Knowledge is power, yet new research suggests that a person's appearance alone can trump knowledge. First impressions are so powerful that they can override what we are told about people. A new study found that even when told whether a person was gay or straight, participants generally identified the person's sexual orientation based on how they looked - even if it contradicted the facts presented to them.

"We judge books by their covers, and we can't help but do it," says Nicholas Rule of the University of Toronto. "With effort, we can overcome this to some extent, but we are continually tasked with needing to correct ourselves." The less time we have to make our judgments, the more likely we are to go with our gut, even over fact, he says.

A series of recent studies, presented at the Society for Personality and Social Psychology (SPSP) annual conference in Austin, shows that appearance shapes everything from whether we ultimately end up liking someone to our assessment of their sexual orientation or trustworthiness. And researchers say that whether a first impression occurs online versus in person is important. While we may be able to size up someone's personality from a Facebook photo, it will often be more negative impression than one formed face-to-face.

Appearance trumps fact

"As soon as one sees another person, an impression is formed," Rule says. "This happens so quickly - just a small fraction of a second - that what we see can sometimes dominate what we know."

In the study on first impressions of sexual orientation, Rule and colleagues showed 100 participants photos of 20 men, identifying them either as gay or straight. The photos had been previously coded based on a consensus opinion on whether the men "looked" gay or straight, which accurately matched to their real-life sexual orientations. The researchers then tested participants' recall of the men's sexual orientations several times to ensure perfect memorization.

After this learning phase, the researchers then showed participants the faces again, varying the amount of time they had to categorize the men's sexual orientations. The less time they had to categorize the faces, the more likely the participants were to categorize the men according to whether they looked gay or straight rather than what they had been told about their sexuality. With more time, however, the participants reverted to what they had learned about the men's sexuality.

"Hence, they seemed to judge by appearance when they were forced to make their judgments quickly," Rule says. "When they were allowed more time, though, they judged according to what they knew about the individuals."

Interestingly, the researchers labeled half the faces with their actual sexual orientation and half with their opposite orientation. They did this to "teach the participants to learn information that was opposite to their perceptions," Rule says. "It was important for us to establish a conflict between perception - how the face looked - and memory - what they knew about the man's sexual orientation."

Rule uses the example of Ricky Martin who for years denied he was gay before finally coming out. In the 1990s, people might see Martin and think "oh, that's a gay guy," he says, "but then you'd recognize that it was Ricky Martin and think 'oh, wait, that's Ricky Martin - he told Barbara Walters that he was straight.' So there's a corrective process there: First impressions continue to assert themselves long after you know relevant information about a person."

Rule presented this study at the SPSP conference, along with a related new study that looked at how people categorized faces as trustworthy or not. In that study, facial appearance was a stronger predictor of whether people viewed someone as trustworthy than descriptive information provided, even even it conflicted.

"Together, these studies help to illustrate the often inescapable nature of how we form impressions of other people based on their appearance," Rule says. "Not only should people not assume that others will be able to overcome aspects of their appearance when evaluating them, but also those of us on the other end should be actively working to consider that our impressions of others are biased."

The virtual bias

"If you want to make a good impression, it is critical that it is done in person," says Jeremy Biesanz of the University of British Columbia. That is the bottom line of his new research that looks at the difference in how we form impressions in person, versus online, by video, or by just watching.

In three studies, Biesanz and colleagues compared the accuracy and bias of impressions formed under different circumstances. The first study analyzed a series of experiments involving more than 1,000 participants who met each other through either a 3-minute speed-dating style interview or by watching a video of the person they are evaluating. They also evaluated their own personalities.

"What we observe here is that the accuracy of impressions is the same when you meet someone face to face or simply watch a video of them," Biesanz, says. "However, impressions are much more negative when you form impressions more passively through watching videotapes." So while people could accurately attribute certain personality traits - for example, extroverted, arrogant, sociable - to others both in person or by video, the magnitude of the positive attributes was lower and of negative attributes was higher via video.

The researchers found similar results in two other studies - one that used the same set-up to compare in-person impressions to those obtained through looking Facebook photos, and another that compared in-person meetings to simply watching someone as a passive observer. In all cases, the passive means of making impressions were as accurate as the active ones. "However, there is an extremely large difference in the positivity of impressions," he says "More passive impressions are substantially more negative."

So while Yogi Berra's mantra of "You can observe a lot by watching" may be true, Biesanz says, mere observation comes at the cost of substantial bias.

From romance to likeability

How we create first impressions is also important in the context of finding a romantic partner. Recent research suggests that whether someone meets a potential mate online versus in person can dramatically change their judging process.

"People are more likely to use abstract information to make their evaluations in hypothetical than in live impression formation contexts," says Paul Eastwick of the University of Texas, Austin, who is presenting results of his studies on gender differences in different romantic contexts. When men and women evaluate potential partners in person versus online, typical gender differences in ideal preferences disappear.

In general, men say they care about attractiveness in a partner more than women, and women say they care about earning prospects in a partner more than men. "But our meta-analysis reveals that men and women do not show these sex differences when they evaluate others in a face-to-face context," Eastwick says. "That is, attractiveness inspires men's and women's romantic evaluations to the same extent, and earning prospects inspires men's and women's romantic evaluations to the same extent."

The research suggests that in live face-to-face settings, people rely more on their gut-level evaluations of another person. "They focus on how that person makes them feel," Eastwick says. "It is very hard to get a sense of this information when simply viewing a profile. This disconnect can cause confusion and distress in the online dating realm, as potential partners that seem terrific 'on paper' prove to be disappointing after a face-to-face interaction."

Beyond the online dating realm, Vivian Zayas of Cornell University, Gül Günaydin of Middle East Technical University, and colleague have found that viewing a photograph can be a a good predictor of how you will judge someone in person. "Despite the well-known idiom to 'not judge a book by its cover,' the present research shows that such judgments about the cover are good proxies for judgments about the book - even after reading it," says Zayas, who is chairing a session at the SPSP conference called "When to Judge a Book by Its Cover: Timing, Context, and Individual Differences in First Impressions."

Zayas' new research shows that initial impressions based on viewing a single photograph accurately predict how a person will feel about the other person in a live interaction that takes place more than 1 month later. "Moreover, participants' initial judgments based on the photograph colored personality judgments following the interaction," Zayas says. "The results showed that initial liking judgments based on a photograph remained unchanged even after obtaining more information about a person via an actual live interaction."