What if we could reliably predict how long you will live for based on how much you smile in photos? First of all, we would see a lot more happy faces in photos of awkward extended family get-togethers. Second, we would have a piece in the puzzle of how we can increase longevity.

In 2010, psychologists Michael L. Kruger and Ernest L. Abel of Wayne State University wanted to see whether we could, indeed, predict lifespan based on smile intensity. To answer this question, they looked at how intensely Major League Baseball (MLB) players smiled in their league photos.

Their results were astonishing—for the 230 players they analysed, they found that players who showed full smiles in their pictures were half as likely to pass away in any given year than those who did not smile at all. Okay, so done deal, right? Let’s start smiling for some extra mileage!

Not so fast. A team of scientists, including psychologists Michael Dufner of the
University of Leipzig and the University of Toronto’s very own Dr. Joanne Mee Hae Chung, reminds us not to be so quick to jump to conclusions. They replicated and extended this study with a much larger MLB sample of 14,101 players and with additional coding resources (like
computer programs that automatically assessed how intense a smile is in a picture). They found that how intensely someone smiles did not significantly predict how long a given MLB player would live for.

While this may seem like a forgone conclusion for some, we need researchers to carry out replication and extension studies like this to keep scientific findings in check. Without them, it can become easy for people to accept and spread conclusions that are not robust, or, put simply, real.

“Replication, strictly, is just about repeating a thing—repeating an experiment, repeating a correlational study exactly how they had done it before,” says Dr. Chung. “[Modifying a study] isn’t actually part of replication—we would call that an extension.”

Dr. Chung describes replication and extension studies as cornerstones of psychological research. “We care about being an incremental science. We want to build on other people’s findings and come up with theories about human behaviour. So one of the ways to do that is to repeat an experiment or a study and see if we get similar results because that will give us more
evidence that the thing is ‘real.’—[that an] effect is robust.”

Differences in results between original and replication/extension studies may be due to elements of chance. “Replication is interesting in different contexts, because you can see the extent to which things hold up. Like you [might] find an effect in certain groups of people, but you may not find it in others,” explains Dr. Chung.

However, contradictions in studies might also be due to pressure on professors to discover new things. Dr. Chung notes that this stress leads to questionable research practices. “Questionable research practices are ways in which you massage your data to make it conform to the conclusion that you want by hypothesizing after the results are known, dismissing results that
conflict with previous research, and continuing to collect data after almost reaching statistical significance, to name a few. People are like, ‘Well, why do people do that?’ and it’s because [. . .] there are a lot of incentives to have publications.”

It’s practices like this that make replications and extensions so crucial to psychology. “Replication is important because we want to have a strong science that’s built on robust findings.”

All of this in sum leads to understandable scepticism around scientific
findings—especially those that sound a little far-fetched. We know now that there is no evidence that smiling will lead to a longer life, but can consistently being happy help?

There is no definite answer on the matter just yet. A 2010 study by Dr. Ed Diener and Dr. Micaela Chan, which reviewed multiple other studies over two decades, came to the conclusion that those who reported higher emotional well-being tended to live longer and be healthier. Conversely, in 2016, Dr. Bette Liu and her team found that self-reports of happiness in a sample of 700,000 women made no difference in death rates. Another 2018 paper by professors Bruce Headey and Jongsay Yong suggested that while unhappiness is linked to a shorter life, how happy you are otherwise has no relationship with your longevity.

In science, it is very, very hard to come to ultimate answers. It is a space full of disagreement, dialogue, and, most importantly, improvement. A scientist should not unconditionally protect a certain theory or study: they should seek to challenge these things.

And so should you. You and everyone else around you deserve to know the truth, and that demands you look past face (get it?) value.