First Impressions

Nalini Ambady. Encyclopedia of Human Relationships. Editor: Harry T Reis & Susan Sprecher. 2009. Sage Publication.

Every interaction with a new person entails a first impression. The impressions we form of others and they of us influence practically every aspect of our lives, including our friendships, romantic relationships, and career prospects. Impression formation occurs rapidly, is often automatic and unconscious, and frequently occurs based on mere glimpses or instantaneous appraisals of “thin slices of behavior.” First impressions have important consequences in many different domains, including in the judgments of relationships, in job interviews, and in assessing others’ personalities. And first impressions are critical because these original evaluations are often lasting, influential, and set the stage for subsequent expectations, behavior, and interactions. This entry first discusses the accuracy and downstream consequences of first impressions in the domains of relationships, deception, job interviews, and personality. This discussion is then followed by an examination of the characteristics of good judges of first impressions.

Thin Slices of Behavior

Thin slices of behavior was coined by Nalini Ambady and Robert Rosenthal to describe brief excerpts of expressive behavior sampled from the behavioral stream that contain dynamic information and are less than 5 minutes long. Thin slices are an excellent way to examine first impressions, and the slices can be sampled from any available channel of communication, including the face, the body, speech, the voice, transcripts, or combinations of these. Hence, static images (e.g., photographs) and larger chunks of dynamic behaviors would not qualify as a thin slice.

The type of judgment being made affects accuracy. Thin-slice judgments are predictive and accurate only to the extent that relevant variables are observable from the thin slice sampled. Using the analogy of an onion, some characteristics, similar to the more visible, transparent outer layers of the onion, are easily observed and judged from thin slices. But other characteristics are hidden, similar to the inner layers of the onion, and are less easily judged from thin slices of behavior. Variables that are more observable and that are revealed through demeanor and behavior, such as extraversion, warmth, and likeability, are more easily and accurately judged from thin slices. In contrast, less observable variables, such as perseverance, are not easily or accurately judged from thin slices. This is because information regarding perseverance is more likely revealed through actions and behaviors that unfold over a relatively long period. Such information is less likely to be gleaned from thin slices of behavior.

Research on thin-slice judgments has had an impact across social, applied, and cognitive psychology and economics and has penetrated the popular literature as well. Thin-slice judgments are particularly useful in examining interpersonal relationships. For instance, judgments based on thin slices have been shown to accurately predict aspects of the doctor-patient relationship, including patient satisfaction and adherence to treatment, the relationship status of opposite-sex pairs interacting, judgments of rapport between two persons, and courtroom judges’ expectations of a defendant’s guilt.

Judging Relationships

Even nonhuman primates show an ability to quickly scan the social environment and recognize relationship patterns among others. These relationship patterns include those of kinship and status. During evolution, being able to make quick and accurate assessments of others’ relationship patterns is important for survival of the species.

Among humans, do first impressions provide signals about different types of relationships? Evidence indicates that people can judge different types of relationships. For example, people (and animals) can judge kinship from minimal cues. They can also judge the status of individuals in an interaction, such as who is the boss and who is the subordinate, from first impressions. These accurate responses depend on a correct interpretation of available verbal and nonverbal cues. Thus, both kinship and status relationships are judged from cues such as posture and gaze. Finally, individuals can judge whether people are strangers, friends, or romantic partners based on minimal cues, such as posture, facial expression, and gaze.

In addition to identifying the type of relationship people share, perceivers also make inferences about the quality of relationships. One important characteristic of the quality of the relationships is the level of rapport between partners. Rapport is defined as the extent to which a relationship is pleasant, engaging, and harmonious. When two people feel rapport toward each other, they are more attentive to each other, more positive in their behavior to each other, and better coordinated in their movements. Can rapport be judged from first impressions? It turns out that people are not good at judging rapport or whether other people are “in sync” from observations of brief video clips. Thus, it seems that people are better at judging types of relationships than they are at judging the quality of relationships.

Another characteristic of the quality of a relationship is the love that exists between the partners. Studies relying on self-reports have found no differences in relationship quality between couples that reported falling in love at first sight and those whose relationships had evolved from friendships. Recent work on speed dating has yielded interesting insights regarding first impressions in romantic relationships. One important insight is that preferences for certain characteristics of an ideal partner expressed before speed dating do not predict the characteristics of the partner selected from the speed-dating event. Thus, predictions before the dating event do not line up with actual choices, suggesting that prior theories and beliefs do not accurately predict first impressions and liking.

How good are people at judging love between others based on their first impressions? People are not good judges of the love between other couples from thin-slice video clips. Research has shown that people who were in love reported being more confident about their ability to judge whether other couples were in love but were actually less accurate than people not involved in romantic relationships. Thus, those in love may feel more confident about their ability to judge love, but are more biased than are those not in love. In sum, although perceivers have little difficulty categorizing the type of relationship that two people share, they are not as good at gauging the quality of that relationship, in terms of the rapport or the love between partners, from first impressions.


How well can deception be judged based on first impressions? Lying is ubiquitous in social life, and most lies are “white lies,” which are relatively harmless and are told to avoid friction and to maintain harmony in relationships. Other lies, however, are less innocuous, such as when liars maliciously manipulate individuals and organizations to advance their own self-interest.

Though people might lie quite often, they are not good at judging when others are lying. Meta-analyses of lie detection ability find average accuracy rate is 54 percent, only slightly better than that afforded by chance guessing.

One important moderator of accuracy in lie detection is the relationship between the liar and the detector. Research reveals that close friends show a substantial and significant improvement in lie detection accuracy (accuracy increased from 47 percent to 61 percent) over time as they get to know each other better, but less close friends show a small decrease in accuracy over time.

Factors such as expertise, experience, and formal training in lie detection do not seem to improve detection accuracy. “Professional lie catchers,” such as police officers, detectives, judges, secret service agents, and parole officers, are no more accurate at detecting deception than are students and other citizens. The one group of individuals that does seem to be more accurate at detecting lies than others is people with elevated levels of depression symptoms, who are better able to spot false reassurances and phoniness.

When perceivers do successfully distinguish truths and lies, they rely heavily on different streams of expressive behavior, including facial displays, gestures, and tones of voice. Are any of these channels particularly revealing of deception? The amount that people can control the information communicated by different channels of communication affects how revealing or “leaky” that channel of communication is considered. Verbal statements are believed to be the most controllable and therefore the least leaky channel of communication, followed, in order, by facial displays, gestures, and vocal tone. Vocal tone may be the leakiest channel of communication because the speaker’s perception differs from that of the listener. Because the voice sounds different to the speaker than to the listener, the speaker has difficultly monitoring and modulating it. Indeed, deception is most accurately detected from changes in the tone of voice compared with other channels of communication.

The bulk of research on deception detection comes from laboratory studies with undergraduate participants in which the liar’s motivation to be successful may be minimal. One meta-analysis of the literature examined whether the cues to deception become more transparent during “high-stakes” lies, when the liar has greater motivation to succeed. This analysis revealed that when liars are more highly motivated to succeed, they become tenser; specifically, they use less eye contact and use higher vocal pitch. This pattern is seen during real-life high-stakes situations, including murder, rape, and arson suspects undergoing police interrogations. Whether this greater transparency during higher stakes situations results in greater perceiver accuracy remains to be determined.

Job Interviews

First impressions are critical in job interviews, as documented in the plethora of books and articles on impression management in the job interview. Recent work suggests that interviewers’ early impressions affect interview outcomes. In general, studies have found that the more favorable the interviewers’ impressions on the preinterview measures are, the more positively they treat the applicant, and the more likely they are to extend an offer to the applicant. Specifically, interviewers use a more positive vocal style with applicants who have made a positive first impression. Moreover, interviewers try to recruit the applicants who have made more positive first impressions by attempting to “sell” the company and the job to a greater extent and providing more information about the company than they do to applicants who have made less positive first impressions. Interviewer behavior also affects applicant communication style, and the more positive the interviewer, the more positive the applicant behavior. Thus, interviewers’ first impressions affect how they conduct the interview, and how interviewers conduct the interview is subsequently related to applicants’ behavior and their evaluation by the interviewer.


The ability to accurately gauge others’ personalities is also central to the development and maintenance of interpersonal relationships. How accurate are our first impressions of other’s personalities? The answer depends on the personality trait that is being judged.

In a pioneering study in 1938, Stanley Estes compared perceivers’ impressions of personality with targets’ self-reported assessments. After viewing two-minute film clips of people engaged in expressive movement, perceivers were able to judge emotionality, inhibition, and apathy at levels above chance. Half a century later, other researchers have examined the accuracy of perceptions of the Big Five personality traits (Extraversion, Agreeableness, Conscientiousness, Neuroticism, and Openness to experience) from first impressions based on brief encounters as well as on videotapes and thin slices. Judgments of extraversion and conscientiousness show the highest correspondence with self-assessments, but neuroticism, agreeableness, and openness were less well judged. Even when targets’ romantic partners or family members are asked to describe targets’ personality, extraversion and conscientiousness emerge as the most accurately judged traits from first impressions.

Another line of research has shown considerable agreement or consensus regarding personality traits of complete strangers. Several studies suggest that unacquainted judges exhibit a surprisingly high degree of consensus in their impressions of a stranger’s personality. Although consensus increases with increased exposure, independent perceivers still agree in their assessments of a target’s personality even in the absence of interaction with that target. Consensus is again particularly strong on two personality traits: extraversion and conscientiousness.

Perceivers can be surprisingly accurate in their impressions even without the benefit of direct interaction. Impressions formed after brief observations lasting a mere 10 seconds are as accurate as those based on 5-minute observations, indicating that personality is often revealed in thin slices of expressive behavior. Moreover, a surprising amount of information about personality is revealed in the environments we construct—whether real or virtual. For instance, first impressions of individuals’ personalities based on their bedrooms or offices have been shown to correlate with their self-reports as well as with close acquaintances’ ratings of the target person’s personality.

Individual Differences: The Good Judge of First Impressions

What are the characteristics of good judges of first impressions? Several factors affect individual differences in the accuracy of first impressions of relationships, including differences in personality and motivation. For instance, individuals who are more highly motivated to understand others and who have greater social skill and competence are more accurate in their first impressions. Conversely, people who are self-preoccupied perform poorly. Knowledge about social relations is also an important moderator. People who have had advanced theatrical training score higher on some tasks that measure the accuracy of first impressions. This could be because their theatrical training sensitizes them to the meaning of particular gestures, facial displays, and vocal patterns.

Women tend to be more accurate judges of first impressions and nonverbal behavior than men do. This female advantage may be the result of socialization and societal expectations. Women may have more knowledge of nonverbal cue meanings and may be more sensitive to such cues.

Increasing evidence also indicates that the most accurate judges of first impressions are people who are socially well adjusted. Certain groups of people with clinical disorders such as autism, schizophrenia, mania, and alcoholism are particularly impaired in their ability to form accurate first impressions.

Even though, overall, both children and adults who enjoy greater interpersonal success are generally better judges of first impressions, individual differences are tempered by cultural and subcultural exposure. For example, people are better at accurately judging targets from their own culture and cultures similar to their own than with more foreign targets.

Self-Fulfilling Prophecies in Interpersonal Relationships

Both accurate and inaccurate first impressions may affect the perceiver’s subsequent behavior toward the target as revealed in the studies on job interviews discussed. Perceiver behavior shapes and constrains targets’ responses, creating self-fulfilling prophecies (also termed interpersonal expectancy effects or behavioral confirmation). A self-fulfilling prophecy is an originally false definition of the situation that evokes behaviors making the false conception come true. Empirical research on interpersonal self-fulfilling prophecies emerged from Robert Rosenthal’s influential work showing that experimenters’ expectations can unwittingly bias the results of their experiments with both human and animal participants. For instance, in one study, the rats of experimenters who were led to believe that their rats were good at running mazes, ran the maze faster than did the rates of experimenters who were led to believe that their rats were not good at running mazes. The false expectations of the experimenter thus evoked behavior from their animal subjects that validated the expectations, creating a self-fulfilling prophecy.

Rosenthal and his colleagues later documented this phenomenon in other interpersonal domains, including the classroom. They showed that the expectations of teachers affected the intellectual performance of students. Teachers were told at the beginning of the year that some students in their class (selected randomly by the experimenters) were going to be “late bloomers” and show gains in intellectual competence in the next few months of school. At the end of the year, those students showed higher intellectual test performance scores than did children in the control group, indicating that teachers’ expectations affected student performance. Although the idea that experimenters’ and teachers’ expectations could create self-fulfilling prophecies was initially met with considerable resistance, eventually enough replications were published to quell most critics. Meta-analyses of the literature on expectancy effects have subsequently shown that these effects are statistically significant and accompanied by a mean effect size that is moderate in magnitude.

Self-fulfilling prophecies have consequential effects on many different types of interpersonal relationships. In one study, for instance, male participants were given a description sheet of a female with a photograph of either an attractive or unattractive woman and were told that they were going to have a telephone conversation with her. After seeing the photograph but before conversing, men expected attractive women to embody more positive traits. After conversing with a female confederate, men’s impressions remained consistent with their initial expectations: Women believed to be attractive were rated more positively, whereas women believed to be unattractive were rated more negatively. Interestingly, independent coders who were unaware of which photograph the men had seen rated the audiotaped conversations of the women who had randomly been paired with the attractive photos more positively. Further analyses of the audiotapes showed that male participants treated partners whom they believed to be less attractive in a less warm, sociable manner, thereby eliciting a more negative reaction from these women than from partners whom the men believed were more attractive. Thus, interpersonal expectations, even over a brief telephone conversation, created a self-fulfilling prophecy that resulted in female participants behaving in ways consistent with their male partners’ initial appearance-based expectations. Similar results have been found in other behavioral confirmation studies that have manipulated impressions or expectancies regarding a variety of target traits, including personality, race, hyperactivity, and mental illness.

Thus, both accurate and inaccurate first impressions lay the foundation for subsequent behavior. First impressions are critical in the formation of fundamental social relationships and influence both whether a relationship is established and the quality of the relationship once it is established.