The Zoom call came in on a spring day in 2021, shortly after Matthew Killingsworth published a paper debunking one of the most widely accepted ideas in the field of behavioural economics: that making more money, above about US$75,000 a year, did not, in fact, make people any happier.
Dr. Killingsworth, a psychologist and senior fellow at the University of Pennsylvania’s Wharton business school, sat on his paper for months, hesitant to go public. Not only was his study likely to stir up a contentious debate, it was also challenging the famous research of a distinguished scientist he’d revered since he was a student.
But Dr. Killingsworth had checked his data over and over again to make sure. The original finding, according to his math, was wrong. Money did buy more happiness, on average, all the way up the income ladder.
On the other end of the call that day was Daniel Kahneman, the celebrated professor emeritus at Princeton University and best-selling author of Thinking, Fast and Slow. Dr. Kahneman, whose work on decision-making earned him a Nobel prize in economics – and later, a Presidential Medal of Freedom – had co-authored the money-doesn’t-equal-happiness paper in 2010, along with a colleague, Angus Deaton, who would also go on to win a Nobel prize.
That first conversation, and the couple that followed, were tense. Dr. Killingsworth was understandably nervous; Dr. Kahneman understandably disgruntled. But rather than simply protest the finding, Dr. Kahneman eventually made a proposal: could the two adversaries collaborate and discover the truth together? Dr. Killingsworth thought it over, and said yes.
Imagine that! Two people with opposing views, armed with their own carefully-researched facts, agreeing to respectfully hash it out.
It hardly seems possible these days, with all the raging in the streets, and insult-flinging on the Internet. How easily will our turkey go down this Thanksgiving, sitting at the table with whichever relative arrives spouting views we find objectionable? Social media’s churlish ways have trained us well: better to hunker down with our own opinions than try for a negotiated truce.
After all, a good-faith effort to find agreement means both sides have to remain open to the possibility that they might be wrong. And who ever likes to be wrong?
Yet if anyone was going to reach out to a critic, it was Dr. Kahneman, who died earlier this year, a few weeks before his 90th birthday. He’d spent decades publishing groundbreaking work on how and why people make irrational judgments and defend flawed choices. And he’d long called out those same blind spots in his scientific peers, criticizing the “absurdly competitive” and hostile nature of a contentious back-and-forth, where nobody ever admitted error, or even that they’d learned something from the other side.
“Controversy,” he declared in a 2003 essay in the journal American Psychologist, “was a waste of time.” What’s more, he wrote, “doing angry science is a demeaning experience – I have always felt diminished by the sense of losing my objectivity when in point-scoring mode.” A more fruitful method, he proposed, would be to research the topic together – an approach he called “adversarial collaboration.”
Neuroscientists, who have long studied the brain to see where consciousness lies, recently turned to adversarial collaboration to get better answers.
Over the last decade, adversarial collaborations have become more common in psychological science. In 2023, for example, researchers published the results of a five-year study pitting two camps against each other to test the leading theories about where consciousness originates in the brain. (In case you are wondering, in very simple terms, the executive-functioning, prefrontal-cortex team conceded points, but not the argument, to the side contending for neural networks in the back of the brain. With no winner declared, a new study is pending.)
As part of another long-term adversarial collaboration, a small group of researchers in the United States and Europe have been releasing their ongoing findings on working memory and aging.
And in 2021, two American psychologists with differing viewpoints published the results of their joint effort to understand whether liberals or conservatives were more likely to be biased against people with different views. (The conclusions: contrary to what either author expected, the liberals tended to be more prejudiced than conservatives against people based on political differences, such as immigration and abortion, although not by much. Conservatives took top-billing when it came to non-political differences, such as fears and preferences, although only sometimes.)
Playing nice with your antagonists isn’t easy, in life, as in science. Bringing adversaries together often produces messy results, and inevitably, leads to more debate.
But when these collaborations work, researchers say, they can advance science in deeper, and more consequential, ways. And what makes them work – a willingness to listen, clear boundaries, and a shared goal – sounds like a worthy experiment for all of us.
How can we ever rescue the fact from fiction, and break society’s bickering, conspiracy-addled habits if we don’t even try?
Smiling, and encouraging others to smile, are not uniquely human practices, as these chimpanzees demonstrate. When naturalist Charles Darwin – who was no stranger to adversarial science in his own day – pointed this out, it would set off decades of debate about facial expressions and behaviour.Colin Braley/Reuters; Punch Magazine
In the study of happiness, it only took a smile to launch a never-ending argument.
Charles Darwin gets credit for starting it all in 1872, by proposing in his book The Expression of the Emotions in Man and Animals that making a facial expression could alter emotions and behaviour – an idea called the “facial feedback hypothesis,” which psychologists have been debating ever since.
In 1988, German psychologist Fritz Strack, together with two colleagues, decided to wade in. They published a quirky experiment designed to investigate whether simply activating the muscles that make a smile could make people feel happier.
In Dr. Strack’s study, the group recruited 92 undergraduate students at the University of Illinois and gave them a pen. They were told to perform certain written tasks while they either held the pen between their lips, preventing a smile, or between their teeth, to force one. (You, too, can try this at home.)
The subjects, who thought they were participating in a study about something else entirely, were then shown Far Side cartoons and asked to score how funny they thought they were. The people with the pen between their teeth, the authors reported, were more amused by the cartoons. The study, published in the Journal of Personality and Social Psychology, concluded that simply activating those smile muscles did appear to increase happiness.
Dr. Strack went on to study other interests, but the “pen-in-the-mouth experiment” was picked up in textbooks, adapted by other researchers, and seeped into mainstream conversations. Even Dr. Kahneman referenced the study in his book. But then, in 2016, a co-ordinated effort by 17 labs in eight countries ran a similar experiment with 2,000 people, one which failed to find that people found jokes funnier when forced to make a smile.
The emotional effects of smiling with a pen in your mouth, as former French prime minister Gabriel Attal is doing here, have been hotly disputed in psychology.Xose Bouzas/Hans Lucas via Reuters
By failing a replication test, the smile study found itself in already crowded company. Around that time, psychological experiments going back decades were being conducted again by new researchers, one by one, and not reproducing the original results.
It now appears that listening to Mozart probably doesn’t make you smarter, posing like a superhero does not make you behave more confidently, and success in adulthood is not linked to your ability to resist eating a marshmallow when you’re a kid. In a 2015 paper published in Science, researchers tried to replicate 100 studies from top-tier journals; less than half clearly validated the first findings.
A closer look at this trend raised significant questions about how research was being conducted – a problem, to be fair, also showing up in areas of science outside of psychology.
Researchers were accused of manipulating data or revising measurements until they found a result they could publish. Other cases were innocent mistakes, discovered when other scientists took a closer look.
To correct these issues, more authors are now preregistering the terms of their experiments in advance – and committing to publish them no matter what – as well as releasing their data to freely allow independent second opinions.
The replication process itself has been controversial, with some researchers pointing out that original studies are rarely repeated perfectly, and that some older findings that were accurate in their time may have evolved as society – and research subjects – changed. This past September, in a rare moment of scientific irony, a 2023 study published in Nature Human Behaviour that tested methods to make psychological research more rigorous had to be retracted by the authors for not following their own advice.
In studies of the brain and behaviour, researchers exploring the same topic can come to very different conclusions depending on their methods, test subjects and starting assumptions.John Lehmann/The Globe and Mail
As a young researcher in the field of behavioural science, Cory Clark, now a visiting scholar at the University of Pennsylvania, observed that back-and-forth rebuttals often never really seemed to advance the science. Like Dr. Kahneman, she became disillusioned watching her colleagues argue over their published contradictions, yet rarely work together to resolve them.
“A lot of these scholars who disagree with one another, they become these enemies,” she says. “They never talk to each other. They never work with each other. They hide from each other at conferences.”
She might as well be describing the world outside science as well. Social media overflows with advice on how to cut people with whom you have conflict out of your life – less often does it seem to deliver on how to compromise with them. A 2019 survey of 2,000 Canadians suggested we think as little of our political opposites as we do the politicians leading their party; according to University of Toronto professor Eric Merkley, who posed the question, about two-thirds of respondents described the other side (in this case, Liberal and NDP voters describing Conservatives, and vice versa) as selfish or hypocritical – leader and supporters alike. Not exactly the foundation for civil discourse.
Scientists are “mere mortals” like the rest of us, as was noted in a 2022 paper published in the Journal of Applied Research in Memory and Cognition, and co-authored by Dr. Clark. They can be beguiled by status and misled by confirmation bias that amplifies facts that support their opinions over ones that don’t. And this happens even though most of what we understand about why people hold to wrong beliefs, and close their ears to reason, actually comes from the diligent work of those same behavioural scientists. It’s hard for any human to shake their nature.
But who better than scientists to find a procedural way to harness our biases and disagreements? In 2020, Dr. Clark co-founded the Adversarial Collaboration Project at Pennsylvania University, to fund and organize more work between rivals.
A successful collaboration, says Dr. Clark, who is now the project’s executive director, leads to results that are more complex, and, she argues, closer to the truth. They rarely reveal that someone is all right, and another person is all wrong. “When you get everyone together, all that nuance bubbles to the surface.”
Often the “adversaries” realize that they don’t disagree as much as they thought. That might be true in the larger world as well: while the ideological gap between the right and left is widening, Canadians and Americans believe their country is more politically polarized than opinion surveys actually suggest.
“When you have a disagreement with somebody about what is true about the world,” Dr. Clark says, “there are often reasons to believe both. We don’t have to be bitter enemies.”
Fritz Strack accepted 2019's Ig Nobel Prize for his study of smiling, whose findings could not be replicated by 17 other labs.Brian Snyder/Reuters
For his part, Dr. Strack, now a professor emeritus at the University of Wurzburg, says he was comfortable with the controversy about his old smile study. (He did note, however, that he believes the 2016 study added one significant element that changed the results: they filmed the participants with a visible camera.)
“I’m always open for new data,” he says, though not because those new results settle the answer once and for all. “They help us understand the phenomenon hopefully on a deeper level.”
So a few years back, when he was contacted by Nicholas Coles, a PhD student at the University of Tennessee who wanted take another run at his experiment, and invite other researchers to join as well, Dr. Strack agreed to offer advice and add his name to the final paper.
Dr. Coles, now an assistant professor at the University of Florida, put out the invitation for an adversarial collaboration, expecting maybe a couple responses. In the end, the Many Smiles Collaboration involved 26 labs in 19 countries, including researchers who disagreed, supported, or were neutral, on the facial feedback hypothesis. (Dr. Coles himself was a skeptic.)
Over e-mail, they hammered out how to conduct the experiment and measure the data, although Dr. Coles – advised by a smaller group of key authors – had to make some final calls. The study ultimately included 3,878 people. And unlike many psychological experiments, which are typically based on Western, usually white, university students, the results included cross-cultural data conducted in 13 languages.
Nicholas Coles worked with Dr. Stack to devise a broad international test of the findings from the original smile study.Christine Baker
A smile coach teaches pupils in Japan, one of the countries surveyed by Dr. Coles and the Many Smiles Collaboration project.Kim Kyung-hoon/Reuters
Two years later, the Many Smiles Collaboration was published in the journal Nature. The paper concluded that mimicking a smile, or making one naturally, appeared to both increase and create happiness, although the effects were relatively small – about the same as looking at a “mildly positive photo.” For some people, smiling intentionally might help ease a frustrating traffic jam or a stressful encounter. But not enough, according to most, though not all, of the authors, to support facial feedback as a serious happiness intervention.
When it came to the study’s version of the pen-in-the-mouth experiment, using subjects who didn’t know they were being forced to smile, the happiness scores were even lower – too low, the authors wrote, to provide unequivocal evidence for the original finding. (Those results did not dissuade Dr. Strack, who still vouches for it, but adds, “I like the critical discussion.”)
To conduct an effective adversarial collaboration, scientists who have been involved in them offer specific advice: Consider the temperament of your adversary. (Let’s face it, in some cases the mood is too malicious, the views too poisonous, to even begin.) Choose a trusted third party to moderate the discussion and keep it respectful. Clearly define where the disagreement lies. Learn the position of your adversary well enough to be able to explain it in a way they agree with. Divide the outstanding issues into small steps. Stay flexible.
Gordon Pennycook, a Canadian psychologist at Cornell University who has participated in an adversarial collaboration, says that there is so much research being published today, “it’s just really easy to play in your own sandbox.” Working with rivals not only leads to better findings, he says, but ideally, “a stronger guarantee that what you’re doing is going to matter.”
Dr. Clark believes that adversarial collaborations could be used to make progress on many unresolved questions in psychological science – ones that, for example, address racial bias, social media’s affects on young people, whether grit can be taught and how much personality changes. As her 2022 paper puts it, “Bad ideas will die faster, and good ideas will elevate with greater clarity.”
Those first Zoom calls between Dr. Killingsworth and Dr. Kahneman back in 2021 were the beginning of many conversations. Early on, Dr. Killingsworth prepared a 100-slide PowerPoint to explain his finding. Dr. Kahneman picked apart the results. “He wasn’t thrilled,” Dr. Killingsworth admits, to be rehashing a result that had been repeated in countless news stories and academic papers going back decades.
But he could tell Dr. Kahneman was also determined to get to the truth. Dr. Killingsworth would send an e-mail at midnight, and wake up to four missives in reply. This would happen sometimes multiple times a week as they combed through the data for nearly two years. “He had been this mythological figure in my mind as a student and then as a professional scientist,” Dr. Killingsworth says. Now, he knew why. “There’s an intensity to the way he engaged that’s different from a lot of people. And it’s fun and challenging.”
A mutual friend and fellow scientist Barbara Mellers, who had originally connected the pair, agreed to mediate the collaboration, participating in the research and pushing the process forward when it stalled. Dr. Mellers, a cognitive psychologist at the University of Pennsylvania who is researching how to use artificial intelligence to support adversarial collaborations, had served the same role decades earlier in one of Dr. Kahneman’s first adversarial collaborations.
That time, the process became so heated, she says, the two scientists stopped speaking to each other for a while, and would communicate only through her. On the money-happiness question, however, other than a few tense moments, the experience was “calm and collected.”
Usually, once you’re working together, she says, “it’s much more difficult to throw insults across the room at the enemy.” But, “somebody needs to keep things moving.” Even if they take more time, she sees these collaborations as a way “to speed up the science.”
Barbara Mellers from the University of Pennsylvania originally put Dr. Killingsworth and Dr. Kahneman, and agreed to mediate a test of their conclusions.Supplied
In March, 2023, the three of them published the results under the title “Income and Emotional Well-being: A Conflict Resolved” in the The Proceedings of the National Academy of Sciences.
It turned out that Dr. Killingsworth was mostly right: the larger one’s income, generally speaking, the higher the happiness. But the collaboration also found what had thrown off Dr. Kahneman’s results. The boost in happiness generally flattened, as he’d proposed, at a certain threshold for the group of people who were the least happy. His data had, however, failed to capture the continuing connection between larger incomes and rising happiness in the rest. And in fact, for the happiest group in the data, more money made them even happier than everyone else.
Dr. Kahneman may have been mistaken about money. But, over the course of his career, he proved he’d been right about another possible route to happiness – having the ability to sort things out with people who hold an opposite belief.
More generally, Dr. Mellers says, “We need to spend more time taking each other’s perspective, and slow down and think a little more deeply about what the issue really is that we disagree on.”
As the science says: listen, ask questions, keep an open mind. Who knows? This Thanksgiving, you might learn something new about that “adversarial” relative, as well as yourself. Don’t forget to smile when you try it.
Who is the happiest person you know?
Some people seem to approach life with a contagious measure of cheerfulness. They respond to adversity with resilience. And they appear to consistently experience joy in small, but important ways.
As part of a year-long investigation of happiness, The Globe and Mail's Erin Anderssen is embarking on a project to interview the "happiest people in Canada," nominated by the people who know them, through work, friendship and family.
To add a name to this list, please send a short paragraph about them and why you consider them the happiest person you know in the box below, or email it to eanderssen@globeandmail.com.
The happy beat: More from The Globe and Mail
Illustration by Illustration by Pete Ryan
Can happiness be taught? Academics who hope to answer that have found reasons to smile
Urban backyards are disappearing. Can we be happy without them?
Five daily practices to keep your mind sharp, according to neuroscientists
YouTube star Adriene Mishler on the future of yoga and the healing power of pets
