Photo by Dustin Whitehead.
I am going to tell you about mind reading, but not the kind you normally associate with this term. I am not going to tell you about magic tricks that will allow you to astonish your friends at your next party. Nor am I going to tell you about telepathy, clairvoyance, or any kind of extrasensory power that creates a psychic connection with anyone. Instead, I am going to tell you about the kind of mind reading you do intuitively every day of your life, dozens of times a day, when you infer what others are thinking, feeling, wanting, or intending. The kind that enables you to build and maintain the intimate relationships that make life worth living, to maintain a desired reputation in the eyes of others, to work effectively in teams, and to outwit and outlast your competitors. The kind that forms the foundation of all social interaction, creating the web of presumptions and assumptions that enables large societies to function. The kind that sometimes feels like your real sixth sense.
Like any sense, this one can be stretched beyond its limits. When others’ experiences are so different, their cultures so foreign, or their history so unknown, our sixth sense clearly fails us. But these humbling experiences are relatively rare for most of us. Far more often, our ability to reason about the minds of others operates so quickly and easily that we hardly even notice we’re using it, or even pause to consider that our assumptions about the minds of others might be wrong.
Just how accurately do people understand each other? For many years, psychologists like me have been trying to answer this question by putting mind reading to the test. We might, for instance, ask a group of people to tell us how much they like you, then ask you to predict how much each of these people will report liking you, and then compare your predictions with the other people’s actual ratings to assess your accuracy. Or we might ask you to look at pictures of people who are happy or sad, proud or ashamed, elated or afraid, to see how accurately you can recognize each emotion.
How well do people perform on these tests? Are we as socially skilled as we think?
To get a sense of your actual abilities, let’s start with what is likely to be a very common and important bit of mind reading: trying to guess another person’s impression of you. Much of our everyday life is spent trying to understand how we’re being evaluated in the minds of others in order to help us create just the right impression. Does your boss think you are intelligent? Do your coworkers like you? Do your employees understand your instructions? Does your neighbor find you trustworthy? Does your spouse really love you? Or perhaps more important if you are young and single, do others think you are attractive?
Answering these questions accurately might be harder than you think. Consider, for instance, a study that analyzed a set of published experiments that all shared the same basic design.(i) In these experiments, people working in a group would be asked to predict how the other group members would rate them on a series of different traits. Researchers then compared these predicted ratings to the other group members’ actual ratings on the very same traits. The traits varied from one experiment to another and included qualities like intelligence, sense of humor, considerateness, defensiveness, friendliness, and leadership ability.
The groups varied in familiarity, with the members of some groups being fairly unfamiliar with one another (such as having met only once, in a job interview) and the members of other groups being very familiar with one another (such as having lived together for an extended time as roommates). If people knew exactly what others were thinking, then there would be a perfect correspondence between predicted and actual ratings. If people were clueless, then there would be no correspondence between the two. Statistically speaking, you measure relationships like these with a correlation, where perfect correspondence yields a correlation of 1 and no correspondence yields a correlation of 0. The closer the correlation is to one, the stronger the relationship.
First, the good news. These experiments suggested that people are pretty good, overall, at guessing how a group of others evaluates them, on average. The overall correlation in these experiments between predicted impressions and the average actual impression of the group was quite high (.55, if you are quantitatively inclined). To put that in perspective, this is roughly the same magnitude as the correlation between the heights of fathers and the heights of sons (around .5). It is not perfect insight into how we are seen by others, but it is also very far from being clueless. In other words, you probably have a decent sense of what others generally think of you, on average.
Now the bad news. These experiments also assessed how well people could predict the impression of any single individual within a given group. You may know, for instance, that your coworkers in general think you are rather smart, but those coworkers also vary in their impressions of you. Some think you are as sharp as a knife. Others think you are as sharp as a spoon. Do you know the difference?
Evidently, no. The accuracy rate across these experiments was barely better than random guessing (an overall correlation of .13 between predicted and actual evaluations, only slightly higher than no relationship whatsoever). Although you might have some sense of how smart your coworkers think you are, you appear to have no clue about which coworkers in particular find you smart and which do not. As one author of the study writes, “People seem to have just a tiny glimmer of insight into how they are uniquely viewed by particular other people.”(ii)
Perhaps, though, getting these broad and general evaluations right is still too much to expect of your sixth sense. What if we tried something simpler still, something specific and concrete that you’ve likely spent a considerable amount of time thinking and learning about? Can you accurately predict how attractive a member of the opposite sex will find you after being shown a photograph of you? You have, after all, lived a full life with yourself, looking at your face in the mirror every morning, and getting a sense of whether people tend to find you attractive or not. At certain points in your life (perhaps you’re at that point right now), you may have thought of little else.
And yet when Tal Eyal of Ben-Gurion University and I ran a series of experiments in which we asked people to predict how attractive they would be rated by a member of the opposite sex based on a photograph we took of them, we found that people’s predictions were no more accurate than chance guessing.(iii) Across two different experiments, the overall correlation between predicted and actual evaluations was 0. It’s not that our volunteers consistently thought they were more attractive than they were actually rated; rather, their predictions simply bore no relation to how they were actually rated on the basis of that photograph. It is often said that love is blind, but our participants did not even have a chance to be blinded by any love. They were just blind to begin with.
The central challenge for your social sense is that others’ inner thoughts are revealed only through the façade of their faces and bodies and language. Just as human beings have evolved the ability to use cues from that façade to see what truly lies beneath—to be mind readers—so, too, have human beings developed a skill to use their façade to mislead and misdirect others—to be liars and deceivers. Anyone who has ever been on the receiving end of “Does my butt look big in these pants?” knows that what you say to someone does not always reflect what you truly believe.
And yet, time and time again, researchers have found that our attempts to guess when another person is telling the truth and when they are lying are just that: little better than guesses. When one group of researchers evaluated decades of studies and hundreds of experiments that measured how well people could distinguish truths from lies, they found that people’s ability to spot deception was only a few percentage points better than a random coin flip: people were 54% accurate overall, when random guessing would have made them accurate 50% of the time.(iv) It’s easy to see how understanding other people can be a daunting task if you are unable to tell when they are lying to you and when they are not.
Illusions of insight
Although it may be challenging, perhaps reading the minds of others still isn’t very much of a problem in everyday life because our mind reading is finely tuned to those we know the best, such as our closest friends, relatives, colleagues, and spouses? Long-married spouses sometimes say they know each other so well that they can complete each other’s sentences. Really getting to know someone puts you in sync with them, you might think, so you’re able to understand each other’s thoughts without even uttering a single word. There is no doubt that friends, coworkers, and romantic partners think they know each other’s minds better than they know the minds of strangers. Is this confidence justified? Do we really know our friends and loved ones as well as we believe we do?
Again, the answer is no, but this answer comes in two parts. The first part is that you are indeed better able to read the minds of close friends and loved ones than those of strangers, although not by all that much. William Ickes, a pioneer in research on mind-reading accuracy, points out that in his experiments, “strangers read each other with an average accuracy rate of 20% percent” when videotaped and later asked to report their moment-by-moment thoughts and feelings.(v) “Close friends and married couples,” he reports, “nudge that up to 35% percent.” So yes, you do know what your spouse or a close friend likes and dislikes more than a random stranger would, but the gain may be surprisingly modest.
The second part of this answer, however, is that the confidence you have in knowing the mind of a close friend or romantic partner far outstrips your actual accuracy. Getting to know someone, even over a lifetime of marriage, creates an illusion of insight that far surpasses actual insight.(vi)
To see both of these results, imagine that you signed up with your sweetheart to participate in an experiment that my colleagues and I conducted, designed as something like the Newlywed Game (Science Edition). You sit in separate rooms, are told that your beloved will never see any of your answers, and are then given a list of 20 questions that measured the extent to which you agreed or disagreed with a wide range of attitudes and opinions. These questions included, “If I had my life to live over, I would sure do things differently,” “I would rather spend a quiet evening at home than go out to a party,” and “Our family is too heavily in debt today.” For each, you reported the extent to which you agreed on a scale ranging from one (labeled “strongly disagree”) to seven (labeled “strongly agree”), with four labeled “neither agree nor disagree.” In the room next to you, your partner is predicting how you will answer all of these questions, and also reporting the number (out of 20) that he or she believes are predicted correctly.
Let’s start with the good news. Partners predicted each other’s exact thoughts better than would be expected by random guessing alone. Our couples had been together for an average of 10.3 years, and 55% were married, so this finding is not particularly surprising. For instance, just by random guessing, couples would have predicted 2.85 items exactly correctly. Couples did better than this, albeit not by all that much. They predicted 4.9 of the items exactly correctly. The correlation between predicted and actual answers was a more encouraging .5. That’s not terrible. If you were a baseball player, this might be the equivalent of hitting a double.
Now for the not-so-good news. Bigger than the gap between actual accuracy and chance accuracy was the gap between how much partners actually knew about each other and how much they believed they knew. Remember, partners were perfectly accurate on an average of 4.9 out of the 20 items. But when we asked them to predict how many they believed they predicted correctly, they said, on average, 12.6 out of 20. Our partners had hit a double on our test, but they thought they had hit a home run.
This common illusion that we know other people better than we actually do can have dangerous consequences. In another experiment, volunteers watched videos of people either lying or telling the truth about whether they were HIV positive. Respondents were fairly confident that they could tell, predicting that they had guessed correctly 70% of the time. In fact, they did no better than what would be expected by chance alone, correctly identifying truths and lies only 52% of the time (when chance is 50%). As people gained more information about the person in the video, they became more confident but did not become any more accurate. When, during the rise of the AIDS epidemic, US Surgeon General C. Everett Koop implored Americans to “know your partner,” he presumably meant accurate knowledge rather than the illusion of knowledge.(vii)
The problem with our sixth sense about the minds of others is not that it is horribly flawed. It falls short of perfection when we test it under challenging circumstances, but it generally performs far better than chance guessing. And compared to the mental abilities of other species on this planet, our ability to think about the minds of others is what truly makes our brains superpowered. The problem is that the confidence we have in this sense far outstrips our actual ability, and the confidence we have in our judgment rarely gives us a good sense of how accurate we actually are.
My goal in Mindwise is to reduce the illusion of insight you have into the minds of others, both by trying to improve your understanding and by inducing a greater sense of humility about what you know—and what you do not know—about others. Only by recognizing the limits of your brain’s greatest sense will you have the humility to understand others as they actually are instead of as you imagine them to be. n
Adapted with permission from Mindwise: How We Understand What Others Think, Believe, Feel, and Want, by Nicholas Epley, published by Knopf. Copyright 2014. All rights reserved.
(i) David A. Kenny and Bella DePaulo, “Do people know how others view them? An empirical and theoretical account,” Psychological Bulletin, July 1993.
(ii) David A. Kenny, "Interpersonal perception: A social relations analysis," New York: Guilford Press, 1994.
(iii) In one experiment we found a slightly positive correlation between predicted and actual accuracy of .23, but we found an equally negative correlation in another experiment (-.24). Combining these experiments, the overall correlation we observe is roughly 0.
Tal Eyal, and Nicholas Epley, “How to seem telepathic: Enabling mind reading by matching self-construal,” Psychological Science, May 2010.
(iv) Charles F. Bond Jr. and Bella DePaulo, “Accuracy of deception judgments,” Personality and Social Psychology Review, August 2006.
(v) Kathryn H. Rollings, Ronan Cuperman, and William Ickes, “Empathic accuracy and inaccuracy,” in Leonard M. Horowitz and Stephen Strack (eds.) Handbook of interpersonal psychology: Theory, research, assessment, and therapeutic interventions, Hoboken, New Jersey: John Wiley & Sons, 2011.
(vi) Kenneth Savitsky, Boaz Keysar, Nicholas Epley, Travis Carter, and Ashley Swanson, “The closeness-communication bias: Increased egocentrism among friends versus strangers,” Journal of Experimental Social Psychology, 2011.
(vii) William B. Swann, David H. Silvera, and Carrie U. Proske, “On ‘knowing your partner’: Dangerous illusions in the age of AIDS?” Personal Relationships, 1995.