8 Truths About Intuition
8 Truths About Intuition
By Matthew Hutson published December 19, 2019 - last reviewed on January 29, 2020
In 1997, IBM’s Deep Blue supercomputer beat the world’s best chess player, Gary Kasparov.
Immediately, some people noted that we were still superior at the ancient Chinese game of Go.
Go, they reasoned, would remain out of robots’ reach for the foreseeable future. Its plurality of
possible moves and the nuances in evaluating even who’s winning put it further out of the realm
of rote combinatorics and into the orbit of intuition, supposedly humanity’s specialty. “It may be
a hundred years before a computer beats humans at Go,” a Princeton astrophysicist told The
New York Times soon after the Kasparov match. “Maybe even longer.” If you follow the news,
you know it took until 2016.
What does the success of DeepMind’s AlphaGo against the best meat-based players say about
intuition, human or otherwise? On one hand, it knocks down some highfalutin’ claims about this
special sense, revealing it to be, as some psychologists have long held, nothing more than pattern
recognition.
On the other, “pattern recognition” does not do full justice to the many patterns intuition
recognizes. Most of human behavior happens automatically, guided by genetics and habit rather
than conscious deliberation. “You could not get by if you walked into a restaurant and you had to
reconstruct from first principles how to behave,” says Valerie Thompson, a psychologist at the
University of Saskatchewan in Canada.
Even for more complex problems, intuition drives decisions, says Gerd Gigerenzer, a
psychologist at the Max Planck Institute for Human Development in Berlin. In working with top
executives at the largest German firms, he finds that “they go through all the data they have—
and they’re buried under data—and at the end the data don’t tell them what they should do.”
Intuition, he says, “is a form of unconscious intelligence that is as needed as conscious
intelligence.”
Despite intuition’s ubiquity, we harbor many mistaken intuitions about intuition. Here we’ll
consider eight facets of unconscious processing—including its application to creativity, morality,
and social interaction—looking at what it does well, where it fails, who uses it, when we trust it,
and how to improve it. Building Deep Blue and AlphaGo required a lot of hard, deliberate
thought, but it also required loads of human intuition and insight. The fact that we hacked
together machines to beat us in a couple of small corners of our own game proves, if nothing
else, that we can hack our own intuitions, too.
A body of research reveals that intuition can be not only faster than reflection but also more
accurate.
We’re fairly good at judging people based on first impressions, thin slices of experience ranging
from a glimpse of a photo to a five-minute interaction, and deliberation can be not only
extraneous but intrusive. In one study of the ability she dubbed “thin slicing,” the late
psychologist Nalini Ambady asked participants to watch silent 10-second video clips of
professors and to rate the instructor’s overall effectiveness. Their ratings correlated strongly with
students’ end-of-semester ratings. Another set of participants had to count backward from 1,000
by nines as they watched the clips, occupying their conscious working memory. Their ratings
were just as accurate, demonstrating the intuitive nature of the social processing.
Critically, another group was asked to spend a minute writing down reasons for their judgment,
before giving the rating. Accuracy dropped dramatically. Ambady suspected that deliberation
focused them on vivid but misleading cues, such as certain gestures or utterances, rather than
letting the complex interplay of subtle signals form a holistic impression. She found similar
interference when participants watched 15-second clips of pairs of people and judged whether
they were strangers, friends, or dating partners.
Other research shows we’re better at detecting deception and sexual orientation from thin slices
when we rely on intuition instead of reflection. “It’s as if you’re driving a stick shift,” says Judith
Hall, a psychologist at Northeastern University, “and if you start thinking about it too much, you
can’t remember what you’re doing. But if you go on automatic pilot, you’re fine. Much of our
social life is like that.”
Thinking too much can also harm our ability to form preferences. College students’ ratings of
strawberry jams and college courses aligned better with experts’ opinions when the students
weren’t asked to analyze their rationale. And people made car-buying decisions that were both
objectively better and more personally satisfying when asked to focus on their feelings rather
than on details, but only if the decision was complex—when they had a lot of information to
process.
Intuition’s special powers are unleashed only in certain circumstances. In one study,
participants completed a battery of eight tasks, including four that tapped reflective thinking
(discerning rules, comprehending vocabulary) and four that tapped intuition and creativity
(generating new products or figures of speech). Then they rated the degree to which they had
used intuition (“gut feelings,” “hunches,” “my heart”). Use of their gut hurt their performance on
the first four tasks, as expected, and helped them on the rest. Sometimes the heart is smarter
than the head.
Lou Brooks, used with permission
Our dreams, those unwilled visions of the night, hold a powerful aura of truth we can’t quite
extinguish. People report they’re more likely to change their travel plans if they dreamed about a
plane crash than if the government announced an actual travel warning. And test-takers can’t
shake the “first instinct fallacy.” Three in four college students reported that when reconsidering
an answer on an exam, their initial choice will usually turn out to be correct. But when erase
marks on actual exams were analyzed, the reverse was true: Twice as many changed answers
went from wrong to right as right to wrong.
“In general,” says psychologist Sascha Topolinski of the University of Cologne in Germany,
“intuition is something emotional that makes you confident in an idea. ‘You cannot take away
this feeling from me. I do not trust this car seller. I can’t tell you why, but I’m confident I don’t
like him.’”
Intuition about the accuracy of an intuition is even more fallible. When people were asked to rate
their confidence that their “gut feelings” had steered them skillfully on a test, confidence ratings
had no relationship with actual performance.
Even when we acknowledge the absurdity of an intuition, we often stick with it. Consider
superstitions. I’m an atheist who knocks on wood while knowing it’s hogwash. “When an
intuition captures attention and triggers emotions, it may be especially hard to shake,” says Jane
Risen, a psychologist at the University of Chicago. She calls maintaining beliefs we know to be
false “acquiescing to intuition.” Intuition may not be magic, but we are truly under its spell.
To have good intuitions in any domain requires a lot of practice. But not all domains are
amenable to good intuitions. First, there must be regularities linking events and outcomes—the
domain must have high “validity.”
Gary Klein, a psychologist at the Washington, D.C., consulting firm MacroCognition, has long
explored the role of wisdom in the intuition of experts such as fire commanders, who can size up
a burning building quickly. “Fires follow the laws of physics,” says Klein.
The global economy is significantly more chaotic, preventing predictability. (As Gigerenzer
notes, five years before the 2007 housing crisis, the president of the American Economic
Association said, “Macroeconomics…has succeeded. Its central problem of depression
prevention has been solved.”)
Whether you should trust your feelings should hinge not on the strength of those feelings—we
have poor intuitions about intuitions—but on the structure of the domain you’re operating in.
Look outward, not inward.
Second, you need clear feedback to hone your intuitive decisions. A review of the literature
shows that weather forecasters, test pilots, and chess masters had more reliable expertise than
psychologists, admissions officers, and judges. Outcomes in the latter’s areas are fuzzier and can
play out long after you’ve made a decision. That goes for much of everyday life, too.“You don’t do
a diary and an Excel file where you write, ‘Okay, on October 1, I made this decision, or I bought
this product,’ and so on,” Topolinski says. We lack hard data about what we do.
Good intuitions in one domain don’t guarantee good intuitions in another. As Gigerenzer puts it,
“A soccer player who has great intuitions about scoring a goal may have bad intuitions about
spending his money. So there cannot be a general test of intuition.” Even within a domain,
expertise can vary between different kinds of tasks.
We can use focused thinking not only to train our intuitive expertise over time but also to invite
or avoid intuitions in the moment. Metaphors and sketches are excellent tools to help us reframe
problems or see solutions more clearly.
Klein coaches people to consider premortems: When considering a plan, imagine from a future
vantage point that it failed and think about what went wrong. This thinking tool makes weak
points real—intuitive objects rather than abstract and ignorable hypotheses.
Philosopher Daniel Dennett of Tufts University has coined the term intuition pumps for thought
experiments meant to reframe problems. But he notes that they can be used for good or for evil.
“One should learn how easy it is to build bogus intuition pumps that will provoke fist-pounding
intuitions that aren’t worth your allegiance,” Dennett says. “But also, intuition pumps can help
you out of imagination blockades. Caution is advised.”
The role of deliberation in honing instincts and knowing when to trust them reveals reflection’s
close collaboration with intuition, in both its development and deployment. “Our reflective
deliberation scaffolds off our intuition, but it goes both ways,” says psychologist Gordon
Pennycook of the University of Regina in Canada. We also tend to use them in tandem.
Lou Brooks, used with permission
Intuition is closely related to another I word, insight. Sometimes the two are conflated, which is
understandable. Both relate to realizations emerging from subconscious processes, offering
guidance and hiding their tracks. But they’re fundamentally different.
According to MacroCognition’s Gary Klein, “Intuition is how we use our experience to know how
to act. Insight runs in the opposite direction. It’s not just drawing on what you know. It’s
changing what you know.”
To that end, we sometimes need to clear intuition out of the way to obtain the sudden solutions
we call insight. Breakthroughs are often counterintuitive. One way to demonstrate the role of
habitual hindrances is to look at magic tricks. Illusions work through mental jujitsu, using our
assumptions against us. To discover how a trick is done, one must relax certain mental
constraints—a good tactic for eliciting insights in general.
In one study, participants watched video clips of a dozen magic tricks, and half received a verbal
clue directing their attention to an assumption. For example, when the magician appeared to
throw a coin from one hand to another before making it disappear, the clue was “transfer to
other hand.” Given such prods to counter their intuition about what they saw, their solution rate
went from 21 percent to 33 percent.
Intuition’s relationship with insight is complicated. It can sometimes indicate when an insight is
possible. A common laboratory test of insight is the remote associates test (RAT): Given three
words, such as cottage, swiss, and cake, can you find a fourth that connects them? (In this case,
cheese.) A variation on this task shows people either a coherent or a random word triad and
makes them guess quickly whether it’s solvable before asking them for a solution. Even in cases
where people can’t summon the solution, they’re better than chance at judging the triad’s
coherence.
Scientists use creative intuition to select which paths to follow toward potential discoveries.
“This is this idea of sensing the right direction,” Sadler-Smith says, “like a radar that says ‘Go
down there, but not down there.’” Nobel laureates have discussed their use of hunches. Michael
S. Brown (Medicine, 1985), has said, “As we did our work, I think, we almost felt at times that
there was almost a hand guiding us.”
But we tend to have poor intuition about how close we are to an insight. In one study,
participants were given math and logic problems, whose solutions required either a central
insight or mere grinding, and were asked to estimate their distance from the solution every 15
seconds. Unlike for the non-insight problems, estimates for insight problems remained fairly flat
until the final “Aha!”
In a second study, participants’ predictions of whether they’d be able to solve insight problems
had no correlation with the truth, unlike for routine algebra problems. Topolinski notes the age-
old attempt to perform a mathematical feat called “squaring the circle” before it was proved
impossible in 1882. “There were many blind tracks that people followed for millennia,” he says.
Similarly, Einstein produced his theories of relativity, “then for the rest of his life he’s concocting
a possible theory of everything.” Such a theory may be out there, but “for his capabilities and his
time, this was a wrong intuition.”
Deliberation is a luxury. In dire situations—say, while being chased by a bear—you don’t have
time to weigh all your options. You follow your first instinct (run, presumably). Anxiety
engendered in any situation similarly pushes you toward fast and frugal reflexes. If you’re truly
in danger, that can be handy. Otherwise, reflection might be better.
Stress’s effects on the brain are mediated in part by the release of the hormone cortisol. In one
experiment, researchers gave participants a cortisol-increasing drug or a placebo, then had them
do something called the cognitive reflection test (CRT). The CRT consists of three questions,
each with an intuitive but wrong answer. For instance, “A bat and a ball cost $1.10. The bat costs
$1.00 more than the ball. How much does the ball cost?” You want to say 10 cents, but a quick
calculation reveals the ball is 5 cents and the bat is $1.05. Most people, even students at elite
colleges, fail to get all three problems right, but cortisol reduced correct answers even further.
Even as stress triggers heuristic thinking—habits and short-cuts—it degrades more sophisticated
intuitive processing. Remember the remote associates test (cottage, swiss, cake)? One study
found that increasing anxiety in participants by showing them hair-raising images scrambled
their intuitions about whether a connecting word existed.
The morbid images may have affected this performance measure, called an intuition index, in
part by lowering participants’ mood. Sadness tends to make people think analytically. We’re sad
when something’s wrong, which may be time for focused problem-solving.
Another scale, with items like “I generally make decisions that feel right to me,” correlated with
better recognition of social norms, as measured by how accurately people estimated their peers’
acceptance of behaviors likes stealing and fighting. And another correlated with greater
creativity on several tasks such as drawing and thinking of uses for a cardboard box.
But people who put faith in intuition also pay a price. They perform worse on tasks requiring
logic. They report having experienced more setbacks resulting from poor decisions, ranging from
missing a flight to getting divorced. They report greater magical thinking—belief in astrology,
ghosts, luck, God, and so on. And in one study, they were more likely to stereotype based on
gender (but only in a positive mood).
Topolinski suggests that people might want to seek careers that match their thinking style. An
accountant won’t get as far relying on her gut as would, say, a counselor. And in any profession,
if you know you put great faith in feelings, you might make room for extra reflection on tasks
where snap decisions can get you into trouble, like getting to the airport.
Some of our deepest-held beliefs involve morality, how we feel people should behave toward one
another. And although they may seem as rock-solid as fact—Thou shalt not kill—they’re just as
guided by intuition as anything else.
We can reason about many of them, but only to a point. For many, especially on controversial or
subtle issues like abortion, it comes down to intuition: It just feels wrong (or right).
Moral intuitions are unavoidable and also valuable, says psychologist Matthew Feinberg of the
Rotman School of Management in Toronto. They drive kindness as well as social justice
movements. “But moral intuitions are also at the heart of many, many problems in society.”
Impassioned gut reactions can derail rational discussion, as opponents are labeled evil.
Many findings highlight the unconscious processing built into moral judgment. Often we base
opinions on things we’d never factor into a deliberate decision. In one study, participants’
approval of sex between cousins depended on whether someone had secretly deployed fart spray
nearby. Visceral repulsion led to moral repulsion.
In another study, participants were asked whether it was okay to push a large man off a
footbridge to block a trolley from killing five other people. If they’d just watched a clip from
Saturday Night Live, versus a documentary, their mood was more positive, and they were four
times as likely to approve. This does not sound like reflection: Thou shalt not kill—unless you’ve
heard a good joke lately.
Of course, morality is based on more than fleeting incidental cues. We also have deeper values
like fairness and loyalty, each an abstraction formed from a lifetime of experience. Psychologist
Jonathan Haidt of New York University has delineated five distinct “moral foundations” that
guide our behavior: fairness, loyalty, authority, purity, and avoiding harm. Studies suggest that
political liberals prioritize fairness and harm avoidance; conservatives favor loyalty, authority,
and purity.
And Feinberg has found that we can shape people’s moral intuitions by catering messages to
their preferred values. When he framed an argument for universal health care in terms of purity
(fewer diseased Americans) versus fairness (health care for all), conservatives expressed more
support for Obamacare. When he framed an argument for military spending in terms of fairness
(fighting inequality) versus authority (American supremacy), liberals expressed more support.
Examples, metaphors, images, and stories can give shape to our own and others’ intuitions not
only in politics but in all realms of life: science, relationships, education. We gain new models of
the world, and thought—conscious and not—fills them in.
As for messages meant to elicit gut reactions, “We see a lot of that on the Internet these days,”
says the University of Saskatchewan’s Thompson. “Memes. That’s exactly what they are.” One
could fairly call memes the fart spray of the internet.
8. You Can Read People by Reading What They Write Online
Humans have strong intuitions about other people. That’s because character judgment has such
dire consequences, and because we have so much experience with it, over our lifetimes and over
evolution. What happens when people-reading goes online? And when it’s limited to the reading
of what others write? Increasingly, we must assess each other via snippets of text, rather than,
say, darting eyes or kind smiles, but that doesn’t hold back our snap judgments.
Generally, when asked to rate a writer’s personality traits based on emails, personal essays,
streams of consciousness, mock diary entries, mock blog posts, Twitter feeds, and dating ads,
readers agree with each other more than chance would allow, indicating that there are cues in
written reports that reliably trigger our intuitions. Which cues do we use?
In dating profiles, studies show, swear words suggest high neuroticism and low
conscientiousness and agreeableness. Angry words suggest the same in tweets. In personal
essays, exaggeration suggests extraversion and openness to experience. Past tense suggests
depression in blog posts, and cognitive words like know suggest it in diaries. Surely deliberation
plays a role in judgments, but I doubt anyone is counting past-tense verbs.
Our judgments about traits from writing samples are also frequently more accurate than chance
would allow. And some people are better than others at reading between the lines. One study, by
Hall of Northeastern University, found that the best judges were female, agreeable,
conscientious, emotionally stable, compassionate, interested in others’ lives, and big readers,
especially of fiction.
Judgments of personality can come from the thinnest of slices—even just an email address.
What’s thinner than an email address? Punctuation. A study found that angry- and happy-
seeming emails have many exclamation points and few question marks, and feminine-seeming
emails have both. Other work found that smileys in formal emails don’t make the writer seem
warm but do make him or her seem incompetent. Meanwhile, adding a smiley with a nose to a
dating profile will win more replies, while adding a noseless smiley will get you fewer ;-). To
think, your future may depend on an emoji.
Neither head nor heart can survive on its own, and negotiating their symbiosis is a challenge
that’s much harder than mastering chess or Go. “‘It’s not about whether intuition or analysis is
best,’” Sadler-Smith tells managers. “The real skill in decision-making, problem-solving,
creativity, whatever, is blending those two things together. And in a way that’s kind of a lifetime
project, isn’t it?”
Pick up a copy of Psychology Today on newsstands now or subscribe to read the rest of the
latest issue.