social psychology's replication crisis- the run of famous findings that didn't repeat when other labs tried them- got going a year or so after the book came out. Some of what the book leans on took a hit. Some of it was never quite what people said it was in the first place.

The famous split of 7% words, 38% tone of voice and 55% face comes from Albert Mehrabian's 1960s studies of how people read liking from tone of voice and facial expression. His own page says that unless someone is talking about their feelings or attitudes, the equations don't apply.

"Drunk-tank pink" (Baker-Miller pink), the bubblegum color painted in holding cells to calm people down, got a proper controlled test in 2015 with 59 prisoners in Swiss cells. Pink cells and plain cells came out the same.

Learned helplessness- giving up on escaping because escape has seemed pointless before- didn't fail so much as flip. Steven Maier and Martin Seligman, who discovered it, wrote in 2016 that the original theory "got it backward". Passivity under long stress turned out to be the default response; what gets learned is control.

The book's identity-priming examples, like Asian women doing better at math when reminded they're Asian and worse when reminded they're women, rest on one or two small studies each. The wider idea, stereotype threat, has had a bumpy run. A 2015 meta-analysis (a study that pools many studies) of gender stereotype threat in schoolgirls found signs of publication bias- positive results getting published, null ones getting filed away. A pre-registered Dutch study of 2,064 students, with its methods locked in before any data came in, found no effect on girls' math scores. The "Obama effect" on test scores the book mentions was contradicted by another study in the very same journal issue.

Growth mindset is the belief that ability can be built, and the belief itself is fine. Short classroom mindset lessons add very little on average- a 2023 meta-analysis put the effect at close to zero once publication bias was accounted for- while a big 2019 US study found a bump of 0.10 grade points, only for lower-achieving students.

The brain-scan study of political partisans reading about their own candidate contradicting himself, from the 2004 election, had 30 people in it. Reading feelings off which brain areas light up- reverse inference- is weaker evidence than it sounds, since most brain areas do many jobs.

Has a line ever flipped you on the spot, the way that beer sign did? Or have you pulled one off yourself and watched someone's whole face change? Leave it in the comments.. the exact words, if you can remember 'em.

Glossary

  • AAB pattern — two beats that set up an expectation and a third that breaks it; common in jokes and music.
  • Afterimage — the ghost color you see on a white surface after staring at a colored one for a while.
  • Amygdala — a small pair of brain structures involved in alarm, surprise and emotional reactions.
  • Anchoring — the pull of the first number mentioned on every estimate that follows it, even when the number is random.
  • Anterior cingulate cortex — a strip of brain along the middle of the frontal lobes that flags conflict, errors and surprises.
  • Base rate — how common something is to begin with, before any other evidence is weighed.
  • Baker-Miller pink — the bubblegum "drunk-tank pink" once painted in holding cells on the claim that it calms people.
  • Cognitive dissonance — the discomfort of holding two clashing beliefs, usually settled by changing whichever one is cheaper to change.
  • Cold empathy — working out what someone else feels without feeling it yourself.
  • Conditional probability — the chance of one thing given that another is already known; the chance of A given B is not the chance of B given A.
  • Conformity — going along with a group's answer or behavior, even against your own judgment.
  • Disrupt-then-reframe — a small oddity in a request (300 pennies instead of $3) followed straight away by the persuasive line ("It's a bargain!").
  • Dollar auction — an auction where the runner-up also pays their bid, so bidding keeps going past the prize's value.
  • Door-in-the-face — asking for something big you expect to be refused, then asking for what you actually want.
  • DRM effect — a false memory of a word that was never on a list, triggered by all the related words that were (Deese-Roediger-McDermott).
  • Foot-in-the-door — getting a small yes first so a later, bigger request on the same theme is more likely to get one too.
  • Fundamental attribution error — explaining what people do by their character instead of their situation.
  • Group polarization — a like-minded group coming out of a discussion more extreme in the direction it already leaned.
  • Groupthink — a tight group's drive for agreement overriding a realistic look at the problem.
  • Growth mindset — the belief that ability can be built with effort, as opposed to fixed at birth.
  • Key stimulus — a trigger that sets off a built-in behavior just by being present, like the red spot on a gull's beak.
  • Kindchenschema — "baby schema": the round head, big low-set eyes and small nose and chin that adults find cute.
  • Leading question — a question worded to push the answer one way.
  • Learned helplessness — giving up trying to escape a bad situation after earlier attempts seemed pointless.
  • Low-ball — getting a yes on attractive terms, then changing the terms after the decision is made.
  • Meta-analysis — a study that pools the results of many earlier studies to estimate an overall effect.
  • Minority influence — a small, consistent group changing the private views of a larger one.
  • Phishing — messages that impersonate someone trustworthy to get a click, a password or a payment.
  • Placebic reason — a "because" that sounds like a reason but adds no information ("because I have to make copies").
  • Pre-registered study — a study whose method and analysis are published before the data are collected, so the results can't be massaged afterwards.
  • Publication bias — positive results getting published more often than null ones, which inflates how strong an effect looks.
  • Reading the Mind in the Eyes test — a test of judging what someone feels from photos of only their eye region.
  • Replication crisis — the discovery, from around 2011, that many well-known psychology findings did not repeat when tested again.
  • Reverse inference — reading a mental state off which brain area lights up; weak, because most areas do many jobs.
  • Rhyme-as-reason effect — rhyming statements being judged as more true than the same statements without the rhyme.
  • Social engineering — manipulating people, rather than machines, into handing over information or access.
  • Social proof — doing what we see people like us doing, especially when unsure how to act.
  • Solicitation purr — the purr cats use when pestering for food, with a cry-like sound hidden inside it.
  • SPICE — Dutton's five ingredients of split-second persuasion: simplicity, perceived self-interest, incongruity, confidence, empathy.
  • Stereotype threat — the idea that being reminded of a negative stereotype about your group drags down your performance.
  • Supernormal stimulus — an exaggerated fake that triggers a built-in response more strongly than the real thing.
  • Teishin — "stopping mind"; the book's word, credited to karate, for the frozen beat after a surprise.
  • Three As — attention, approach and affiliation; the con man's map of where influence works.
  • Unity — Cialdini's seventh principle: people say yes more easily to those they see as one of their own.
  • Wason selection task — the four-card logic puzzle that shows people looking for confirmation instead of for what would prove them wrong.

Sources

Comments

Post a Comment

← Newer