Think Again: The Power of Knowing What You Don't Know

by Adam Grant

  * I think the absurdity was best captured by humorist Richard Brautigan: “Expressing a human need, I always wanted to write a book that ended with the word Mayonnaise.” He wrote that line in the penultimate chapter of a book, and delightfully went on to end the book with the word—but deliberately misspelled it “mayonaise” to deprive the reader of closure. Human need, unfulfilled.

  * Had thought earlier about showing my edits throughout the book, but didn’t want to inflict that on you. Slogging through half-baked ideas and falsified hypotheses wouldn’t be the best use of your time. Even if you’re a huge fan of Hamilton, you probably wouldn’t love the first draft—it’s much more exciting to engage with the product of rethinking than the process.

  * Too whimsical. Early readers want more gravitas here—several have reported that they’re handling dissent differently now. When they confront information that challenges their opinions, instead of rejecting it or begrudgingly engaging with it, they’re taking it as an opportunity to learn something new: “Maybe I should rethink that!”

  * Challenge network says updating a “fun fact” from the book is too trivial.

  * A big unanswered question here is when rethinking should end—where should we draw the line? I think the answer is different for every person in every situation, but my sense is that most of us are operating too far to the left of the curve. The most relevant data I’ve seen were in chapter 3 on superforecasters: they updated their predictions an average of four times per question instead of twice per question. This suggests that it doesn’t take much rethinking to benefit from it, and the downsides are minimal. Rethinking doesn’t always have to change our minds. Like students rethinking their answers on tests, even if we decide not to pivot on a belief or a decision, we still come away knowing we’ve reflected more thoughtfully.

  * For my part, I had assumed the phrase “blowing smoke up your arse” came from people gifting cigars to someone they wanted to impress, so you can imagine how intrigued I was when my wife told me its real origin: In the 1700s, it was common practice to revive drowning victims with tobacco enemas, literally blowing smoke up their behinds. Only later did they learn that it was toxic to the cardiac system.

  * I started not with answers but with questions about rethinking. Then I went looking for the best evidence available from randomized, controlled experiments and systematic field studies. Where the evidence didn’t exist, I launched my own research projects. Only when I had reached a data-driven insight did I search for stories to illustrate and illuminate the studies. In an ideal world, every insight would come from a meta-analysis—a study of studies, where researchers cumulate the patterns across a whole body of evidence, adjusting for the quality of each data point. Where those aren’t available, I’ve highlighted studies that I find rigorous, representative, or thought provoking. Sometimes I’ll include details on the methods—not only so you can understand how the researchers formed their conclusions, but to offer a window into how scientists think. In many places, I’ll summarize the results without going into depth on the studies themselves, under the assumption that you’re reading to rethink like a scientist—not to become one. That said, if you felt a jolt of excitement at the mention of a meta-analysis, it might be time to (re)consider a career in social science.

  * This looks like good news for countries like the United States, where self-assessments came fairly close to reality, but that doesn’t hold across domains. In a recent study, English-speaking teenagers around the world were asked to rate their knowledge in sixteen different areas of math. Three of the subjects listed were entirely fake—declarative fractions, proper numbers, and subjunctive scaling—which made it possible to track who would claim knowledge about fictional topics. On average, the worst offenders were North American, male, and wealthy.

  * My favorite example comes from Nina Strohminger, who once lamented: “My dad called this morning to tell me about the Dunning-Kruger effect, not realizing that his daughter with a Ph.D. in psychology would certainly know the Dunning-Kruger effect, thereby giving a tidy demonstration of the Dunning-Kruger effect.”

  * There’s an ongoing debate about the role of statistical measurement issues in the Dunning-Kruger effect, but the controversy is mostly around how strong the effect is and when it occurs—not whether it’s real. Interestingly, even when people are motivated to accurately judge their knowledge, the least knowledgeable often struggle the most. After people take a logical reasoning test, when they’re offered a $100 bill if they can correctly (and, therefore, humbly) guess how many questions they got right, they still end up being overconfident. On a twenty-question test, they think they got an average of 1.42 more questions right than they actually did—and the worst performers are the most overconfident.

  * That reaction can vary based on gender. In Basima’s study of investment professionals, impostor thoughts helped the task performance of both men and women, but were more likely to spur extra teamwork among men. Men were driven to compensate for their fear that they might fall short of expectations in their core tasks by doing extra collaborative work. Women were more dependent on confidence and more likely to feel debilitated by doubts.

  * I was studying the factors that explain why some writers and editors performed better than others at a travel guide company where I was working. Performance wasn’t related to their sense of autonomy, control, confidence, challenge, connection, collaboration, conflict, support, self-worth, stress, feedback, role clarity, or enjoyment. The best performers were the ones who started their jobs believing that their work would have a positive impact on others. That led me to predict that givers would be more successful than takers, because they would be energized by the difference their actions made in others’ lives. I went on to test and support that hypothesis in a number of studies, but then I came across other studies in which generosity predicted lower productivity and higher burnout. Instead of trying to prove them wrong, I realized I was wrong—my understanding was incomplete. I set out to explore when givers succeed and when they fail, and that became my first book, Give and Take.

  * It’s possible to change even your deep-seated beliefs while keeping your values intact. Psychologists recently compared people who walked away from their religions with those who were currently religious and never religious. Across Hong Kong, the Netherlands, New Zealand, and the United States, they found a religious residue effect: people who de-identified with religion were just as likely to keep volunteering, and gave more money to charity than those who were never religious.

  * If you choose to make fun of yourself out loud, there’s evidence that how people react depends on your gender. When men make self-deprecating jokes, they’re seen as more capable leaders, but when women do it, they’re judged as less capable. Apparently, many people have missed the memo that if a woman pokes fun at herself, it’s not a reflection of incompetence or inadequacy. It’s a symbol of confident humility and wit.