How fast do people form a first impression of your face? What the research says

Very fast. In Willis and Todorov's 2006 experiments, judgments of trustworthiness, competence, likeability, attractiveness and aggressiveness made after a 100-millisecond look correlated highly with judgments made with no time limit, and longer exposure mostly raised confidence rather than changing the verdict. The impression also depends on the particular photo: in Todorov and Porter's 2014 study, different photos of the same person produced impressions that varied as much as, or more than, differences between people. What you can control is the photo: expression, distance and the moment you pick. These are quick impressions, not accurate verdicts on character.

By escapethematrix.app · Updated October 15, 2026

You do not get to decide when someone starts judging your face. Research on first impressions suggests the judgment has usually formed before you have said a word, and that it is built from the specific photo or moment they saw, not from a fixed "you". This guide summarizes what the key studies measured, what they did not show, and which parts of a photo you can control.

escapethematrix.app publishes Presence, an iPhone appearance app, disclosed near the end. Every claim below is taken from the abstract of the cited study.

The short version

1. A tenth of a second is enough

Willis and Todorov ran five experiments, each focused on one trait judgment, and varied how long unfamiliar faces were shown. Judgments made after a 100-millisecond exposure correlated highly with judgments made with no time constraint. For attractiveness, likeability, trustworthiness, competence and aggressiveness, giving people longer did not significantly increase that correlation (Willis & Todorov, 2006).

Longer looks did change two things. From 100 to 500 ms, judgments became more negative, response times fell and confidence rose. From 500 to 1,000 ms, judgments and response times did not change significantly (with one exception), but confidence still rose for some judgments, which the authors read as extra time boosting confidence rather than changing the verdict. Longer exposure also led to more differentiated impressions of the person.

What this does not say: that the impression is right. It says that a snap verdict exists, and that people hold it more firmly with time.

2. What the snap judgment is made of

Oosterhof and Todorov analyzed trait judgments of emotionally neutral faces and found two dimensions were enough to describe them: valence, approximated by trustworthiness, and dominance. Their modelling found that valence judgments are more sensitive to features resembling expressions that signal whether a person should be avoided or approached, while dominance judgments track features signaling physical strength or weakness. The authors describe face evaluation as an overgeneralization of adaptive mechanisms, which can account for judgments that are rapid "yet not necessarily accurate" (Oosterhof & Todorov, 2008).

Practical reading: people respond to cues that look like friendly or unfriendly expressions, even on a face that is trying to be neutral. A resting expression that reads as tense or flat in a still frame can be taken as "avoid" by a stranger who knows nothing else about you.

3. Your photo changes the impression

Most first-impression studies used one image per person and treated differences as stable traits. Todorov and Porter tested that assumption. Images of the same individual led to different impressions, with variance within a person comparable to or exceeding variance between people for a range of social judgments. Preferences for images also shifted with context, such as choosing a photo for online dating versus a political campaign, and the biases created by that selection were evident even after a 40-ms presentation (Todorov & Porter, 2014).

This is the most useful finding for anyone choosing a profile picture. The photo is not a neutral record of your face. It is one sample, and it can be a better or worse one for the impression you want.

4. What you can change in a photo

Expression. Golle, Mast and Lobmaier morphed faces along attractiveness and happiness. Judgments of attractiveness were strongly influenced by smile intensity, and a happy expression could even compensate for relative unattractiveness (Golle et al., 2014). These were computer-morphed faces judged side by side, so it shows a direction, not a promise about your own photo.

Distance. Bryan, Perona and Adolphs found that photographs taken from within personal space drew lower investments in a trust game and lower ratings of trustworthiness, competence and attractiveness than photographs taken from farther away. The effect replicated across studies that controlled for image size, expression and lighting (Bryan et al., 2012). Our guide on why you look different in photos covers the distance numbers in detail.

Gaze and lighting. We did not find a study we could check that puts a number on either for photos, so we make no numeric claim. In the distance studies above, lighting was held constant, which tells you it is a separate variable, not that it does not matter. Soft, even light and a lens at eye level are sensible defaults, not proven effects.

What to do with this

1. Assume the first look is fast. Your profile photo, thumbnail or video still is read in a fraction of a second, so the image you pick matters more than the effort you put into the caption.
2. Take several photos and choose. Within-person variation is as large as between-person variation in the Todorov and Porter data, so choosing among several photos gives you a real chance to pick a better sample.
3. Choose for the context. The photo that suits a dating profile may not suit a professional page, and viewers pick differently in each setting.
4. Check your resting expression. The Oosterhof and Todorov model says trustworthiness judgments respond to features resembling approach or avoid expressions, and Golle and colleagues found smile intensity moved attractiveness judgments. So compare a neutral frame with a relaxed smile and see which reads better to you.
5. Shoot from farther back. About 1.5 m with the 2x lens, which our photo guide explains, keeps you well outside the close range where Bryan and colleagues found lower ratings.
6. Do not over-read the result. A quick impression is not a verdict on you, and the same face can score differently on a different day. If your sleep, grooming or posture varied between shots, see what sleep loss does to your face and the first impression checklist.

Where Presence fits (disclosure: our app)

Presence is an iPhone app published by escapethematrix.app. It analyzes a selfie on your phone and gives First Impression, Aura and Grooming scores, so you can compare shots taken the same way. Treat any score as feedback on one photo, not a measure of you, which is the same lesson as the research above.

Sources

1. Willis J, Todorov A. First impressions: making up your mind after a 100-ms exposure to a face. Psychological Science. 2006;17(7):592-598. PubMed
2. Oosterhof NN, Todorov A. The functional basis of face evaluation. PNAS. 2008;105(32):11087-11092. doi:10.1073/pnas.0805664105. PubMed
3. Todorov A, Porter JM. Misleading first impressions: different for different facial images of the same person. Psychological Science. 2014;25(7):1404-1417. doi:10.1177/0956797614532474. PubMed
4. Golle J, Mast FW, Lobmaier JS. Something to smile about: the interrelationship between attractiveness and emotional expression. Cognition & Emotion. 2014;28(2):298-310. doi:10.1080/02699931.2013.817383. PubMed
5. Bryan R, Perona P, Adolphs R. Perspective distortion from interpersonal distance is an implicit visual cue for social judgments of faces. PLOS ONE. 2012;7(9):e45301. PubMed

This guide explains perception research. It is not medical or psychological advice. If worry about your appearance is affecting daily life, talk to a doctor or mental health professional.

Frequently asked questions