HEALTH ASYMMETRY IN THE FACE 227
display of health, there is a complex interaction between health status, physical
attractiveness, social stereotypes, inferences regarding sexual activity, and a host
of other factors. Importantly, the distinction between perceived health and real
health can be blurred by attractiveness status.
In the present study, we were concerned with the left-right organisation of
the appearance of health in the face, on the assumption that there is a
neuroanatom ical underpinning in the brain of the observer for this type of
assessment. First we analysed the health judgements across all our observers,
and then we compared these judgements to previously obtained attractiveness
judgements (Zaidel et al., 1995) of the same set of faces.
METHOD
Participants
A total of 24 right-handed undergraduate students, 12 females and 12 males,
enrolled in introductory psychology courses volunteered to participate in the
experiment in exchange for partial course credit.
Materials
A total of 38 pairs of faces (21 women, 17 men) consisting Ofleft-Ieft and right-
right composites from a previous study were the stimuli (Chen et al., 1997;
Zaidel et al., 1995). Each composite was created on a Macintosh computer from
a head-on, symmetrically lit face, by aligning each facial half with its own
mirror image, hence left-left and right-right.
PROCEDURE
On a Macintosh computer screen, a pair of left-left and right-right faces (of the
same individual) was presented side-by-side for 10 seconds. There were 38 such
trials and women’s and men’s faces were intermixed within the series of trials.
The laterality of a facial composite on the computer screen was counterbalanced
within each face sex. The task for each participant was to decide which member
of the pair appeared healthier or if there was no difference between the two
faces.
RESULTS
ANOVA of health judgements
The frequency of selecting the left-left, right-right, or “same” (i.e., no
difference) was determined for each of the participants, and the mean was
entered into a repeated-measures ANOVA. (The responses of the female and
male viewers did not appear to be significantly different.) The within-subject