Once we know something, it becomes hard to imagine what it feels like not to know it.
More quotesThe participants preferred photographs with dilated pupils and could not report why. Their explanations arrived anyway: expression, warmth, character.
That gap limits what an interview can tell us about visual preference. People describe cues they noticed and reasons available to consciousness. Lighting, gaze, spacing or another small alteration may influence the response without appearing in the account.
Testing the finished image can reveal the effect. It cannot settle whether using it is acceptable. “People cannot tell what changed” should increase ethical scrutiny rather than trigger a designer’s little victory dance.
The study involved a specific set of photographs and male participants. Attraction outside that setting involves far more than pupil size. There is no universal recipe here, only evidence that an unnamed visual difference can influence judgement.
Consent matters because retouching changes how strangers may infer warmth, health or interest from a real person. The subject should understand the intended alteration, especially where an image sells a relationship, service or political claim.
In a portrait test, alter one detail and record preference, attention and what viewers understood. Keep the unedited image beside the result. If the change clarifies a truthful expression, there may be a defensible use. If it manufactures a quality the subject cannot support, the explanation offered by viewers makes the deception harder to see.
The test sheet retains two columns: chosen image and stated reason. They do not always agree. A later round repeats the comparison with fresh images and participants, leaving the same two columns in place.