University of Delaware A series of five studies investigated the young infant's ability to produce identifiable emotion expressions as defined in differential emotions theory. Trained judges applied emotion-specific criteria in selecting expression stimuli from videotape recordings of 54 1- to 9-month-old infants' responses to a variety of incentive events, ranging from playful interactions to the pain of inoculations. Four samples of untrained subjects from two different populations confirmed the social validity of infants' emotion expressions by reliably identifying expressions of interest, joy, surprise, sadness, anger, disgust, contempt, andjeaj;. Brief training resulted in significant increa'sesTin the accuracy of discriminatio n of infants' negative emotion expressions for low-accuracy subjects. Construct validity for the eight emotion expressions identified by untrained subjects and for a consistent pattern of facial responses to unanticipated pain was provided by expression identifications derived from an objective, theoretically structured, anatomically based facial movement coding system. There has been little systematic investigation of the infant's ability to encode or produce facial expressions of emotion, except for smiling and laughing (Emde, Gaensbauer, & Harmon, 1976; Spitz & Wolf, 1946; Sroufe & Wunsch, 1972). Investigations of the newborn (Emde et al., 1976; Peiper, 1963) and the born blind (Dumas, 1932; Eibl-Eibesfeldt, 1972; Goodenough, 1932; Thompson, 1941) have suggested that the expressions of startle, neonatal smiling, disgust, and distress are present at birth. Darwin's (1872, 1877) observational studies,