Composite facial expressions were prepared by aligning the top half of one expression
(e.g., anger) with the bottom half of another (e.g., happiness). Experiment 1 shows
that participants are slower to identify the expression in either half of these composite
images relative to a "noncomposite" control condition in which the 2 halves are misaligned.
This parallels the composite effect for facial identity (A. W. Young, D. Hellawell,
& D. C. Hay, 1987), and like its identity counterpart, the effect is disrupted by
inverting the stimuli (Experiment 2). Experiment 3 shows that no composite effect
is found when the top and bottom sections contain different models' faces posing the
same expression; this serves to exclude many nonconfigural interpretations of the
composite effect (e.g., that composites are more "attention-grabbing" than noncomposites).
Finally, Experiment 4 demonstrates that the composite effects for identity and expression
operate independently of one another.