'Thin slices' of life

For the past several years, bars and restaurants around the country have been hosting a new type of dating ritual. At events organized by companies like HurryDate and 8MinuteDating, dozens of eligible singles meet up for a round of lightning-quick "dates"--sometimes as short as three minutes each--with 10 or 20 other people. At the end of each date, participants mark on a card whether or not they'd like to see the other person again.

When the evening is over, event staff correlate the results and provide contact information to any mutually interested pair.

The speed-dating concept rests on a simple premise: that a few minutes can be plenty of time to size a person up and evaluate compatibility.

It's no secret that people often judge each other based on immediate intuitions. We make split-second judgments of strangers all the time: When a student decides to drop a class based on minutes of a professor's teaching or when we meet someone at a party, our first impressions count. For as long as parents have admonished their children "don't judge a book by its cover," most of us have been doing just that.

In the past 15 years, though, researchers have become interested in trying to systematically answer the questions: How much can those fleeting first impressions really tell us? How much of people's personality is it possible to intuit within a few seconds, or minutes, of meeting them?

The psychologists are finding that in some cases our social intuition is indeed amazing--we can sometimes pick up a remarkable amount of information about a person's personality or skills in just a few seconds. At other times, though, we can all be social dunces--oblivious to these nuances.

Old questions, new answers

Until the mid-1980s, psychologists mostly weren't interested in answering those questions at all, says Nalini Ambady, PhD, a social psychologist at Tufts University. Researchers wanted to study the processes by which people judged others, not how accurate those judgments were, she explains.

Dave Kenny, PhD, a psychologist at the University of Connecticut, was one of the first researchers to look into the accuracy of first impressions. In a 1988 study, he examined whether people's first impressions of strangers' personalities matched up with the strangers' self-ratings, scores on personality tests or other measures of personality.

Kenny based his research on a 1966 study by two University of Michigan psychologists, Warren Norman, PhD, and Lewis Goldberg, PhD. In that study, which was published in the Journal of Personality and Social Psychology (Vol. 4, No. 6), Norman and Goldberg asked University of Michigan students to rate their peers' personalities on the first day of class, before the students had had a chance to interact. They found that the students' ratings of one another tended to agree with their self-ratings, particularly on the traits "sociable" and "responsible."

This result was, Kenny says, overlooked for more than 20 years, until he and his colleagues picked up and extended the work in 1988.

"At the time, it was so counterintuitive and not what people expected--even Norman and Goldberg hadn't expected it--that they downplayed it," Kenny says. "That's just not where the field was at the time."

In his study, which appeared in the Journal of Personality and Social Psychology (Vol. 55, No. 3), Kenny divided more than 250 students in a psychology class into groups of four. After making sure none of the group members knew each other and had not had a chance to talk, the students rated all members of their group, including themselves, on five factors that each represented a personality trait. The factors were sociable (representing extroversion), good-natured (representing agreeableness), responsible (representing conscientiousness), calm (representing emotional stability) and intellectual (representing culture). Again, as in the Norman and Goldberg study, the strangers' first impressions of each other correlated significantly with self-ratings for the traits "extroversion" and "conscientiousness."

Self-ratings, however, don't necessarily provide accurate descriptions of everyone's personality.

"Look at all those movie stars who say they're really shy," Kenny points out. "Plenty of people think they're less extroverted than they really are."

With this in mind, he and colleague Maurice Levesque, PhD, conducted another study, also published in the Journal of Personality and Social Psychology (Vol. 65, No. 6). This time, groups of four strangers rated each other on the five personality traits. Then, the strangers met in pairs and were videotaped talking to each other. Later, judges watched the extensive videotapes and rated each subject's extroversion, based on the amount of time he or she spent talking, the number of arm movements and other factors. Again, the strangers' first-impression ratings of extroversion strongly correlated with people's rated levels of ext level extroversion as seen on the videotape.

In the 14 years since Kenny's studies, numerous researchers have replicated and expanded his results. Overall, the studies have indicated that people are good at sensing a stranger's level of extroversion or sociability. Some have suggested people can sense the other four traits as well (particularly conscientiousness and agreeableness), but those results are much more mixed, and the conclusions murky.

This makes sense, says Ambady, who analogizes personality to the layers of an onion. "The layers near the top," she says, "are the easiest to pick up."

Extroversion may be the most obvious of personality traits, but there are some researchers who think that other, more internal traits can also be visible.

Frank Bernieri, PhD, a social psychologist at Oregon State University, agrees that extroversion is the most easily visible trait. But, he notes, that fact could be partially explained by the experimental designs used by most researchers in the area.

"The contexts in which we've asked people to evaluate others are social interactions where, of course, extroversion would be the most obvious trait," he says.

In fact, Bernieri firmly believes that all aspects of personality are embedded in behavior.

To test this, he and his student Amber McLarney-Vesotski, PhD, now a psychology instructor at Alpena Community College in Michigan, devised situations in which they believed the other four aspects of a person's personality would become more apparent. In one, for example, a stranger asks the subject his or her opinion of a Rorschach-blot-like painting--a task designed to determine the subject's degree of openness. In another, the subject plays a game with an obstinate, irritating stranger--designed to determine neuroticism or calmness and emotional stability. The researchers then showed 10-second video clips of these situations to a panel of college-student judges, and the judges were able to predict all five aspects of the subjects' personalities at a better-than-chance level.

Beyond personality

Sometimes, we aren't looking to divine someone's overall personality or intelligence based on a first impression; we simply want to know how good they will be at a particular skill or set of skills, like teaching. Tufts psychologist Nalini Ambady has found that students, for example, are surprisingly good at predicting a teacher's effectiveness based on first impressions. She creates these first impressions with silent video clips of teachers--clips she calls "thin slices."

In a 1993 study published in the Journal of Personality and Social Psychology (Vol. 64, No. 3), Ambady and a colleague videotaped 13 graduate teaching fellows as they taught their classes. She then took three random 10-second clips from each tape, combined them into one 30-second clip for each teacher and showed the silent clips to students who did not know the teachers. The student judges rated the teachers on 13 variables, such as "accepting," "active," "competent" and "confident." Ambady combined these individual scores into one global rating for each teacher and then correlated that rating with the teachers' end-of-semester evaluations from actual students.

"We were shocked at how high the correlation was," she says. It was 0.76. In social psychology anything above 0.6 is considered very strong.

Curious to see how thin she could make her slices before affecting the student judges' accuracy, Ambady cut the length of the silent clips to 15 seconds, and then to six. Each time, the students accurately predicted the most successful teachers.

"There was no significant difference between the results with 30-second clips and six-second clips," Ambady says.

In a later experiment in the same study, she cut out the middleman--the global variable--and simply asked students to rate, based on thin-slice video clips, the quality and performance of the teachers. Again, the ratings correlated highly with the teachers' end-of-semester evaluations. Ambady also replicated her results with high school teachers.

Of course, one could argue that the true measure of a teachers' effectiveness is not what their students say about them, but how much those students learn. Ambady, acknowledging this, has tried to measure whether students actually learn more from teachers who give a first impression of effectiveness.

In an as-yet-unpublished study, she videotaped groups of five participants, one of whom was randomly assigned to be the "teacher." The teacher spent time preparing a lesson, and then taught students a mathematical language in which combinations of letters represent different numbers, as in 10=djz or 3=vfg. The students took a test at the end of the lesson to measure their knowledge of the new language. Then, as before, strangers watched 10-second video clips of the teachers and rated them on the same variables as in the first study. The thin-slice ratings of teacher effectiveness, Ambady says, significantly predicted students' performance on the test.

"Students learned more from teachers who were seen in the thin slices as having the qualities of a better teacher," Ambady says.

A cautionary note

Do all of these studies, then, suggest that people should simply trust their gut instincts and, when meeting strangers, go with their first impressions? Is what you see really what you get?

Well, it's not quite that simple. First, there is the inevitable counterexample. First impressions can, sometimes, be dreadfully wrong. Serial killer Ted Bundy was a seemingly normal, attractive man who fooled not only his victims, but two women with whom he had long-term relationships, Bernieri points out.

"We can be distracted by the most visible and salient aspects of a person's personality," Bernieri says. Physical attractiveness and charisma (a large part of which is extroversion) can hide the inner layers of the onion.

"I believe that personality is truthfully encoded within the first 30 seconds of behavior," Bernieri says, "but that doesn't mean we're going to accurately get all of it all the time." Dave Kenny agrees. "People can certainly make inaccurate judgments, sometimes tragically so," he says.

It's also important to remember that all first-impression studies deal with aggregates, not individuals. "When we talk about accuracy, we're not looking at single judgments, we're looking at the average of a lot of judgments," says Kenny.

When a study indicates that people, in the aggregate, can judge whether strangers are extroverts or introverts at better-than-chance levels, it does not mean that everyone is equally good at doing so. "People do vary in their social intelligence," says Bernieri. "There are dunces among us who just never get it."

Even people with high "social intelligence" might vary in their ability to accurately judge a stranger, depending on something as simple as how they feel when they are making the judgment.

In a 2002 study published in the Journal of Personality and Social Psychology (Vol. 83, No. 4), Ambady found that people who were induced into a happy mood by watching a scene from a happy movie were able to more accurately predict a teacher's effectiveness from a thin-slice video clip than were people who were induced into a sad mood. Ambady thinks that this might be because people who are in a sad mood don't trust their snap judgments--they might come to the same immediate first impression as someone in a happy mood, but then doubt themselves and start second-guessing.

Making and judging first impressions is also, of course, a two-way street. The judge, the judged and the rapport and similarity between the two all affect the accuracy of a first impression. Few studies have specifically looked at cross-cultural first impressions, but research on nonverbal sensitivity has shown that people from similar cultures are better at judging each others' personalities based on nonverbal cues than people from very different cultures.

"An extroverted Chinese person might look completely different than an extroverted American," says Bernieri. "But even though they might not seem loud and talkative to us, they'll be louder and more talkative than another Chinese person. The key to judging someone accurately is that you have to be able to compare within cultures, not between cultures."

Overall, then, psychologists have accumulated a decade-and-a-half's worth of evidence that people are, collectively, better than you might think at deciphering a stranger's personality and abilities based on a first impression. But individually, it seems, we're still on our own.

"I would never give anyone a blanket statement that they should trust their first impressions, or not," says Nalini Ambady. "That's too dependent on the person, the context of the first impression, everything."