By Elizabeth R. Tenney, Jennifer M. Logg, and Don A. Moore

The $10 billion self-help industry offers to inspire us to be optimistic, confident, and therefore successful. We are buying it. Especially in the US, we tend to value optimism. This may be because we think we see evidence of optimism’s power all around us: We see winning athletes attribute their success to confidence; optimistic political candidates attract support and win elections; and in school, more confident students perform better on tests. Optimism often precedes success, and so it may be tempting to put your money on the optimistic horse over the pessimistic one. But that is the thing about predictions and correlations: the cues that predict success (e.g., optimism) might not be the same factors that cause the success. Unless optimism actually causes success, then it would be a mistake to devote money and energy to inspiring optimism rather than, say, developing skills. In a series of studies, we set out to first test whether people believed that optimism caused better performance. First, we asked participants to read short stories in which the protagonists needed motivation to perform or needed to make decisions. For example, in one story, the protagonist was undergoing heart surgery. When the protagonist needed motivation (for rehabilitation purposes), people prescribed optimism: if the true chance of success was 70%, then the protagonist should believe it was 85%. But when the protagonist was deliberating and needed to make a decision (about planning for after the surgery), people endorsed realism. These results revealed that people prescribed optimism selectively: when they thought it could help performance. 

Then we put those beliefs to the test. Did optimism really help performance as much as people expected? We asked one group (we’ll call these people “predictors”) how well test-takers would do. We asked another group to be test-takers. The experimental manipulation led some test-takers to expect to be among the top performers (optimistic), others to expect to be among the worst (pessimistic). We used several different types of tests: math, age-guessing, and even Where’s Waldo puzzles. Predictors believed that test-takers induced to be optimistic would perform better than those induced to be pessimistic. We made it clear to them that, thanks to random assignment, everything else about the test-takers was similar. Predictors were still willing to bet their money on the optimists. 

In fact, the results showed that optimism failed to increase test performance—at least in the tasks we tested. Those induced to be optimistic did search longer for Waldo, but they didn’t actually find him any more often. They also did not do better at math or at guessing ages. So predictors’ bets on the benefits of optimism failed to pay off. Were there some sorts of people whose performance benefited from optimism? Not that we could find in our data. It is possible that we happened to choose tests impervious to optimism. However, we explicitly selected them because we expected them to show a positive effect of optimistic beliefs.

Why do people believe that optimism is more useful than it really is? It may be that we see optimistic people achieving their goals and we interpret optimism as playing a causal role when it is maybe just a byproduct of knowing you have actual skills. Another possibility is that we might believe it is more enjoyable to be optimistic, and this makes us prone to seeing the upside of optimistic thinking even where it does not exist. Whatever its cause, our evidence suggests that we are, at least sometimes, too optimistic about the benefits of optimism.


Elizabeth R. Tenney is an Assistant Professor in the Management Department at the University of Utah Eccles School of Business. She studies overconfidence and other biases that affect people's social interactions and decisions. Her website with links to her publications can be found here

Jennifer Logg is a PhD candidate in Management at the University of California, Berkeley’s Haas School of Business. Her research examines how people make judgments about themselves and the world, why they make biased judgments, and how they can improve their accuracy. 

Don Moore is on the faculty at UC Berkeley's Haas School of Business. He studies overconfidence, including when people think they are better than they are, when they think they are better than others, and when they are too sure they know the truth. He can be reached at [email protected].