You trust a face like yours

Abstract The appraisal of trustworthiness from facial appearance a stranger is critical for successful social interaction. Although self-resemblance considered significant factor affecting the perception trustworthiness, research yet to be conducted on whether this theory applicable natural unfamiliar faces in real life. We examined aspect by using state-of-the-art deep convolutional neural network face recognition measure similarity large sample people with evaluators. found that more they resembled rater, trustworthy were evaluated if same sex as rater. Contrarily, when was opposite sex, did not affect ratings. These results demonstrate an important our judgments especially same-sex