Alex Todorov

We may be taught not to judge a book by its cover, but when we see a new face, our brains decide whether a person is attractive and trustworthy within a tenth of a second. "We decide very quickly whether a person possesses many of the traits we feel are important, such as likeability and competence, even though we have not exchanged a single word with them", said Princeton University psychologist Alex Todorov. These snap judgments predict all kinds of important decisions. For example, politicians who simply look more competent are more likely to win elections. Yet the character judgments we make from faces are as inaccurate as they are irresistible. 
How do we form first impressions and why do we rely so much on them - I asked these questions Alexander Todorov, one of the world’s leading researchers on the subject.

Alex, how do we form first impressions?
People use all kinds of appearance cues to make snap judgments of others: facial appearance, emotional expressions and any other non-verbal behaviors, clothing, and, of course, verbal behaviors.

How much time do we usually need to make up our minds about others? 
These judgments based on appearance cues are made very rapidly and from minimal information. In my lab, we have shown that presenting a face of a stranger, somebody you have never seen before, for less than 100 milliseconds (that is, less than 1/10th of a second), is sufficient for people to decide whether the stranger could be trusted. There is no evidence that these judgments are accurate, but we cannot help it. We effortlessly make these judgments.

A pair of faces that show the variations in appearance that shape our impressions of trustworthiness.

What do we pay most attention when we form first impressions - eyes, smile, facial features....?
It is a complicated question. Typically, we describe face perception as holistic. That is, the sum of the parts provides more information than any of the parts. For example, smiling faces are perceived as friendlier. But a smile can make a face look sinister in the proper configuration of other facial features. To see how, draw a schematic face with the eyebrows pointing like this V and add a smile.

What people think could read from the faces of others?
As I said before, there is no good evidence that judgments from still images of faces are accurate. In real social interactions, we get much more information from dynamic verbal and non-verbal behaviors and the specific context of the interactions. This still leaves room for errors in impressions, but some of these may have a kernel of truth. People make all kinds of inferences from facial appearance like how intelligent the person is, how trustworthy and dominant, etc. It very much depends on the social context. If we care about a specific trait (e.g., extroversion), we’ll use appearance cues to make inferences about this trait.

Which facial features are perceived as credible and trustworthy?
It is not about the faces. It is about the people, but unfortunately the first access of information we have is the people’s faces. Typically, feminine faces and faces that appear to be happy are perceived as more trustworthy.

Are first impressions always accurate? Could we rely on them when making a judgement?  
First impressions can be highly misleading. If your decision has consequences, you better find other information to back up this decision. For example, if you are deciding whether to hire someone, letters of reference will provide much more valuable information then your impression. Letters of reference summarize relevant observations over a long-term time frame. Your impression captures a short glimpse of the person that can be distorted by multiple factors. For example, some people are naturally shy and would not perform that well in an interview setting. Yet, a shy but brilliant person may be what you really need for your job. 

Why then we pay so much attention to first impressions?
First impressions feel like perceptions not inferences. We believe what we see. But what we “see” in others is constrained by many factors, including our own biases.

How first impressions affect our everyday decisions?
There are many situations, in which we have brief interactions with strangers. In these situations, our decisions are likely to be guided by our first impressions. There are many experiments showing effects of first impressions on various choices. For example, people are more likely to invest money in others when these others appear to have trustworthy faces. This is even the case when there is information on the return of the investment. Of course, this is in the context of one-shot experimental games, where reputation doesn’t play a large role. In real life long-term economic transactions, reputation is what really matters.

Has the way we form first impressions changed over the years?
What has changed is that there is more and more personal information online and we form impressions from this information. This may have unfortunate consequences in some contexts. You would not want your potential employer to see a picture of you drunk with your buddies.

How could we learn to create a positive first impression? 
There are different impressions for different occasions. Being at a party with a goal of having fun is different from being at a job interview with a goal to get a job. You need to know what is expected from you and be prepared.

Have you ever formed a wrong first impression about someone?
I try not to rely on my first impressions. In some contexts, these may be more or less irrelevant or inconsequential. It does not matter whether your impression is accurate or not if is an impression of a stranger you met on the street and you are never going to meet again. Of course, if you are in a dangerous neighborhood late at night, that may change the consequences and your first impression may be the best information you got. When I make decisions whether to admit a graduate student to work with me or to hire a manager of my lab, I weigh heavily in my decisions their letters of recommendation, as well as any evidence for their academic competence like grades, test scores, and prior research experience. There is a lot of work in psychology showing that unstructured job interviews provide very little useful information if any. Yet, people overweight their impressions from such interviews in their decisions.

On what are your working currently? 
In my lab, we have multiple ongoing projects at any moment of time. We build computer models of first impressions that can help us visualize what changes in the face lead to different impressions. We conduct functional neuroimaging experiments to study how faces are represented in the brain. We do behavioral experiments to figure our how we change our minds about others over time as we learn more information about them. Right now, I am working on a paper showing that different images of the same person can lead to very different impressions and that these impressions can bias subsequent decisions.