This article reports on three experiments on the controversial topic o
f ''context effects'' in the judgment of emotion from the face. In Exp
eriment 1 (N = 169) subjects were shown either a happy, sad, or angry
anchor face as context followed by a target slide of a ''neutral'' fac
e. In Experiment 2 (N = 119) subjects were shown an anchor of a happy
or angry face as context and a sad face as target. In Experiment 3 (N
= 180) subjects were shown an anchor of a happy, sad, or surprised fac
e as context and an angry face as target. All experiments used facial
expressions from Ekman and Friesen's Pictures of Facial Affect (1976).
Dependent measures included intensity ratings of pleasure and arousal
dimensions (Mehrabian & Russell, 1974); a judgment of the intensity o
f sir specific emotions expressed (happy, sad, angry, afraid, disguste
d, and interested); and categorical judgments of emotions. Significant
context effects were observed for the neutral target and, with smalle
r effects, for the angry and sad targets on dimensional and intensity
ratings. The magnitude of the context effect depended on both the targ
et and anchor facial expressions. Greater categorical agreement of emo
tion was obtained for the target when another face was provided as a c
ontext than when the target face was shown alone. These results provid
e an independent replication and extension of recent research (Russell
, 1991; Russell and Fehr, 1987) on the relativity of facial affect jud
gment.