Why You’ll Always Think Your Big Changes Are Behind You

Think about your favorite singer or band. What would you pay me today for a ticket to one of their concerts ten years from now? OK, now think about a different group—the one that was your favorite ten years ago. (If your answer to both questions is the same act, just take your Dylan bootlegs and try another quiz, OK?) What would you pay to go hear this year’s favorites, this week?

If you’re like the respondents in this study in the current issue of Science, you’ll be willing to pay more to see your current favorite in ten years than you are to see your old love right now. To be precise, 170 people were willing to pay an average of $129 to see today’s favorite in 10 years, but only $80 to see their once-favorite band from a decade ago. That 61 percent difference is, in essence, a strange bet—knowing that they value the 10-year-old band less, they still think they will value the current band more, when it is ten years in the past. This hunch—”I’ve changed a lot to get here, but now I’m done”—is, as the rest of the paper describes, a pretty strong and pervasive prejudice among the 19,000 people the researchers tested. The authors—Jordi Quoidbach, Daniel T. Gilbert and Timothy D. Wilson—have come up with a lovely name for it: “The end-of-history effect.”

The End of History, Francis Fukuyama’s 1992 book, argued that Western-style liberal democracy might well represent the end-point of humanity’s forward march, as it could prove the final, lasting form of government on which all nations would settle. (In the early 1990s, with the Soviet Union recently collapsed, a recession that seems like a mild headcold compared to now and Islamic terrorism faint on the radar, this kind of thing seemed plausible).

Twenty years on it looks foolish indeed to assume that social change would not continue into the future, as it always has in the past. But, as the Science paper nicely documents, “history is over” does seem to be the default setting for understanding one’s personal life.

Via interactive Web surveys, Quoidbach et al. asked people (the vast majority women—do men not take Internet surveys?) to play through different versions of the 10-years-ago versus 10-years-hence procedure. They asked some respondents to estimate how their personalities had changed and would change; they asked others about basic values (pleasure, success, security); they asked others about preferences for music, food, vacation and friends. Consistently, they report, they saw the same effect. People reported significant change in the past, but expected little to no significant change in the future.

You can think of a lot of reasons why most people would like to preserve the illusion that they’re coherent beings. Psychologically, it’s hard to feel sure of yourself if you know your self could be a different person in a few year’s time. Then, too, there are the practical and psychic pressures of a society whose institutions depend on consistency over time. My mortgage binds me to do, think and feel in 20 years just as I did when I signed it. So does a marriage vow, or any solemn promise. When voters keenly scrutinize political candidates’ biographies for clues about how they will behave in office, it’s with the presumption that people stay roughly the same throughout their lives.

Why, then, can we face up to our inconsistencies in the past but not expect more in the future? I suspect part of the answer lies with the human mind’s great talent for spinning stories, especially about itself. No matter how much your past is full of zigs and zags as you changed in personality, values and tastes, you can always package it as a coherent narrative. I can describe what happened to me as a natural evolution (my principles drove me to rethink my stance on marriage equality) or I can describe the change as a dramatic break (my experience in the war really changed me). It doesn’t matter which; when I tell the tale, it will have the reassuring coherence of story. In other words, all autobiographies are coherent—not because people are, but because stories are. The future, being unknown, cannot be shaped in this way. Quite the opposite: Sitting on a story of ourselves that makes sense of the past, we can’t be open to the idea of future change. It would mess up the narrative. (Of course, once the change has occurred, we—artful storytellers that we are—will find a way to work it in, or to leave it out of the tale.)

According to Quoidbach et al., one possible explanation for their results, is self-flattery: “most people believe that their personalities are attractive, their values admirable, and their preferences wise,” they write, “and having reached that exalted state, they may be reluctant to entertain the possibility of change.” But I think there’s more to their second suggestion: “People also like to believe that they know themselves well, and the possibility of future change may threaten that belief.”

POSTSCRIPT 1/8/13: Jordan Ellenberg, a mathematician, is skeptical about most of these experiments. (Hat tip Gary Marcus for this.)

Ellenberg notes that the experimenters asked people to make specific predictions about various traits in the future, and then combined these into an estimate of overall future change. That’s a mathematical mistake, he says. Knowing that some general change will occur doesn’t mean you know where and how it will. This is because even if people expect to change in someway they can’t be sure exactly in which way it will happen. So imagine a person who is asked “do you expect to change a lot in the next ten years?” and then is asked about food, music, hobbies, friends and vacations. If that person says “yes!” to Question 1 but then says “no” to all five of the specifics, that person is doing their math correctly, and displaying no cognitive bias at all, Ellenberg argues. From their inability to predict five specific changes, you cannot infer a general refusal to predict change of some kind.

Only one of the studies in the paper, asked people to compare apples to apples: The one in which volunteers were asked to compare past personality changes to their estimated future change. Here’s hoping the authors (and other interested parties, since the data is freely available here) will weigh in.

Follow me on Twitter: @davidberreby

Quoidbach, J., Gilbert, D., & Wilson, T. (2013). The End of History Illusion Science, 339 (6115), 96-98 DOI: 10.1126/science.1229294