Content uploaded by Lori A. Roggman
Author content
All content in this area was uploaded by Lori A. Roggman
Content may be subject to copyright.
Developmental Psychology Copyright 1990 by the American Psycho/ogical Association, Inc.
1990, Vol. 26, No. 1,153--159 0012-1649/90/$00,75
Infants' Differential Social Responses to
Attractive and Unattractive Faces
Judith H. Langlois, Lori A. Roggman, and Loretta A. Ricser-Danner
University of Texas at Austin
Two studies were conducted to examine infants' social responses to attractive and unattractive faces.
In Study I, 60 12-month-olds interacted with a stranger who wore a professionally constructed
attractive or unattractive mask. The infants showed more positive affective tone, less withdrawal,
and more play involvement with the stranger in the attractive condition. In Study 2, 43 12-month-
olds played with an attractive and an unattractive doll. The infants played significantly longer with
the attractive doll. These results extend and amplify earlier findings showing that young infants
exhibit visual preferences for attractive over unattractive faces. Both visual and behavioral prefer-
ences for attractiveness are evidently exhibited much earlier in life than was previously supposed.
Recently, we reported data demonstrating that young infants
both discriminate among faces on the basis of adult-judged
physical attractiveness and exhibit visual preferences for attrac-
tive, compared with unattractive, female faces (Langlois et al.,
1987). These results, which have been found by others as well
(Samuels & Ewy, 1985; Shapiro, Eppler, Haith, & Reis, 1987),
were surprising and important because most researchers in this
area have assumed that standards of attractiveness were learned
gradually through cultural transmission and were not evident
until 3 or 4 years of age (e.g., Adams & Crane, 1980; Berscheid
& Walster, 1974; Dion, 1973; Dion & Berscheid, 1974; Langlois
& Stephan, 1981), or even later (Cavior & Lombardi, 1973;
Cross & Cross, 1971). In contrast to this assumption, however,
these visual preferences of young infants suggest that prefer-
ences for attractive faces are evident much earlier than was pre-
viously supposed and may even be innate.
This demonstration of discrimination and visual preference
for attractive faces by infants was a significant first step in chal-
lenging the prevailing assumption that such discriminations
and preferences were only gradually learned through cultural
transmission. It is not clear, howeve~ whether these early ap-
This research was supported by Grant HD 21332 from the National
Institute of Child Health and Human Development to Judith H. Lang-
lois. Study 1 was presented at the International Conference on Infant
Studies, Washington, DC, 1988.
We thank our strange~ Annabelle Arteaga, for her willingness to go
far above and beyond the normal call of duty in Study 1. We also thank
Paul Smith for constructing the masks and Leslie Lawson for her make-
up skills in Study 1. We are grateful to Jean Ritter for obtaining the
attractiveness ratings of the masks in Study 1 and to Rita Casey for
drawing the faces of the infants in Study 2. Finally, we thank all of the
experimenters, camera old-raters, infants, and parents who so gra-
ciously gave up their Saturdays to participate in the studies reported
here.
Lori A. Roggman is now at the University of Arkansas, and Loretta
A. Rieser-Danner is now at Villanova University.
Correspondence concerning this article should be addressed to Judith
H. Langlois, Department of Psychology, The University of Texas, Mezes
Hall 330, Austin, Texas 78712.
pearing visual behaviors of infants represent preferences for at-
tractiveness that are analogous to those seen in adults and older
children (e.g., Dion, 1973, 1977; Dion & Berscheid, 1974;
Langlois & Downs, 1979; Langlois & Stephan, 1977). Before
accepting that differential infant visual behavior toward attrac-
tive and unattractive faces reflects infant preferences for attrac-
tiveness that are similar to preferences demonstrated with
adults and older children, two assumptions must be accepted.
The first is that when infants consistently look longer at one
type of stimuli than another; they are indeed demonstrating a
preference. Although such an assumption is commonly made in
the infant perception literature (e.g., Banks & Salapatek, 1983;
Fantz, 1965; Fantz & Fagan, 1975), the validity of the assump-
tion has been questioned by some (e.g., Haith, 1980), in part,
because discrimination and preference cannot be separated in
the paired presentation technique used in these studies.
The second assumption required is that visual preferences
are functionally equivalent to the preferences older children and
adults show for attractive individuals. Infant visual behaviors
toward attractive and unattractive faces are in' fact consistent
with the results of studies of older children and adults in which
differential looking time has been used as a measure of prefer-
ence for attractiveness: Like infants, both adults and older chil-
dren
look
longer at faces
judged as
attractive compared with
faces judged as unattractive (e.g., Dion, 1977; Hildebrandt &
Carman, 1985; Hildebrandt & Fitzgerald, 1978, 1981; Kleck
& Rubenstcin, 1975; Power, Hildebrandt, & Fitzgerald, 1982).
Thus, the responses of young infants are, at least superficially,
similar to those of older children and adults. Hewers, in the
literature with older children and adults, preferences are more
typically indicated by verbal statements of liking (e.g., Ber-
scheid, Dion, Walster, & Walster, 1971; Dion, 1973; Kleck,
Richardson, & Ronald, 1974; Langlois & Stephan, 1977) and
behavioral observations or ratings of social responsiveness, so-
cial involvement, and social rejection (e.g., Anderson & Bern,
1981; Elder, Van Nguyen, & Caspi, 1985; Reis, Nezlek, &
Wheeler, 1980; Reis et al., 1982; Smith, 1985; Snyder; Tanke,
& Berscheid, 1977).
The purpose of the studies reported here was, first, to test
153
154 J. LANGLOIS, L. ROC~MAN, AND L. RIESER-DANNER
infant preferences for attractive faces by using dependent vari-
ables that are more directly tied to variables previously docu-
mented in the literature than are looking time and visual dis-
crimination. The infant behaviors examined here were selected
to be conceptually similar to behaviors indexing preferences
and social responsiveness exhibited differentially toward attrac-
tive and unattractive targets by adults and older children. The
second purpose was to test infant preferences for attractive faces
in social situations to determine if they generalize beyond arti-
ficial testing situations with flat, two-dimensional photographs
as stimuli. Two studies were designed to examine the behavioral
responses of infants. In Study 1, infants interacted with a
stranger in the presence of their mother. Attractiveness of the
stranger was manipulated by having thin, lifelike, latex theater
masks constructed by a professional mask maker to be either
attractive or unattractive. In Study 2, infants played with attrac-
tive and unattractive dolls whose faces were constructed by trac-
ing facial features from photographs of infants who had been
rated by adults as more and less attractive.
Study 1
Method
Subjects. Eighty-nine 12-month-old infants were recruited from the
infant subject pool maintained by the Children's Research Laboratory
at the University of Texas. Twelve-month-olds rather than younger in-
fants were selected because this is about the earliest age at which social
behaviors such as play, approach, and avoidance of strangers can be
observed. Twenty-nine subjects were eliminated from the data analyses:
4 because of equipment failure, 5 who showed intense distress and did
not complete the session, 3 who spent one or more episodes out of view
of the videocamera~ 4 who were accompanied by adults other than their
primary caregivers, and 13 who were observed on 2 days when the mask
was misapplied and did not adhere properly to the stranger's face. Thus,
a final sample of 60 infants was obtained, 26 in the attractive condition
( 18 gifts, 8 boys) and 34 in the unattractive condition ( 15 girls, 19 boys).
All of the infants were full term, healthy, and from middle-class families.
Attractiveness manipulation. Because there might be behavioral
differences between attractive and unattractive individuals asked to in-
teract with infants (e.g., Dion & Stein, 1978; Hildebrandt & Fitzgerald,
1978; Langiois & Casey, 1984; Langiois & Downs, 1979; Snyder et al.,
1977), a single female stranger interacted with all oftbe infants, using a
strict, rehearsed script to standardize her behavior. The attractiveness
of the strenger was manipulated through the use of professional theater
masks. Both attractive and unattractive versions of the mask were worn
by the stranger to control for any effects of wearing a mask per se.
A professional mask maker created the masks using the stranger's
face as a basic mold (e.g., Buchman, 1971). The female stranger was an
attractive woman, and a cast of her face served as the attractive mold.
Alterations to this basic mold to create the unattractive version of her
face were based on previous research (Cunningham, 1986) that identi-
fied measurements of facial features that predict ratings of attractive-
ness. Specifically, the eyes were narrowed and moved closer together, the
eyebrows were lowered, and the nose was lengthened. The unattractive
version of the mask was designed so that the feature sizes were well
within the normal range of attractiveness: Our goal was to make the
stranger appear unattractive, but not to appear abnormal or deformed
in any way. The masks were made of very thin latex so that the stranger's
face appeared quite real and lifelike. Thus, she was able to talk and smile
without appearing strange. Three versions of each basic mask were
made. The stranger, wearing each of the masks, was rated for attractive-
ness from the videotapes in each condition. A total of 25 introductory
psychology students judged the attractiveness of the stranger using a Lik-
err-type scale ranging from not at all attractive (1) to very attractive (5).
The reliability of the attractiveness ratings was .99, as estimated by co-
etficient alpha. The mean rating of the stranger in the attractive and
unattractive conditions was 3.16 and 1.16, respectively,/(24) = 2.33,
p < .05. Thus, the attractiveness manipulation was successful, and com-
ments from the raters revealed that none of the students thought the
stranger was deformed or abnormal looking.
The stranger was blind at all times to which mask she was wearing,
so that knowledge of the attractiveness condition of the session could
not bias her behavior toward the infant. The stranger could not differ-
entiate between the masks when she was wearing them because only the
external appearance of the mask changed. The interior of each mask
was designed to he identical so that there were no differential tactile
cues. The stranger was never told which mask she was wearing, and all
glass and other shiny surfaces in the building were occluded. Outside of
her interaction in the experimental session, the stranger was never left
to interact with individuals who were not part of the research team. The
choice of the mask to he worn on any given testing day was random
within the availability of masks (the masks were so thin that they tore
after one or two applications).
To ensure that the stranger's behavior was in fact equivalent across
attractiveness conditions, a random sample of tapes was selected and
the length of the stranger's conversation with the infant was timed. Con-
versation length was selected as the most sensitive indication of differ-
ences in the stranger's behavior because conversation length was the
only stranger behavior that was always available in the tapes: The face
of the stranger was not visible in two of the three episodes, so facial
affect could not be coded in these episodes, and the only words that
varied in the stra%~er's script were references to the specific toys with
which individual infants played. There were no significant differences
in the length of time the stranger spent talking to the infants in the two
attractiveness conditions, F- 3.10, ns.
Procedure. Application of the mask was a very tedious process. One
graduate student experimenter was trained by the professional mask
maker in application and make-up procedures. The application proce-
dure itself took approximately 2-3 hr, limiting the number of days per
week sessions could be conducted because of the restrictions imposed
by the schedules of all the necessary personnel. Generally, infant sub-
jects were scheduled back-to-back across a 4 or 5-hr time period on
Saturdays, during which time the stranger wore the same mask.
On their arrival at the laboratory, the mothers and infants were led
by a female experimenter to a waiting area where the study was ex-
plained in detail. Questions were answered, and informed consent was
obtained. The mothers were informed about the nature of the masks
but were asked not to comment on or respond to the masks in any ob-
servable way so that they would not inadvertently provide the stranger
with information about which mask she was wearing, and so that the
infants would not respond to the stranger on the basis of reactions of
their mothers.
Once mothers and babies were ready, they were loci to a playroom
furnished with two chairs, a small table with ma~a_~ines, and a set of
age-appropriate toys spread out in the middle of the floor. The mothers
were given instructions to spend approximately 1 rain (or longer; if
needed) playing with their infants with the toys (Episode 1). Mothers
were told to move to a pre-assigned chair when they felt that their infant
was comfortable. They were encouraged to respond to the bids of their
child but not to initiate interaction after moving to the chair. After giv-
ing full instructions, the experimenter left the room.
Following the mother's move to the chair, infants were given a 2-rain
warm-up period with the toys (Episode 2). At the end of this 2-rain
period, the female stranger entered the room, greeted the mother and
infant, and sat in the chair next to the mother (Episode 3). The stranger
engaged the mother in conversation for 1 rain, then began a l-rain
SOCIAL RESPONSES 155
scripted conversation with the infant (Episode 4). After a timed signal,
the stranger moved to the floor and attempted to engage the baby in play
with the toys for a 3-rain period (Episode 5). At the end of this period,
the stranger engaged the baby's attention toward a wall hanging with
pockets containing small toys. Once the baby's attention was directed
toward the wall, the stranger picked up the baby and walked toward the
wall (Episode 6). From the time of the pickup, the stranger and infant
spent I rain talking about and playing with the toys from the wall hang-
in_g. Finally, the stranger returned the infant to the floor and returned to
her chair for 30 s (Episode 7). At the end of this 30-s period, the original
experimenter returned to the playroom and the session ended.
Two independent cameras were focused on the playroom simulta-
neously. One camera was focused on the entire room so that the location
of the infant relative to the stranger and mother was in view; the moth-
er's and stranger's chairs faced away from the camera. A second camera
was focused on the infant's face for a close-up view. A special effects
generator combined both video images into a split-screen image for cod-
in~ In order to film through a one-way mirror and still be sure that the
stranger could not see which mask she was wearing, the experimenter
covered the mirror with black material, leaving only a small section
at
the top of the mirror left open for filming purposes. The height of this
open section, howev~ prevented the stranger from seeing herself.
Dependent variables.
The frequency and intensity of several behav-
ioral variables were coded from the videotaped records. The frequency
variables included infant approach to, withdrawal from, and smiling
toward the stranger.
The intensity of infant behaviors was recorded toward both the
stranger and the mother. The affective tone displayed by the infant dur-
ing each episode was initially coded on an I l-point scale of defined
emotional expression ranging from extreme negative to extreme posi-
tive affect. Only a small percentage of the infants who completed the
session exhibited the extremes of behavior, however. Therefore, scores
were collapsed into a I (distress ranging from vocal protest to full inten-
sity cry) to 3 (pleasure ranging from smiling to laughter) intensity scale.
The remaining behavioral responses of intensity were developed from
Ainsworth, Blehar, Waters, and Wall's (1978) scales of attachment be-
havior, but were defined more precisely to facifitate unbiased coding
(e.g., Ritter & Langlois, 1988). Each behavior was coded on an 8-point
ordinal scale for each episode. Stranger-directed behaviors included
avoidance, resistance, and play involvement. Mother-directed behaviors
included proximity seeking and contact maintenance, behaviors that
typically increase in intensity when infants are anxious or distressed
(Ainsworth et al., 1978). The definitions and reliabilifies of each of the
dependent variables can be found in Table 1.
Coders were naive to the hypotheses of the study. In addition, the face
of the stranger was not visible to the coders in two of the three episodes
(Episodes 3 and 7) to ensure that the attractiveness of the stranger did
not bias the coding of the behavior of the infant. Analyses performed
on individual episodes (see Footnote 2 later in the article) revealed as
many or more significant effects in the predicted direction for the oc-
cluded episodes as for the unoccluded one, indicating that the coders
were not biased by the cue of the stranger's face. For all of the coding, a
second observer coded a randomly selected subset of 25% of the ses-
sions. Interobserver reliabifity was estimated using intraclass corre-
lations assuming random raters, a conservative assessment of reliability
with high generalizability. Infant approaches toward the str~, as de-
fined, happened so rarely that they could not be reliably measured and
were not, therefore, included in any further analyses. For the frequency
variables, reliability averaged. 79; for the intensity measures, reliability
averaged .75.
Results
Infant responses to the stranger were compared in Episode 3
when the granger first entered the room, Episode 5 when the
stranger approached and initiated play with the infant, and Epi-
sode 7 when the stranger released the infant after pickup. These
were the three episodes in which infants were predicted to re-
spond to the facial appearance of the stranger: on first impres-
sion, during initial interactions, and al~er close contact? In-
fants' behavior toward their mothers was also compared be-
tween attractiveness conditions in the same episodes.
The data were analyzed using 2 X 2 × 3 univariate analyses
of variance (ANOVAS) and multivariate analyses of variance
(MANOVAS), with sex of infant and attractiveness of the manger
as between.subjects factors and episode as a within-subjects fac-
tor. A MANOVA was performed on the correlated dependent
variables of proximity seeking and contact maintenance; AN-
OVAS were performed on uncorrelated variables. Because there
were different numbers of boys and girls in each condition, the
regression approach to ANOVA and MANOVA was used to con=
trol for chance effects due to unequal cell sizes.
The analyses of the intensity measures revealed several pre-
dicted effects. A significant main effect for condition indicated
that the infants showed more positive affect (affective tone) in
the attractive condition than in the unattractive condition, F(I,
56) = 6.28, p < .01 (see Table 2). The only other significant
effect for this variable was a main effect for episode, F(2, 112) =
4.46, p < .01, which indicated that the infants' affective tone
was more negative during the last episode (Episode 7) than dur-
ing either Episodes 3 or 5.
A significant main effect for condition showed that the infants
were more involved in play (play involvement) with the manger
in the attractive compared with the unattractive condition, F(1,
55) = 7.85, p < .01 (Table 2). A significant episode main effect
also indicated that the infants were more involved in playing
with the stranger in Episode 5 than in the other two episodes,
F(2, 110) = 16.34, p < .01. Finally, a marginally significant
effect for condition was obtained for resistance. Infants were
more resistant toward the stranger in the unattractive condition,
F(1, 55) -- 3.00, p < .08 (Table 2).
The frequency of the infants" social responses to the stranger
differed between the experimental condition in which the
stranger wore the attractive mask and the one in which she wore
the unattractive mask (Table 2). A significant main effect for
condition revealed that the infants showed more frequent with-
drawal from the stranger in the unattractive than in the attrac-
tive condition, F(I, 56) = 4.37, p < .04 (Table 2). The main
effect for condition was modified by a significant interaction
between condition and episode, F(2, 112) = 3.43, p < .04. Tests
of simple effects showed that Episode 5 accounted for most of
the condition effect, F(I, 56) = 4.09, p < .05. Means composing
this interaction for the attractive and unattractive conditions
were, respectively, .04 and .09 in Episode 3, .23 and .88 in Epi-
sode 5, and .08 and .18 in Episode 7. Significant effects for epi-
sode, F(2, 112) = 5.97,p < .01, and for the Sex × Episode inter-
action, F(2, 112) -- 3.10, p < .05, also indicated that more fre-
quent withdrawal was evident in Episode 5 than in the others,
especially by girls.
' Episode 6 (stranger pickup) was not coded because there was no
variance in this episode: The behavior of the infants was constrained by
the stranger as she held them; they could not approach or avoid the
stranger nor seek contact with the mother.
156
J. LANGLOIS, L. ROC~MAN, AND L. RIESER-DANNER
Table
1
Variable Definitions and Reliabilities
Variable Definition Reliability
Frequency measures
Smile
Withdrawal
Intensity measures
Affective tone
Play
involvement
with stranger
Avoidance to stranger
Resistance to stranger
Proximity seeking
toward mother
Contact maintenance
toward mother
Frequency oflips clearly up-turned in response to stranger.
Frequency of turning or moving away from stranger.
Scored from 1 (protest vocalizations to full intensity crying) to 3
(observable pleasure, no distress).
Scored from 0 (no involvement, including no looking) to 4
(accepts toy) to 7 (vocalizes and offers toys 2 or more times).
Scored from 0 (looks, approaches or shows toys to stranger) to 4
(looks away or turns away 2 times) to 7 (never looks or
responds in any way and turns completely away for more than
5 s).
Scored from 0 (unresponsive to proximity or contact with
st~nt~e*) to 4 (one strong push away with negative
vocalization and arched back) to 7 (persistent hitting and
pushing away of stran~ or offered toys).
Scored from 0 (no movement toward mother) to 4 (approach
more than
3/4 distance to morn without contact) to 7
(full
approach ending with contact for 15 s or more).
Scored from 0 (no contact) to 4 (baby initiates contact, leans or
touches room briefly) to 7 (baby initiates contact; clings more
than l rain).
.77
.80
.98
.76
.65
.68
.94
.72
No relevant significant effects were obtained from the MA-
NOVA performed on the intensity of proximity seekin$-contact
maintenance or from the ANOVA performed on the frequency
of smile. 2
Study 2
Method
The purpose of Study 2 was, again, to go beyond recording infant
visual behavior by examining behavioral differences toward attractive
and unattractive faces. A second purpose was to test infant responses
to facial attractiveness in yet another context. A third purpose was to
replicate the findings of Study 1 with stimuli that could not possibly
influence the behavior of the infants. In Study 2, the faces were on cloth
dolls and the behavior examined was the amount of time infants spent
holding and playing with the dolls. Because dolls do not behave, their
Table
2
Significant Attractiveness Effects
Attractive Unattractive
condition condition
Variable
M SD M SD
Study I
Intensity measures
Affective tone 2.72 0.36 2.42 0.58
Play involvement 3.87 1.02 3.15 0.82
Resistance 0.35 0.56 0.82 1.25
Frequency measures
Withdrawal 0.12 0.33 0.38 0.68
Study 2
Play with doll (s) 97.60 80.12 71.07 48.79
attractiveness could not provide differential behavioral cues that might,
in turn, influence infant behavior.
Subjects.
Fifty-six 12-month-old infants participated (28 girls and 28
boys)) All the infants were full term and healthy and were recruited
from subjects participating in a variety ofother research projects at the
Children's Research Lab (23 of these subjects had participated in Study
1). Infants were asked to participate if they had completed their partici-
pation in other studies but did not appear fatigued or distressed. The
procedure was described to the parents, and informed consent was ob-
tained.
The final data analyses were based on forty-three 12-month-olds (22
girls and 21 boys), including 18 who had participated in Study I. Data
from infants who did not play for at least 60 s were excluded from the
final analyses. Data from 13 infants (6 girls and 7 boys) were excluded
for this reason.
Stimuli.
Two 10-in. cloth dolls were constructed. The dolls had ink-
drawn faces that were schematic drawings of faces from photographs of
infants who had been previously rated for physical attractiveness by a
panel of at least 40 adults (a ~ .90). One face was traced from a photo-
graph of an infant rated as highly attractive (M = 4.09, scale from 1 to
5). In order to match the second face to the first on dimensions such as
face shape and feature placement, the second face consisted of features
2 By including episode as a repeated measures variable, the sample
size dropped significantly for the variables of resistance, smile, contact
maintenance, and proximity seeking, because the coders could not see
the infants in all episodes (they occasionally wandered offcamera). The
power of these tests was therefore not a great as those performed on the
other variables. When univariate analyses for each episode ~ per-
formed on these variables, significant or marginally significant condi-
tion effects in the predicted direction were found also for smiling toward
the stranger and for contact maintenance toward mothers.
3 Six-month-olds were initially included to provide data parallel with
those from the visual preference studies (Langlols et al., 1987). Howeve~
the 6-month-olds did not play with the dolls (23 played for less than 60
s) so they were dropped from further consideration.
SOCIAL
RESPONSES
157
from two photographs of infants
rated as
relatively unattractive (M =
2.10andM= 2.71, scale from I to 5).
The two dolls were identical in every way except for their faces and
the
color of the buttons on their
clotl~. Both dolls had yellow yarn
hair and gray clothing, but the attractive doll had black buttons and the
unattractive doll had white buttons. Ohsetvers were told to keep
track
of which doll was which by looking at the buttons.
Procedure.
Infants were placed in a seat attached to a table. The two
dolls were placed on stands within reach of the infant. The stands kept
the dolls upright and in easy view, but the dolls could be easily removed
from the stands by the infants. The leR and right positions of the dolls
were counterbalanced to avoid side bias. Both dolls were replaced on
the stands if either doll was drepped to the floor. The parent filled out a
questionnaire while he or she was seated a few feet away from the infant.
Observers used stopwatches to record the total duration of any inten-
tional physical contact with the dolls. Observation continued until the
subject fussed or protested or until 5 min had passed, whichever was
sooner.
For reliability purposes, all of the observers recorded data from eight
videotaped observations. The intraclass correlation, assuming random
raters, was .97. In addition, two naive observers, who had not previously
been involved in the
study
and who had not observed the infants in the
laboratory, recorded data from 19 additional videotapes. Their in-
traclass correlation with the original live observations was .90.
Results
Data were analyzed using a repeated measures ANOVA, with
attractiveness of the dolls as a within-subjects factor and infant
sex as a between-subjects factor. There was a significant effect
of doll attractiveness, F(l, 39) = 4.73, p < .04, and of sex, F(l,
39) = 5.31, p < .03. Infants contacted the doll with the attractive
face more than the doll with the unattractive face (Table 2).
Also, girls played with the dolls (M = 204.76 s,
SD
= 132.17 s)
more than the boys did (M = 134.23 S,
SD
= 54.88 s). The ef-
fect of doll attractiveness did not interact with the effect of in-
fant sex.
General Discussion
The results of these two studies were consistent and straight-
forward: Infants interacted differently with the attractive female
stranger and inanimate doll than they did with the unattractive
stranger and doll. Infants showed significantly more positive
affective tone, more play involvement with the stranger, less
withdrawal, and marginally less resistance toward the stranger
in the attractive condition. The results of Study 2 paralleled the
play findings of Study 1: Infants played significantly more with
an attractive compared with an unattractive doll.
The 12-month-olds in these two studies exhibited differential
social behaviors indicative of social preferences toward attrac-
tive and unattractive faces. These results complement and ex-
tend those we reported earlier showing that infants visually dis-
criminate between and visually prefer attractive female faces.
The results of the two studies reported here are important be-
cause they substantiate the preference interpretation of the vi-
sual behavior of infants. They also weaken an alternative inter-
pretation of the visual preference data, namely, that the ob-
served
visual
behavior of very
young infants is only an
incidental, temporary consequence of the sensitivity of the hu-
man visual system to certain aspects of visual stimuli such as
curves, concentricity, and vertical symmetry that may be more
characteristic of attractive than unattractive faces (e.g., Fantz,
F~n, & Miranda, 1975; Fisher, Ferdinandsen, & Bornstein,
1981). Although the origin of visual preferences for attractive
faces may lie in innate perceptual sensitivities, the social behav-
iors of infants suggest that preferences for attractive faces are
not merely artifacts of these perceptual processes, nor are they
limited to visual behavior in experimental situations with little
relevance for social interaction and behavior.
Another issue these results address, albeit indirectly, is the
continuity of preferences for attractive faces from those seen in
3- and 6-month-olds to those seen in older children and adults.
The developmental psychologist is always faced with a dilemma
in demonstrating continuity in an organism that changes dra-
matically over time. Continuity of specific behaviors can rarely
if ever be demonstrated empirically, given the major cognitive
and behavioral changes that take place in the human over devel-
opment (e.g., Lerner, 1986; Scarr, 1985; Sroufe & Waters,
1977). Instead, functional continuity must be inferred from the
organization of behaviors across different situations, contexts,
and levels of development (e.g., SackeR, Sameroff, Cairns, &
Suomi, 1981; Sroufe & Waters, 1977). At ages 3- and 6-months,
infants have very limited ways ofexpressing preferences. Visual
behavior at these ages is meaningful because infants are not mo-
bile and cannot talk or otherwise express social preferences to-
ward unfamiliar persons in their environment. By 12 months
of age, however, infants are capable of many more social, cogni-
tive, and motoric acts, yet they still have not received much ex-
posure to television or other means of cultural transmission of
attractiveness stereotypes and preferences. Most 3- and 4-year-
olds are verbal and play with other children; adults, obviously,
have a large behavioral repertoire with which to indicate prefer-
enc~i.
In each of these ages, behavior differentiating attractive and
unattractive faces is seen, Thus, very young infants look longer
at
attractive faces; older infants play more with attractive hu-
mans and dolls and show more withdrawal and resistance, and
less positive affect toward unattractive faces; children and
adults verbally indicate more liking for attractive individuals
(e.g., Dion, 1973; Langlois & Stephan, 1977) and behave prefer-
entially toward them (e.g., Dion, 1974; Langlols & Downs,
1979; Snyder et al., 1977). Although no longitudinal work is
available to demonstrate individual differences in the continuity
of preferences for attractive faces, the behaviors of 3-, 6-, and
12-month-old infants, older children, and adults seem to repre-
sent functionally equivalent preferences for attractive faces.
The magnitude of the differences in infant social responses to
attractive and unattractive strangers was
not l~, 4 although
the
infants' play behavior directed toward the dolls showed rela-
tively large differences as a function of attractiveness. It is not
the case that infants seek interaction with attractive strangers
and avoid interaction with unattractive strangers in an either-
or fashion. Rather, differences in infant behavior toward the
stranger were more subtle, just as differences in visual behavior
4Tbe very short length of each episode mitigated against
larger
differences being obtained. In addition,
larger
diffe~nces in behavior
would likely be obtained with more extreme exemplars of attractive and
unattractive faces.
158 J. LANGLOIS, L. ROGGMAN, AND L. RIESER-DANNER
toward attractive and unattractive faces were subtle (although
quite consistent across infants). The magnitude of the differ-
ences in these visual and social behaviors toward attractive and
unattractive faces are reasonable, given that it would not be
adaptive for infants to refuse to look at or interact with adults
who may become a caregiver. Indeed, these infant preferences
for attractive faces may be apparent only for unfamiliar faces:
Because of the importance of the attachment system to the sur-
vival of the infant, attractiveness is not likely to influence infant
behavior toward familiar caregivers and parents, although it is
possible that infant attractiveness may influence the behavior of
adult caregivers toward infants (Hildebrandt & Carman, 1985;
Langlois & Casey, 1984).
The results of these two studies reveal information about the
generality of infant visual preferences for attractive faces to
other more social contexts, yet they do not indicate why infants
prefer attractive faces or what stimulus dimensions underlie at-
tractiveness in faces. Although the question of what constitutes
beauty has interested scientists for many years (e.g., Darwin,
1859; Galton, 1878), and although some knowledge has been
obtained regarding the facial measurements that predict attrac-
tiveness ratings (e.g., Cunningham, 1986; Hildebrandt & Fitz-
gerald, 1979), no significant progress has been made in our un-
derstanding of the functional significance of attractiveness. As
work accumulates that suggests that the origins of standards of
and preferences for attractiveness lie not in the gradual cultural
transmission of currently prevailing norms but rather in innate
or very early acquired mechanisms, establishing the stimulus
dimensions and functional significance of attractiveness be-
comes more crucial. What information do attractive and unat-
tractive faces differentially communicate to young infants, chil-
dren, and adults alike? Why are cross-cultural judgments of at-
tractiveness more similar than different, despite common
assumptions to the contrary (e.g., Bernstein, Lin, & McClellan,
1982; Cunningham, 1986, Johnson, Dannenbring, Anderson,
& Villa, 1983; Kleck et al., 1974; Langlois & Stephan, 1977;
Maret, 1983; Richardson, Goodman, Hastorf, & Dornbusch,
1961; Stephan & Langlois, 1984; Weisfeld, Weisfeld, & Cal-
laghan, 1984)? Why is facial attractiveness an influence on hu-
man visual behavior as early as 3 months of age and on social
behavior as early as 12 months of age? It is to these questions
that future research must now turn.
References
Adams, G. R,, & Crane, P. (1980). An assessment of parents' and teach-
ers' expectations of preschool children's social preference for attrac-
tive or unattractive children and adults. Child Development, 51, 224-
231.
Ainsworth, M. D. S., Blchar, M., Waters, E., & Wall, S. (1978). Patterns
of attachment. Hillsdale, NJ: Eflbaum.
Anderson, S. M., & Bern, S. L. (1981). Sex typing and androgyny in
dyadic interaction: Individual differences in responsiveness to physi-
cal attractiveness. Journal of Personality and Social Psychology, 41,
74-86.
Banks, M. S., & Salapatek, P. (1983). Infant visual perception. In
M. M. Haith & J. J. Campos (Vol. Eds.), Infancy and developmental
psychobiology (pp. 435-571). In P. H. Mussen (Ed.), Handbook of
child psychology. New York: Wiley.
Bernstein, I. H., Lin, T., & McClellan, P. (1982). Cross- vs. within-racial
judgments of attractiveness. Perception & Psychophysics, 32, 495-
503.
Bersebeid, E., & Walste~ E. (1974). Physical attractiveness. In L. Ber-
kowitz (Ed.), Advances in experimental social psychology (pp. 157-
215). New York: Academic Press.
Bersebeid, E., Dion, K. K., Walst¢~ E., & Walster, G. W. (1971). Physi-
cal attractiveness and dating choice: A test of the matching hypothe-
sis. Journal of Experimental Social Psychologg 7, 173-189.
Buchman, H. ( 1971 ). Stage makeup. New York: Watson-Guptill.
Cavior, N., & Lombardi, P. A. (1973). Developmental aspects of judg-
ments of physical attractiveness in children. Developmental Psychol-
ogy, 8, 67-71.
Cross, J. E, & Cross, J. (1971). Age, sex, race, and the perception of
facial beauty. Developmental Psychology,, 5, 433-439.
Cunningham, M. R. (1986). Measuring the physical in physical attrac-
tiveness: Quasi-experiments on the sociohiology of female facial
beauty. Journal of Personality and Social Psychology,, 50, 925-935.
Darwin, C. (1859). On the origin of species by means of natural selec-
tion. London: John Murray.
Dion, K. K. (1973), Young children's stereotyping of facial attractive-
ness. Developmental Psychology,, 9, 183-188.
Dion, K. K. (1974). Children's physical attractiveness and sex as deter-
minants of adult punitiveness. Developmental Psychology,, 10, 772-
778.
Dion, K. K. (1977). The incentive value of physical attractiveness for
young children. Personality and Social Psychology Bulletin, 3, 67-
70.
Dion, K. K., & Berscheid, E. (1974). Physical attractiveness and peer
perception among children. Sociomet~ 37, 1-12.
Dion, K. K., & Stein, S. (1978). Physical attractiveness and interper-
sonal influence. Journal of Experimental Social Psychology,, 14, 97-
108.
Elder, G. H., Van Nguyen, T, & Caspi, A. (1985). Linking family hard-
ship to children's fives. Child Development, 56, 361-375.
Fantz, R. L. (1965). Visual perception from birth as shown by pattern
selectivity. Annals of New York Academy of Science, 118, 793-814.
Fantz, R. L., & Fagan, J. E (1975). Visual attention to size and number
of pattern details by term and preterm infants during the first six
months. Child Development, 46, 3-18.
Fantz, R. L., Fagan, J. E, & Miranda, S. B. (1975). Early visual selectiv-
ity. In L. B. Cohen & P. Salapatek (Eds.), Infant perception: From
sensation to cognition (pp. 249-341 ). New York: Academic Press.
Fisher, C. B., Ferdinandsen, K., & Bornstein, M. H. (1981). The role
ofsyrnmetry in infant form discrimination. Child Development, 52,
457--462.
Galton, E (1878). Composite portraits. Journal of the Anthropological
Institute of Great Britain & Ireland, 8, 132-142.
Haith, M. M. (1980). Rules that babies look by. Hillsdalc, N J: Erlbaum.
Hildebrandt, IC A., & Carman, T. (1985). The distribution of caregiver
attention in a group program for young children. Child Study Jour-
nal 15, 43-55.
Hildebrandt, K. A., & Fitz~rald, H. E. (1978). Adults' responses to
infants varying in perceived cuteness. Behavioral Processes, 3, 159-
172.
Hildebrandt, K. A., & Fitzgerald, H. E. (1979). Facial feature determi-
nants of perceived infant attractiveness. Infant Behavior and Develop-
ment, 2, 329-339.
Hildebrandt, K. A., & Fitzgerald, H. E. ( 1981). Mothers' responses to
infant physical apiw, arance. Infant Mental~alth Journal 2, 56-61.
Johnson, R. W., Dannenbring, G. L., Anderson, N. R., & Villa, R. E.
(1983). How different cultural and geographic groups perceive the
attractiveness of active and inactive feminists. The Journal of Social
Psychology, 119, 111-117.
Kleck, R. E., Richardson, S. A., & Ronald, C. (1974). Physical appear-
SOCIAL RESPONSES 159
ance cues and interpersonal attraction in children. Child Develop-
ment, 45, 305-310.
Kleck, R. E., & Rubenstein, C. (1975). Physical attractiveness, per-
ceived attitude similarity and interpersonal attraction in an opposite
sex encounter. JournalofPersonality and Social Psycholog3z, 31, 107-
114.
Langlois, J. H., & Casey, R. J. (1984, April). Baby beautiful: The rela-
tionship between infant physical attractiveness and maternal behav-
ior. Paper presented at the fourth biennial International Conference
on Infant Studies, New York.
Langlois, J. H., & Downs, A. C. (1979). Peer relations as a function of
physical attractiveness: The eye of the beholder or behavioral reality?
Child Development, 50, 409--418.
Langlois, J. H., & Stephan, C. (1977). The effects of physical attractive-
ness and ethnicity on children's behavioral attributions and peer pref-
erences. Child Development, 48, 1694-1698.
Langlois, J. H., & Stephan, C. ( 1981). Beauty and the beast: The role of
physical attractiveness in the development of peer relations and social
behavior. In S. S. Brehm, S. M. Kassin, & E X. Gibbons (Eds.), Devel-
opmental social psychology: Theory and research (pp. 152-168). New
York: Oxford University Press.
Langlois, J. H., Roggman, L. A., Casey, R. J., Ritter, J. M., Rieser-Dan-
her, L. A., & Jenkins, V. Y. (1987). Infant preferences for attractive
faces: Rudiments of a stereotype? Developmental PsychologTz, 23,
363-369.
Lerner, R. M. (1986). Concepts and theories of human development.
New York: Random House.
Maret, S. M. (1983). Attractiveness ratings of photographs of Blacks by
Cruzans and Americans. The Journal of Psychology, 115, 113-116.
Power, T. G., Hildebrandt, K. A., & Fitzgerald, H. E. (1982). Adults'
~sponses to infants varying in facial expression and perceived attrac-
tiveness. Infant Behavior and Development, 5, 33--44.
Reis, H. T., Nezlek, J., & Wheeler, L. (1980). Physical attractiveness in
social interaction. Journal of Personality and Social Psychology,, 38,
604-617.
Reis, H. T., Wheeler, L., Spiegel, N., Kernis, M. H., Nezlek, J., & Perri,
M. (1982). Physical attractiveness in social interaction: 2. Why does
appearance affect social experience? Journal of Personality and So-
cial Psychologg, 43, 979-996.
Richardson, S. A., Goodman, N., Hastorf, A. H., & Dornbusch, S. M.
(1961). Cultural uniformity in reaction to physical disabilities. Amer-
ican Sociological Revie~, 26, 241-247.
Ritter, J. M., & Langlois, J. H. (1988). The role of physical attractiveness
in the observation of adult-child interactions: Eye of the beholder or
behavioral reality? Developmental Psycholog~, 24, 254-263.
Sackett, G. P., Sameroff, A. J., Cairns, R. B., & Suomi, S. J. (1981).
Continuity in behavioral development: Theoretical and empirical is-
sues. In K. Immelmann, G. Barlow, L. Petrinovich, & M. Main
(Eds.), Behavioral development: The Bielefeld Interdisciplinary Proj-
ect (pp. 23-58). New York: Cambridge University Press.
Sarnuels, C, A., & Ewy, R. (1985). Aesthetic perception of faces during
infancy. British Journal of Developmental Psychologg, 3, 221-228.
Scarf, S. (1985). Constructing psychology. American Psychologist, 40,
499-512.
Shapiro, B. A., Eppler, M., Haith, M. M., & Reis, H. (1987, April).
.4n event analysis of facial attractiveness and expressiveness. Paper
presented at the meeting of the Society for Research in Child Devel-
opment, Baltimore, MD.
Smith, G. J. (1985). Facial and full-length ratings of attractiveness re-
lated to the social interactions of young children. Sex Roles, 12, 287-
293.
Snyder, M., Tanke, E. D., & Berscheid, E. (1977). Social perception and
interpersonal behavior. On the self-fulfilling nature of social stereo-
types. Journal of Personality and Social Psycholog~, 35, 656-666.
Sroufe, L. A., & Waters, E. (1977). Attachment as an organizational
construct. Child Development, 48, 1184-1199.
Stephan, C., & Langlois, J. H. (1984). Baby beautiful: Adult attributions
of infant competence as a function of infant attractiveness. ChildDe-
velopment, 55, 576-585.
Weisfeld, G. E., Weisfeld, C. C., & Callaghan, J. W. (1984). Peer and
self-perceptions in HOPi and Afro-American third- and sixth-graders.
Ethos, 1Z 64-83.
Received August 8, 1988
Revision received March 3, 1989
Accepted March 6, 1989 •
A preview of this full-text is provided by American Psychological Association.
Content available from Developmental Psychology
This content is subject to copyright. Terms and conditions apply.
