Facial Emotion Recognition Task

The Facial Emotion Recognition task measures how accurately people identify the six basic emotions (happiness, sadness, anger, fear, disgust and surprise) in photographs of faces. On each trial a face appears for 5 seconds and the participant clicks or taps the word that best describes its expression. The six buttons stay on screen after the face disappears, and there is no time pressure for answering. Accuracy for each emotion is the main outcome; the response time is recorded as well.

The design follows the Ekman 60 Faces Test from the FEEST (Young et al., 2002): 60 photographs, ten for each emotion, each shown for 5 seconds, with no time limit for the answer (procedure as described by Khosdelazad et al., 2020). The six categories go back to Ekman and Friesen's (1971) cross-cultural work on basic emotions.

This task ships without face photographs. No validated face set allows its photographs to be passed on through a public library, so each researcher adds a set they are licensed to use, for example FACES, the Radboud Faces Database, KDEF or Ekman and Friesen's Pictures of Facial Affect. Until a photo is added, its slot shows a grey placeholder card with the expected file name, and the data mark that trial with placeholder = true. The Documentation tab explains how to add the photos and summarises each set's terms. Adding photos needs the lab.js builder, which is available on the Pro plan for tasks copied from the library.

The task has 6 practice trials (one per emotion, from a separate poser, without correctness feedback) and 60 main trials in random order. A fixation cross and a variable blank interval (500–1,500 ms) come before each face.