Case File 203 · Investigation Psychology

Are Microexpressions Reliable for Detecting Lies?

By Murder Book Editorial TeamPublished by Gungrounds, the studio making Murder BookPublished 3 October 2026Updated 3 October 2026

No, microexpressions are not a reliable way to detect lies. A microexpression is a very brief facial expression, lasting roughly 1/25 to 1/5 of a second, thought to flash when someone tries to conceal an emotion. The idea is real and has some laboratory support, but the leap from "a brief expression appeared" to "this person is lying" is not justified by the evidence. Studies find true microexpressions are rare, that they show up in truthful people too, and that spotting them does little to improve lie detection. A fleeting expression can tell you an emotion may be present. It cannot tell you why, and it is not proof of deception.

Key takeaways

Microexpressions are brief facial movements, around 1/25 to 1/5 of a second, linked to the theory that concealed feelings can leak out.
In a key frame-by-frame study, true microexpressions appeared in only about 2% of expressions and in a minority of participants.
A detailed critique argued that the assumptions behind microexpression theory, including that such expressions are frequent, uncontrollable and specific to lying, are not well supported.
Emotional leakage, when it occurs, signals an emotion, not its cause; honest people under stress leak too.
No single facial cue reliably separates lies from the truth.

Where does the idea of microexpressions come from?

The concept grew out of the work of psychologists Paul Ekman and Wallace Friesen, whose 1969 paper "Nonverbal leakage and clues to deception" first set out the idea that feelings a person tries to hide can escape through nonverbal behaviour. The face, they argued, is both the best place to send a false message and a place where true feelings can briefly break through. Over the following decades this grew into the popular image of the microexpression as a hidden tell, helped along by television and training products. The underlying theory of leakage, and its limits, is covered in What Is Behavioural Leakage?.

How often do microexpressions actually appear?

Rarely, according to one of the most careful tests of the idea. In a 2008 study published in Psychological Science, Porter and ten Brinke asked participants to view emotional images and respond with either genuine or deceptive expressions: faking a feeling, neutralising it or masking it. The researchers then analysed 697 expressions frame by frame, a total of 104,550 frames.

Masked emotions did produce more inconsistent expressions and a higher blink rate than genuine ones, so some leakage was real. But true microexpressions, lasting between 1/25 and 1/5 of a second, were shown by only about 22% of participants and in just 2% of all expressions, and only ever in the upper or lower half of the face rather than the whole face. Untrained observers did only slightly better than chance at spotting the deception. In other words, the classic microexpression is uncommon, partial and easy to miss, even in a controlled setting designed to produce it.

Is microexpression theory sound?

A pointed critique came in 2018, when Judee Burgoon published "Microexpressions Are Not the Best Way to Catch a Liar" in Frontiers in Psychology. She set out the chain of assumptions the theory needs in order to work: that deception reliably produces a negative emotion, that this emotion produces an outward expression including microexpressions, that such expressions cannot be controlled, that they are valid and specific indicators of deceit, and that they happen often enough to be detected.

Burgoon argued that each link is questionable. Deception does not always feel bad; it can produce relief or even pleasure, and some liars feel aroused without any clear emotion at all. Expressions can be masked or managed. And, as the 2008 study showed, full microexpressions are infrequent. A competing opinion piece by Matsumoto and Hwang defended a role for microexpressions in reading intent, so the debate continues, but the strong popular claim, that a fleeting expression exposes a lie, outruns the evidence.

Does any facial leakage mean anything?

Sometimes faces do leak, but in a probabilistic way, not as a switch that flips when someone lies. In a 2012 study titled "Darwin the Detective," ten Brinke, Porter and Baker examined videotaped pleas from 52 people appealing for the return of a missing relative, half of whom were later convicted of murdering that person. Muscles associated with genuine grief were contracted more often in the faces of honest pleaders than deceptive ones, and deceivers sometimes showed expressions inconsistent with sincere distress.

That is a measurable difference across a group, studied with specialised coding, not a tell any observer could read in the moment. It describes tendencies, with plenty of overlap between the groups, and it still cannot classify a single face with confidence. The lesson is the same as for gaze and gesture: patterns exist, but individual certainty does not.

Why is a brief expression not proof of lying?

Because an expression reveals, at most, that an emotion may be present, and emotions have many causes. A flash of fear during questioning might come from guilt, but it might equally come from the stress of being suspected, from grief about the event, or from the simple fear of not being believed. That last possibility is the heart of the Othello error, discussed in Is Nervousness a Sign That Someone Is Lying?: an innocent person's distress can look exactly like a guilty person's. A microexpression does not come with a label explaining which emotion it is or where it came from.

Popular belief	What research suggests
A microexpression exposes a hidden lie	At most it suggests an emotion; it does not reveal the cause
Microexpressions happen constantly	True microexpressions were rare in a detailed frame-by-frame study
They cannot be controlled or masked	People can neutralise and mask expressions; leakage is partial
Watching for them makes you a strong lie detector	Observers performed only slightly above chance

What is more useful than hunting for microexpressions?

Content and corroboration. Because no facial cue classifies a lie reliably, the stronger approach is to examine what a person says, how much of it can be checked against independent evidence, and whether the account stays consistent under careful questioning. That is why the research consensus, set out in Why There Is No Single Reliable Sign of Deception, warns against reading guilt off a face. For the wider picture, including the related myths about posture and gesture, see Can Body Language Tell You If Someone Is Lying?, and for how well anyone can judge lies at all, see Can You Actually Tell When Someone Is Lying?.

Does Murder Book use microexpressions as a lie detector?

Murder Book: The Art of Interrogation is an upcoming narrative interrogation game from Gungrounds, an independent studio in Zagreb, Croatia. Its press kit names Emotional Leakage among the behaviours players examine, and development screenshots show casebook entries such as Acting Emotion, which links a failure to sustain a performed emotion to cognitive load theory. In keeping with the research, these work best as prompts to look closer, not as verdicts.

A flicker on a suspect's face is a reason to ask another question and check the story, not a confession. Cases are designed for focused 5–10-minute sessions, so the emphasis is on weighing an account rather than freezing a single frame. Murder Book is in development for PC (Steam) and mobile, with release currently planned for the end of Q2 2027.