In 1946, a psychologist named Solomon Asch conducted one of the more elegant experiments in social psychology's history. He gave two groups of participants a list of adjectives describing a hypothetical person and asked them to form an overall impression. Both groups received the same six adjectives — intelligent, industrious, impulsive, critical, stubborn, envious — but in different orders. The first group read them in the order listed above, beginning with positive traits. The second group read them in reverse order, beginning with the most negative.
The impressions formed were dramatically different. The first group described a capable, energetic person whose impulsiveness and stubbornness were seen as manageable flaws. The second group described a problematic, difficult person whose intelligence was qualified by their fundamental negativity. The information was identical. The meaning attributed to it was not. The initial framing — the first few adjectives — had structured everything that followed.
This is the primacy effect in impression formation, and it has been replicated, extended, and contextualized in thousands of subsequent studies. First impressions are not merely influential — they are structurally determinative of how all subsequent information about a person is processed. The initial impression creates a framework that filters incoming data, weights it selectively, and bends ambiguous evidence toward consistency with itself. Understanding why this happens is not merely an intellectual exercise. The first impression you make in a job interview, a new team, or a first meeting with an important client may shape how that person perceives everything you do for months or years.
"You never get a second chance to make a first impression." — Will Rogers
Key Definitions
Primacy Effect: Solomon Asch's (1946) finding that information presented first in a sequence exerts disproportionate influence on overall impressions, because it establishes an interpretive frame through which subsequent information is processed.
Halo Effect: Edward Thorndike's (1920) concept describing the cognitive bias in which a positive (or negative) overall impression of a person systematically biases the evaluation of that person's specific, independent characteristics.
| First Impression Mechanism | Effect | Research Basis |
|---|---|---|
| Primacy effect | First information anchors all subsequent judgment | Asch impression formation studies (1946) |
| Confirmation bias | We seek information that confirms the initial impression | Wason (1960); widely replicated |
| Halo effect | One positive trait generalizes to unrelated traits | Thorndike (1920); strong in interview research |
| Attribution asymmetry | We explain consistent behavior as dispositional, not situational | Fundamental attribution error research |
| Schema resistance | Once a schema forms, contradicting information is dismissed or reinterpreted | Social cognition research (Fiske & Taylor, 1991) |
Confirmation Bias: The tendency to search for, interpret, and remember information in ways that confirm existing beliefs. The primary maintenance mechanism of first impressions once formed.
Thin-Slicing: Nalini Ambady's term for the ability to make accurate judgments about a person from very brief behavioral exposures. Accurate for some characteristics, systematically inaccurate for others.
Bayesian Updating: A statistical framework describing how rational belief revision should occur in response to new evidence. Applied to social cognition, it provides a normative model against which the conservatism of actual impression updating can be measured.
Solomon Asch and the Architecture of First Impressions
Asch's 1946 paper, 'Forming Impressions of Personality,' published in the Journal of Abnormal and Social Psychology, remains one of the foundational documents of social cognition research. Its influence extends not just through the primacy effect finding but through the theoretical framework Asch proposed to explain it.
Asch argued, contra earlier associationist accounts that treated impressions as simple aggregations of trait evaluations, that impression formation is a configural, meaning-making process. The overall impression is not a sum of parts but an organized gestalt — a structured whole in which each element takes on meaning from its relationship to the others. The initial elements, because they arrive before the structure is formed, have disproportionate influence on the structure itself. Later elements are interpreted within the context the early ones created.
This insight was significant: it placed impression formation in the domain of active interpretation rather than passive recording. The mind is not a camera; it is an interpreter. And interpreters, as hermeneutic theorists have long noted, always begin from somewhere — from a pre-understanding, a frame, a context. In social cognition, the first few data points provide that frame.
The Central Trait Effect
Asch's research also established the 'central trait' effect: some personal attributes exert disproportionate influence over the overall impression regardless of their position in the sequence. In his studies, 'warm' versus 'cold' had a far greater effect on overall impressions than other trait variations. Participants who received 'warm' in their list attributed a constellation of positive social qualities to the hypothetical person; those who received 'cold' attributed an entirely different character.
The warm-cold dimension corresponds to what subsequent research identified as the primary dimension of social perception: communion, or relational warmth. Research by Nick Ambady, Amy Cuddy, and colleagues on social perception has consistently found that two dimensions — warmth and competence — organize the vast majority of variance in person perception. Warmth is evaluated faster, carries more emotional weight, and is harder to update than competence evaluations. A person perceived as cold from the outset faces a particularly persistent first impression because warmth is a central trait.
The Halo Effect: How One Good Thing Floods Everything Else
Edward Thorndike's 1920 observation about military performance ratings introduced the halo effect concept, but its full implications for social cognition were not appreciated until subsequent decades of experimental work. The halo effect is not a minor bias affecting edge cases — it is a pervasive, systematic distortion that affects virtually all global impressions of complex entities.
Richard Nisbett and Timothy Wilson's 1977 study demonstrated the halo effect with particular elegance. They showed students a videotaped interview with a lecturer who was either warm and personable or cold and distant. After watching the tape, students were asked to rate the lecturer on various specific attributes: his appearance, mannerisms, and accent. Participants who had seen the warm version rated all three attributes significantly more positively than those who had seen the cold version — despite the specific physical attributes being identical in both videos.
Critically, when asked whether their liking for the instructor had influenced their ratings of his specific attributes, participants denied it. They believed they had assessed the attributes independently. The halo effect is largely invisible from the inside — it operates beneath the level of conscious awareness and does not feel like a bias. This is precisely what makes it so persistent.
In professional settings, the halo effect has documented consequences for performance evaluation, hiring decisions, and salary determination. Research by Landy and Sigall (1974) found that identical essays were rated significantly higher in quality when they were believed to have been written by attractive authors. Subsequent research has replicated halo effects in interview settings, performance reviews, and academic grading. The effect compounds the primacy effect: the positive or negative valence of the initial impression creates a halo that biases all subsequent specific evaluations.
Confirmation Bias: The Engine of Impression Persistence
If the primacy effect explains how first impressions are formed and the halo effect explains how they generalize across attributes, confirmation bias explains how they persist despite contradictory evidence. It is the primary maintenance mechanism of first impressions, and its operation in impression maintenance is more insidious than its better-known operation in belief persistence generally.
The basic mechanism of confirmation bias in social perception involves three processes operating simultaneously. Selective attention means that information consistent with the existing impression is more likely to be noticed and attended to. Selective interpretation means that ambiguous behaviors are interpreted in ways consistent with the impression — the same act is seen differently depending on whether it is performed by someone liked or disliked, trusted or suspected. Selective memory means that impression-consistent information is more accessible in recall than inconsistent information.
Together, these processes create what social psychologists call 'behavioral confirmation': the first impression is not just maintained against evidence, it becomes self-fulfilling. Mark Snyder, Elizabeth Tanke, and Ellen Berscheid's 1977 study demonstrated the full mechanism. Male participants who believed they were speaking on the phone with an attractive woman (based on a fake photo) behaved more warmly toward her; the women, in turn, responded with increased warmth. Independent judges who listened only to the women's side of the conversation rated the women who had been treated as attractive as significantly more friendly and likable — not because they were, but because the interaction had made them so. The first impression created the evidence that confirmed it.
Thin-Slicing and Accuracy: When First Impressions Are Right
The persistance of first impressions would be less troubling if they were reliably accurate. Nalini Ambady's research on thin-slicing raises this possibility. Her finding that very brief behavioral observations predict certain outcomes at above-chance levels suggests that first impressions sometimes contain genuine information.
The research on thin-slicing accuracy has been replicated and extended in multiple domains. Ambady and colleagues found that thirty-second video clips of teachers predicted end-of-term student evaluations. Studies on first-meeting impressions found that personality ratings made in seconds correlated with ratings made after extended acquaintance for extraversion and conscientiousness. Even more striking, a 2005 study found that naive raters could predict election outcomes from photographs of candidates' faces better than chance — suggesting that the cues that trigger first impressions contain real information about characteristics that affect real-world performance.
The important qualification is that thin-slicing accuracy is highly domain-specific. It works for characteristics that are genuinely and consistently expressed through observable behavior and appearance. It fails, often dramatically, for characteristics that are not — honesty, specific domain competence, intelligence when decoupled from communication fluency. And the subjective experience of forming a first impression does not accurately distinguish between accurate and inaccurate ones. People feel equally confident in first impressions that turn out to be accurate and those that turn out to be wrong.
Bayesian Updating: The Normative Model and Its Failure
Bayesian reasoning provides a useful normative framework against which actual impression updating can be measured. A rational Bayesian agent updates beliefs in proportion to the evidential weight of new information. Strong disconfirming evidence should produce large updates; weak disconfirming evidence should produce small updates. The prior strength of the belief should affect how much updating occurs, but it should not prevent updating when the evidence is clear.
Actual impression updating is far more conservative than the Bayesian model predicts. Research on impression change consistently finds what psychologists call 'conservatism' — the tendency to update beliefs less than the new evidence warrants. A person who has formed a strong negative impression updates it only partially in response to clearly positive disconfirming behavior, assigning more variance to situational factors ('they were probably trying to impress someone') than to trait revision.
The mechanisms behind this conservatism include the confirmation bias processes described above, but also a deeper feature of mental models: they are actively used to guide behavior, and frequent revision of actively used models is cognitively expensive. A stable mental model of a person, even an inaccurate one, provides predictive value. The value of maintaining a working model, even at the cost of some accuracy, may systematically outweigh the value of accuracy-driven revision in the economics of social cognition.
Resetting a Bad First Impression
Given the architecture of impression persistence — primacy effects, halo distortion, confirmation bias maintenance — how can a bad first impression actually be changed? The research is specific about what does and does not work.
Simple time passage does not reliably change first impressions. If additional interactions provide only ambiguous information, confirmation bias processes will interpret that ambiguity to maintain the original impression. What is needed is information that is both clearly inconsistent with the initial impression and attributable to stable characteristics rather than situational factors.
Research on disconfirmation and impression revision suggests several components of effective impression reset. First, the disconfirmation needs to be substantial and repeated across multiple contexts. A single counter-example is too easily explained away ('they were just having a good day'). Consistent disconfirmation over time, in different situations, is harder to attribute to situational factors.
Second, directly addressing the initial impression — 'I think we may have gotten off on the wrong foot, and I'd like to change that' — can accelerate updating by making the revision explicit and conscious rather than leaving it to the slow process of automatic updating. This requires social courage but leverages metacognition: bringing the impression formation process into awareness where it can be examined rather than allowing it to operate beneath conscious scrutiny.
Third, seeking contexts where the initial impression-creating dynamic is absent helps. If a bad first impression was formed in a high-pressure, adversarial context, a low-stakes collaborative context creates conditions under which different information can emerge and a different impression can form alongside the original one.
The Role of Warmth in Impression Recovery
Asch's central trait research, and the broader research on warmth as the primary dimension of social perception, has a specific implication for impression recovery. Warmth evaluations are both formed quickly and updated reluctantly. The impression that someone is cold, closed, or unfriendly is particularly persistent because warmth is central — it colors all other attribute evaluations and is the first dimension assessed.
Research by Cuddy, Fiske, and Glick on the stereotype content model suggests that low-warmth impressions are especially resistant to revision because they trigger threat responses that maintain vigilance rather than openness to new information. The practical implication: impression recovery after a cold or hostile initial encounter requires sustained, patient, genuinely warm behavior over time. It cannot be accomplished with a single gesture of friendliness, no matter how genuine.
Practical Takeaways
Understand that first impressions are structurally influential, not merely influential. The information you present first creates the interpretive frame for everything that follows, which means the order and framing of your self-presentation matters — not just the content.
Recognize the halo effect in your own evaluations. When you have a generally positive or negative overall impression of someone, actively seek specific evidence about the particular attribute you are evaluating rather than letting the halo do the work. This is a deliberate effort; it will not happen automatically.
Treat impression-disconfirming information as more important than impression-confirming information when evaluating someone. Given that confirmation bias systematically overweights the latter, deliberate attention to the former compensates for the bias rather than amplifying it.
When you need to reset a bad first impression, commit to a sustained, consistent strategy rather than a single corrective gesture. The impression maintenance mechanisms are robust; the disconfirmation needs to be substantial and repeated to overcome them.
Do not confuse confidence in first impressions with accuracy. The felt certainty of a first impression judgment is not correlated with its accuracy. Maintaining calibrated uncertainty — especially for characteristics that are not reliably expressed through observable behavior — is both epistemically and practically appropriate.
References
- Asch, S. E. (1946). Forming impressions of personality. Journal of Abnormal and Social Psychology, 41(3), 258-290.
- Thorndike, E. L. (1920). A constant error in psychological ratings. Journal of Applied Psychology, 4(1), 25-29.
- Nisbett, R. E., & Wilson, T. D. (1977). The halo effect: Evidence for unconscious alteration of judgments. Journal of Personality and Social Psychology, 35(4), 250-256.
- Snyder, M., Tanke, E. D., & Berscheid, E. (1977). Social perception and interpersonal behavior: On the self-fulfilling nature of social stereotypes. Journal of Personality and Social Psychology, 35(9), 656-666.
- Ambady, N., & Rosenthal, R. (1992). Thin slices of expressive behavior as predictors of interpersonal consequences: A meta-analysis. Psychological Bulletin, 111(2), 256-274.
- Ambady, N., & Rosenthal, R. (1993). Half a minute: Predicting teacher evaluations from thin slices of nonverbal behavior and physical attractiveness. Journal of Personality and Social Psychology, 64(3), 431-441.
- Willis, J., & Todorov, A. (2006). First impressions: Making up your mind after a 100-ms exposure to a face. Psychological Science, 17(7), 592-598.
- Cuddy, A. J. C., Fiske, S. T., & Glick, P. (2008). Warmth and competence as universal dimensions of social perception: The stereotype content model and the BIAS map. Advances in Experimental Social Psychology, 40, 61-149.
- Kahneman, D. (2011). Thinking, Fast and Slow. Farrar, Straus and Giroux.
- Fiske, S. T., Cuddy, A. J. C., & Glick, P. (2007). Universal dimensions of social cognition: Warmth and competence. Trends in Cognitive Sciences, 11(2), 77-83.
- Gladwell, M. (2005). Blink: The Power of Thinking Without Thinking. Little, Brown.
- Lord, C. G., Ross, L., & Lepper, M. R. (1979). Biased assimilation and attitude polarization: The effects of prior theories on subsequently considered evidence. Journal of Personality and Social Psychology, 37(11), 2098-2109.
Frequently Asked Questions
What is the primacy effect in impression formation?
The primacy effect in impression formation was established by Solomon Asch in his landmark 1946 paper on impression formation, published in the Journal of Abnormal and Social Psychology. Asch presented participants with lists of trait adjectives describing a hypothetical person and found that the order in which the traits were presented dramatically affected the overall impression formed. When the list began with positive traits ('intelligent, industrious, impulsive...'), the person was rated far more positively than when the identical traits were presented in reverse order ('envious, stubborn, critical...'). The initial information established an interpretive frame that shaped how all subsequent information was understood. A person described first as 'intelligent' was seen as impulsive in an interesting, energetic way. The same trait presented last, after negative descriptors, was seen as dangerous unpredictability. The primacy effect is the foundation of modern first impression research.
What is the halo effect and how does it distort impressions?
The halo effect is a cognitive bias in which an overall positive (or negative) impression of a person influences how specific attributes of that person are evaluated. It was first described by psychologist Edward Thorndike in 1920, who noticed that military officers' ratings of their men on different characteristics were suspiciously highly correlated — those rated highest on intelligence also tended to be rated highest on physical appearance, leadership, and personal qualities, in ways that seemed to reflect one general impression rather than independent assessments. In impression formation, a positive first impression creates a halo that systematically biases subsequent evaluations of the person's specific qualities. Research by Richard Nisbett and Timothy Wilson (1977) demonstrated the halo effect in controlled experiments: participants who viewed a likable instructor rated his accent, mannerisms, and physical appearance significantly more positively than participants who viewed the same instructor being unlikable — despite the specific attributes being identical.
How does confirmation bias maintain first impressions?
Confirmation bias — the tendency to search for, interpret, and remember information in ways that confirm existing beliefs — is the primary cognitive mechanism that makes first impressions persistent. Once an initial impression is formed, it creates a framework that biases the processing of all subsequent information about the person. Ambiguous behavior is interpreted in ways consistent with the existing impression. Impression-consistent information is weighted more heavily and remembered more accurately than inconsistent information. Impression-disconfirming information is explained away or attributed to situational factors. Research by Mark Snyder and colleagues on behavioral confirmation demonstrated that the effect is not purely cognitive — it can become self-fulfilling, as the person holding the impression behaves in ways that elicit behavior consistent with their expectation from the target.
Are first impressions ever accurate?
Research on thin-slicing accuracy suggests that first impressions are sometimes surprisingly accurate and sometimes dramatically wrong, depending on the characteristic being assessed. Nalini Ambady's research established that very brief behavioral observations predict certain outcomes — teaching effectiveness, surgical liability, personality dimensions like extraversion — at above-chance levels. Characteristics that are genuinely expressed through observable behavior and appearance are assessed reasonably accurately from brief exposures. Characteristics that are not — like honesty, intelligence, or domain-specific competence — are not assessed accurately from brief exposures, even though observers typically feel confident. The practical implication is that first impressions are selectively accurate in ways that are not obvious from the subjective experience of forming them, which makes uncritical trust in them unreliable as a general strategy.
How can you reset a bad first impression?
Resetting a bad first impression is genuinely difficult but not impossible, and the research on impression updating provides specific guidance. The most important finding is that inconsistency with the initial impression needs to be both clear and attributed to stable characteristics of the person rather than situational factors. Ambiguous disconfirming behavior tends to be explained away; unambiguous, repeated disconfirming behavior in multiple contexts is harder to explain away. Research by Chester and DeSoto (1979) on impression revision found that consistent, repeated disconfirmation over time gradually updates impressions. Actively drawing attention to the specific earlier impression and explicitly noting its inaccuracy — a form of meta-communication about the interaction — can accelerate the update. Seeking a different context, where the initial impression-creating dynamic is absent, also helps reset the perceptual baseline.