I have watched some magnificent certainty from café awnings. Entire transport systems redesigned between two mouthfuls of toast. Economic policy settled by a man who cannot locate his loyalty card. The confidence arrives promptly. The supporting material appears to be on a replacement bus.
University of Glasgow research gives us a useful piece of the explanation, without giving anybody a licence to diagnose a stranger’s honesty by staring at their forehead. A paper first published on 21 October finds that movements around the eyes and eyebrows can change how confident a speaker is perceived to be.
The researchers used computer-generated faces answering yes-or-no questions. They varied upper-face movements while keeping spoken content and intonation constant. Forty-four participants took part: 22 White British native English speakers and 22 Chinese native Mandarin speakers. This was a controlled study of judgements about faces, not a census of every culture with an eyebrow.
Eye closure and raised brows were associated with greater perceived confidence; brow lowering and tightening around the eyes with greater doubt. The two groups showed a broadly shared pattern, with differences in the weight given to particular cues. That is interesting. It suggests the upper face is doing conversational work even when the words and the way they are spoken stay the same.
It does not establish that the confident-looking speaker knows the answer. Nor does it turn blinking into a lie detector, a job qualification or an emergency substitute for reading the document. The study measured how confident people appeared. It did not award their claims a certificate of correctness.
I find this reassuringly human. We are trying to work out what somebody means, and a face supplies clues. That is not foolish. The trouble begins when a clue about presentation gets promoted to head of evidence. You can raise your eyebrows as high as the café canopy; the arithmetic beneath them still needs to add up.
The study’s 44 participants and two language groups also put a sensible boundary round the conclusion. A useful experimental finding is not proof that every audience, setting and speaker will work identically. There is no need to make it carry a global consultancy on its back.
So I shall continue watching the face, listening to the words and checking what supports them. A well-timed blink may make an answer look assured. If the answer is that the missing train will arrive in five minutes, I would still like to see the timetable.
