According To Study: Do We Actually Think We Won't Change In The Future?

The words 'according to a study' can be used to justify any old garbage, and the average reader has seen enough bullshit 'studies' reported in the media to be suspicious about them all. So this is a new series for TheVine. So far we've looked at whether cute cats actually increase productivity, whether whether bacon actually causes cancer and whether being spiritual makes you more likely to have psychological problems?. This week: are people terribly bad at imagining how much they'll change in the future?


One of the things about being human is that we change. I'm not quite the same person I was a decade ago; I mean, 2003 Tim (who was something of an indie snob who really liked Radiohead a lot) wouldn't have imagined that 2013 Tim would unironically dig a song like Taylor Swift's 'We Will Never Ever Get Back Together'. Beyond all the things I've learnt in that decade, I think I'm more outgoing and confident than I was in 2003, perhaps a little more open to new things. So will I change just as much from now to 2023? It doesn't feel to me like I will. But unless a TARDIS conveniently materialises in my living room tomorrow, it'll be 2023 before I know how much I'll have changed.

The reason I'm telling you all this is that a research paper titled The End Of History Illusion (which you can read here), by Jordi Quoidbach and colleagues, got prominent coverage from the New York Times and Time Magazine recently. The New York Times argued that Quoidbach's study showed that "when we remember our past selves, they seem quite different. We know how much our personalities and tastes have changed over the years. But when we look ahead, somehow we expect ourselves to stay the same." Time Magazine said that "even though most people acknowledge that their lives have changed over the past decade, they don’t believe change is constant. Against all evidence, most people seem to believe that who they are now is pretty much who they will be forever." This idea that we think we won't change in the future but actually we change a lot - the End of History Illusion of the paper's title - sounds like another one of those classic intuitively counter-intuitive findings from social psychology. It's a neat life lesson: "don't be so sure of the future". Malcolm Gladwell's probably already writing a book about it.

However, the way it was reported in the papers is misleading. At best. This is not actually the New York Times' fault; it's because the research was published in an academic journal called Science. Science is basically a rubbish place to find out about solid psychology research, because it's a) highly prestigious and very good for a scientist's career to get published there; and b) because it likes to publish very short articles. Because any psychologist with an eye on climbing the academic ranks would dearly love to be published in Science, they're more likely to spin their findings to make the editors at Science think "hey, that's awesome!". And because Science likes to publish very short articles, the gory details of the study are often missing. And those gory details definitely missing in the case of 'The End Of History Illusion', which doesn't even report the raw averages, or the questions participants were actually asked. The article by the New York Times' science reporter, John Tierney, is a reasonably faithful summary of the spin the authors of the article put on it. It's also incredibly misleading, because the spin is misleading. Tierney likely didn't have access to the gory details (which you can apparently only access from an online repository of data if you have a university IP address) or didn't understand them, because the gory details actually completely upend the message you get from his article. This is what Tierney said: "when we remember our past selves, they seem quite different. We know how much our personalities and tastes have changed over the years. But when we look ahead, somehow we expect ourselves to stay the same". This, on the other hand, is a more accurate summary of the research, if you look at the gory details: "when we remember our past selves, they seem slightly different. We know how much our personalities and tastes have changed over the years. But when we look ahead, we expect ourselves to also be slightly different (though we very slightly underestimate how different)."

So what's the research basically about? In one study, the researchers got French adults from the age of 18 to 68 to fill in a personality survey, which asked questions about whether they saw themselves as conscientious, neurotic, agreeable, etc. They then either asked the participants to fill in a survey about their personality 10 years ago, or to fill in a survey on what they thought their personality would be like in ten years. The clever thing the researchers did was to take, say, the difference between average 18 year old's assessment of their personality now versus in ten years time, and compare it to the difference between the average 28 year old's assessment of their personality now versus ten years ago. This way, the researchers could see whether people's predictions of how much they'd change in a decade resembled people's memories of how much they changed in a decade. Doing the research, it seemed to the researchers that people's assessments of how much their personalities had changed were reasonably accurate, because the researchers compared that data to previous research where people actually took the same personality test 10 years apart - the amount of change in people's memories was broadly comparable to the actual change in personality test scores over 10 years. However, compared to this data, people seemed to underestimate how much they'd change in the next decade. In further studies, the researchers seemed to find a similar effect when, instead of looking at personality, they looked at how much people's core values change (how important are things like success, money, sex, etc), or looked at simple facts about their strongest preferences (their favourite band or their best friend, for example). Put it this way: 38-year-olds thought their core values had changed more compared to when they were 28 than 28-year-olds thought their core values would change in the next 10 years. 46-year-olds were more likely to think their best friend would still be their best friend when they were 56 compared to how many 56 years olds said they had the same best friend when they were 46. And so forth!

Unfortunately, because Quoidbach and colleagues published their research in sexy prestigious Science, rather than a normal boring psychology journal, it's surprisingly difficult to figure out what was actually going on in their experiments; having enough information to figure out what's going on is exactly why reports of traditional psychology journal articles are invariably long and boring. While I seem to be unable to access the detailed data in the data repository (despite having the university IP address that's meant to give you access), an awesome person going by the name 'deinst' who commented at another blog criticising this paper could view it and discussed it in a comment. In the first study, which looked at personality changes, participants were asked how much they agreed with statements like "I see myself as extroverted, enthusiastic" on a scale of 1 to 10. 'deinst' reports that the average amount of change between current personality and 10-years-ago personality is 1.45. 1.45 isn't very much change at all on a scale of 1 to 10 - it's something like the difference between "I agree very strongly that I am extroverted" and "I agree quite strongly that I am extroverted". Importantly, the average amount of predicted change between current personality and 10-years-in-the-future-personality is 1.2, according to 'deinst'. On a scale of 1 to 10, the difference between a change of 1.45 (remembered change) and a change of 1.2 (predicted change) is basically bugger all. 

Remember that the researchers called this very small change The End Of History Illusion (probably inspired by Fukuyama's in-hindsight-amusing 1989 argument that history was over because communism was vanquished). 'People, it seems, regard the present as a watershed moment at which they have finally become the person they'll be for the rest of their lives', they (misleadingly, judging by their data) say in the abstract of the paper. But if deinst's stats are accurate, people do think their personalities will change, and are actually pretty damn accurate at knowing about how much their personalities will change (I'd be happy to be off by .25 out of 10, really). What the research did find was that people very slightly (and consistently, if the statistics are accurate) underestimate how much their personalities would change. But not by very much! So in my imaginary world where science articles are always reported accurately, the headline would be "PEOPLE REASONABLY ACCURATE AT PREDICTING HOW MUCH THEY'LL CHANGE IN A DECADE, BUT THEY UNDERESTIMATE SLIGHTLY". Sadly, the idea of the "End Of History Illusion" - that we mistakenly think we'll never change ever again - is much sexier than "people slightly underestimate how much they'll change". It made a big splash, which is what Science wants. But it's really no good for science, the discipline, if research where the journal and researchers make it as hard as possible for readers to figure out the real story not only gets published but also gets heaps of publicity.

The other thing that bugs me about this research, beyond the spin around it being so exaggerated, is that it assumes everyone's the same. Have you noticed that I've used the words "we" or "people" a lot? The researchers consistently use these words too; they talk over and over again about what "people" do in their study. But, of course, people aren't all the same. People who speak different languages and who experience different cultures may respond differently to questions. In particular, the participants in Quoidbach and colleagues' study were people who visited the website of a French TV show, Leurs Secrets Du Bonheurs, which as far as I can tell, was a documentary that followed the lives of six French people who were trying to follow scientific methods to improve their overall levels of happiness. It may be that there are differences between people who go to the websites of French TV documentaries and the rest of the world; perhaps worrying about whether you're happy is related to feeling like you won't change very much. Maybe - sacre bleu! - the questions that they asked participants translated poorly to French. But the other thing here is that - even if we assume that French people who want to be happier are basically like us - research like this study will take everybody's results and add them all together, not worrying about the differences between people. Maybe smart people (like you, dear reader!) are less likely to underestimate how much they'll change. Maybe anxious people are more likely to underestimate how much they'll change. Remember that the researchers found a small but consistent underestimation of how much we'll change. Maybe there's a small amount of (somewhat idiotic) people who assume they'll never change ever again, and the researchers found the effect entirely because of how those people skewed the data. All of which is me making stuff up. But this is research that could be fairly easily done! Quoidbach et al. could spend a week looking at their data again to analyse the relationship between personality attributes and extent of personality change and see whether any of this matters. I'd read a paper about that. I mean, as long as they publish it in a proper psychology journal with all the gory details.

In some ways, perhaps the take-home message here is that your personality doesn't actually change all that much in a decade, on average. The difference between being very strongly extroverted and quite strongly extroverted, for example, isn't that much really. But it's a shame, because the End of History Illusion is a nice idea, isn't it? In the end, I think the reason people were drawn to the New York Times article was that they liked to think they could change, that they weren't stuck in a rut. Even though you can't imagine the future being all that different right now doesn't mean that it can't be - won't be - different.

Of course, this is still true whether or not the 'End of History Illusion' is real. You can still change. You're not stuck in a rut. Just because the average person's personality doesn't change much doesn't mean that your personality won't change. Go out and seize the day and so forth!


Tim Byron, Lead image via Shutterstock.

P.S. Further critiques of this study (on logical/mathematical grounds) are available here and here.

profile of TimByron