the Psychonomic Society
1978, Vol. 12 (1)
Smiling, body position, and
McGINLEY, PATSY McGINLEY,
Wyoming, Laramie, Wyoming 82071
Two experiments were conducted. In
I, regardless of her body position, awoman
majority of the time (70%) was seen as more interpersonally
awoman who seldom smiled (20%). When
woman seldom smiled, she was
when she displayed open body positions
closed body positions. In Experiment 2, the closed body position/smiling
eye gazes were monitored while they viewed the slides of
smiling condition, subjects looked
total looking time.
McGinley, leFevre, and McGinley (1975) and
McGinley, Nicholas, and McGinley (1978) found
female stranger was evaluated more positively when she
displayed open as opposed to closed body positions.
Although body positions were manipulated, there was
no attempt to systematically manipulate other non-
verbal information about the stranger.
Given the importance of the face in the expression of
emotions, it seems that a valuable extension of the
McGinley et al. (1975, 1978) studies would be to inves-
tigate.the concomitant effects of body position and facial
expression on interpersonal attraction. Obviously, how-
ever, facial expressions per se are multitudinous, and a
specific aspect of them , such as smiling, would be a
reasonable choice for an initial investigation .'
Though smiles can be used for deviated purposes such
as deceit (Ekman &Friesen, 1975), they are more com-
monly associated with friendliness, happiness, and joy
(Izard, 1971). Without discussing the possible innateness
of smiling and its inherent relationship to pleasant inter-
nal stimuli or to the removal of unpleasant internal stim-
uli, it seems safe to say that from early infancy humans
are exposed to others who smile and that those smiles
are usually temporally associated with rewarding inter-
personal interactions that satisfy physical and social
needs. As a result of these associations, the smile becomes
aconditioned stimulus that elicits emotional responses
that commonly stimulate evaluative positive overt re-
sponses. This interpretation is generally consistent with
Lott and Lott 's (1972) theory of interpersonal liking.
Empirically speaking, a female stranger who expresses
open body positions
seen as more interpersonally
attractive than one who displays closed body positions.
If smiling elicits positive evaluative responses, the effect
of combining smiling and body positions should lead to
higher attraction when a person expresses open body
positions and smiles a lot than when the person expresses
Requests for reprints should be sent to Hugh McGinley,
Department of Psychology , University of Wyoming, Laramie,
Wyoming 8207 1.
closed positions and smiles or when the person smiles
little and displays either open or closed body positions.
In the first experiment, subjects were shown slides of
a stranger displaying either open or closed body positions
and either smiling or not smiling. Experiment 1prompted
the second experiment, in which the subjects' eye-gaze
behavior was monitored to see if the stranger's smiling
or unsmiling face was attended to differentially:
Subjects. Ninety-six female student s from introductory
psychology classes at
University of Wyoming volunteered to
participate in the experiment. The subjects met in small groups
of from two to seven people.
Materials and Equipment. We used 50 color slides
stranger, a Kodak 860Z projector, two Hunter timers, the Inter-
personal Judgment Scale (IJS) (Byrne , 1971), bipolar scales of
evaluations, and a postexperimental questionnaire.
The 50 slides consisted
position slides of
the stranger and 10 slides each of open body position/smiling,
open body position/nonsmiling, closed
and closed body position/nonsmiling. Two independent judges
selected the slides from over 400 taken of the female model
stranger. Only slides consistently rated as depicting the stranger
in open, neutral, or closed body positions and smiling or not
smiling were used. The criterion for rating consistency was that
both judges agreed on a slide's rating during an initial and retest
judgment. Within the smiling category,
matched (i.e., a body position was presented in both a smiling
and nonsmiling mode). The open body slides showed the model
in various combinations of the following: leaning backward, legs
stretched out , knees apart , one ankle crossed over the other
. knee, elbows far away from her body, hands held outward, and
arms held outward from her body either at her side or elevated .
The closed body slides showed that model in some combination
of the following: elbows next to her body, arms crossed, hands
folded in her lap, knees pressed together, feet together, and legs
crossed at either the knees or the ankles. In the 10 neutral slides,
the model was typically leaning back , her legs slightly stretched
out and slightly apart, hands resting on her lap or on her thighs ,
and arms held relatively close to her sides.
The US (Byrne, 1971) consisted of six seven-alternative items
having to do with intelligence , personal liking, working together,
personal adju stment, morality , and knowledge of current events .
As is typical for attraction research using the US, the ratings '
22 McGINLEY, McGINLEY, AND NICHOLAS
with different superscripts differ at p <.01.
The standard error
the mean for the IJS-Attraction and
Evaluation-Attraction data, respectively, were .44 and .67.
For both sets
data, the higher the score the more positive the
Perhaps, given the relatively short viewing time for
each slide (5.5 sec), subjects who saw the smiling stranger
were attracted to her smile and spent less time looking
at her body and, therefore, were not influenced by body
position. The largest difference in the evaluations occur-
red when the stranger expressed closed positions and
Mean Attraction Scores for Body Position
and Smiling Conditions, Experiment I
because the subjects indicated they knew the stranger).
For the US score, there was a borderline body position
effect [F(1,89) =3.61, P< .07] and a smiling effect
.OOl]. However, the Body Position
by Smiling interaction was noteworthy [F(1 ,89) =3.66,
p < .07] . To avoid a Type
error (which is likely where
there is a borderline level
significance for a main effect
and a borderline interaction
another variable with that
main effect), simple effects analyses were conducted.
Analyses showed that the stranger was rated as most.
attractive when she smiled; when she smiled a majority
the time, there was no effect for body position. When
she smiled very little (20%), she was viewed as less
attractive. The latter effect, however, was affected by
body position; the stranger was seen as less attractive in
the closed than in the open body positions. The bipolar
evaluation data led to similar findings. There
was a significant smiling effect [F(1,89) =12.42, p <
.001] and, although there was no main effect for body
position, the Body Position by Smiling interaction was
significant [F(1,89) =3.96, p < .05]. Simple effects
analyses for these data led to the same conclusion as did
those for the US data. Means for both attraction meas-
ures are shown in Table 1. In summary, when there was
a higher percentage of smiling by the stranger (70%),
body position did not matter; body position made a
difference if she smiled little (20%). The latter conclusion
was supported by the McGinley et al.
ing that a nonsmiling stranger displaying open body posi-
tions was seen as more evaluatively positive than on dis-
playing closed positions.
the smiling variable on the body position
variable can be seen in Figures 1 and 2. Two judges read
the responses to the postexperimental questionnaire.
There was no evidence that the subjects were intuitively
aware of the hypothesized relationship between smiling,
body position, and interpersonal attraction.
made with the liking and working together items were summed
as a single measure
interpersonal attraction. A high score indi-
cated interpersonal attraction (positive evaluation). A secondary
attraction was made by having the subjects rate the
stranger on four seven-interval bipolar scales
(sweet-sour, clean-dirty, bad-good, and cruel-kind). The ratings
were summed to a single score from 4 (negative) to 28 (positive).
The postexperimental questionnaire consisted of six ques-
tions designed to gather information about the subjects' percep-
the experiment. The questions were (a) what were you
supposed to do; (b) what kind of results do you think we expect
to find; (c) were your judgm ents about the woman based on how
you felt about her or on how you thought we expected you to
respond (explain); (d) what was the purpose of asking you to
make judgments about the woman ; (e) specifically, how do you
think we expected you to feel about the woman;
other comments do you have?
Procedure. When the subjects arrived at the psychology
laboratory, they were met by a female experimenter wearing a
white laboratory coat. She introduced herself and explained that
the experiment, as the subjects had been told prior to volunteer-
ing, involved person perception. The subjects were told that they
would receive limited information about a female stranger and
that the information was in the form of 20 color slides that had
been taken of the woman while she conversed about various
topics. The experimenter then told the subjects that they would
be asked to make judgments about the woman, based on the in-
formation received. The lights were dimmed and the subjects
were shown the slides. The slides were shown on a wall screen
to a projected size of approximately 120 x 180 cm. Each slide
was shown for 5 .5 sec; the inter slide interval was approximately
.75 sec. Following the slide presentation, the subjects made
interpersonal judgments about the stranger and filled out the
The experimental manipulation was made using four dif-
ferent sets of slides: (a) Open body position/smiling, where
10 neutral body positon slides were interspersed with 10 open
body position slides of the stranger. The stranger was smiling in
all 10 of the open body posiiton slides and in four of the neutral
body position slides. (b) Closed body position/smiling, where the
same 10 neutral slides were interspersed with 10 closed body
position slides. All of the closed body position slides and four
the neutral body slides showed the stranger smiling. (c) Open
body position/nonsmiling, where 10 neutral body position slides
were mixed in with 10 open body position slides. None of the
open body position slides showed the stranger smiling; however,
she was shown smiling in four of the neutral body position
slides. The open body positions were matched to those shown in
the open body position/smiling condition. (d) Closed body
position/nonsmiling, where the 10 neutral slides were inter-
spersed with 10 closed body position slides. None of the closed
body position slides showed the stranger smiling. As in the other
conditions, the stranger was smiling in four of the neutral body
position slides. The body position manipulation, then, consisted
of 50% neutral and 50% either open or closed body position
slides, while the smiling manipulation consisted
(4/20) or 70% (14/20) smiling across the 20 slides.
Hypothesis. In accord with prior results (McGinley et al.,
1975, 1978), we hypothesized a body position effect where the
stranger would be rated as more interpersonally attractive when
she expressed open body positions than when she expressed
closed body positions. Also, we speculated that the stranger
would be evaluated more positively when she smiled frequently
with respect to the attraction elicited by the stranger, the order
of the experimental conditions from high to low would be
smiling-open, smiling-closed, nonsmiling-open, and nonsmiling-
Attraction scores were analyzed by analysis
ance with unequal ns (three subjects' data were omitted
either smiled frequently or seldom (Table I) . If these
differences were also associated with significant differ-
ences in where subjects looked at the stranger, one could
conjecture that the body position effect was negated in
the smiling conditions, not because of a strong effect of
the smiles per se, but because the smiles resulted in the
subjects' inattention to the body position. To test this
possibility, we monitored the subjects' eye-gaze patterns
while they viewed the slidesof the stranger.
Subjects. Thirty-nine female students from the University
of California at San Diego volunteered to serve as subjects. They
were obtained through bulletin board advertisement and direct
solicitation. The subjects ranged from freshmen to graduate
students but were mostly freshmen and sophomores. Each sub-
ject was seen individually and was paid $2 for her participation.
Materials and Equipment. With minor exceptions, the ma-
terials and equipment were the same as those used in Experi-
ment 1. The bipolar scales were not used and only the slides
from the closed body position/smiling and closed body position/
nonsmiling conditions were used. The slides were projected to a
size of approximately 80 x 120 em. The slide time was 5.5 sec,
while the interslide time was .75 sec. Also, a Whittaker Eye
View monitor and pupillometer system was used to record the
subjects' eye movements." The eye movement data were reo
corded on a Sony 3650 videocorder and then analyzed at a later
date. These data were scored for time spent looking at the face
of the stranger. The times were recorded in seconds and tenths
of seconds and were twice measured by a singlejudge. The data
that were used in the analyses were the average of the two
face-viewing times. The interrater coefficient of reliability for
the two time measures was .99.
Procedure. The procedure was the same as in Experiment I,
except that in this experiment the subject was told that her
pupillary response would be monitored while she viewed the
pictures of the stranger and thought about what kind of person
the stranger was. Also, for each subject, the eye monitoring
equipment had to be aligned; the alignment took between
2 and 5 min.
Results and Discussion
The attraction data showed that subjects who saw the
closed body position/smiling slides were more interper-
sonally attracted to the stranger than were those who saw
the closed body position/nonsmiling slides. The means
were, respectively, 10.61 and 7.95 [t(37) =4.94, p <
.0 I] . This result is consistent with that in Experiment 1.
The looking data were based on 38 subjects, as the
data for one subject could not be used. Analysis of these
data revealed no difference in face-viewing behavior be-
tween subjects who viewed the smiling pictures and those
who viewed the stranger smiling very little . On the aver-
age, subjects spent about 55% of their viewing time look-
ing at the stranger's face. Face-viewing times ranged
from 30% to 95% of the total looking time. Usually, the
subjects looked at the stranger's face, scanned across
various parts of her body, then looked back at her face.
Although Experiment 2 was conducted as a descrip-
tive study, we thought that a smiling face would attract
more visual attention. However, the smiling and nonsmil-
ing faces attracted similar visual attention. Perhaps the
subjects modified their eye-gaze behavior because they
our interest in where they looked while
viewing the slides. This possibility, however, was not
SMILINGAND BODY POSITION 23
Figure 1. Examples of open body positions with a smiling
(a) and nonsmiling stranger (b).
supported by responses to the postexperimental ques-
tionnaire. When the experimenter introduced the experi-
mental task to the subjects, he spoke
pupil for size changes but did not mention recording eye
gaze. A majority
the subjects, in their responses to
the postexperimental questionnaire, referred to the rela-
Examples of closed body positions with a smiling
(a) and nonsmiling stranger
tionship between pupil size and emotional state,
one mentioned that the experimenter might be interested
in where she was looking while viewing the slides.
body position and smiling are aspects of
human behavior that frequently and consistently occur during
interpersonal interactions, very little is known about their
effects on the perceived attraction of the interacting individuals.
While the results of the present experiments shed some light on
these variables, the generalizations that can be
made from these effects are limited because : (a) the lack of a
difference between open and closed body position when the
stranger smiled frequently may have been due to a limiting
effect of the attraction scale; (b) a single stranger was used and
since physical attractiveness is known to be related to inter-
personal attraction, the results may have been confounded with
the subjects' perceptions of the stranger's physical attractiveness;
and (c) only female subjects and a female stranger were used .
With these limitations in mind, the results of the two experi-
ments indicate that both smiling and body position influence
interpersonal attraction, but that these variables interact such
that high smiling negates the attraction effect of body position
on interpersonal attraction. Overall, these results warrant the
further investigation of the concomitant effects of personal
nonverbal behaviors on interpersonal attraction.
D. The attraction paradigm. New York: Academic
W. V. Unmasking the face.
Englewood Cliffs, N.J : Prentice-Hall, 1975.
C. E. The face
emotion. New York: Appleton-
LoTI, A. J., &LOTI, B. E. The power of liking: Consequences
of interpersonal attitudes derived from a liberalized view of
secondary reinforcement. In 1. Berkowitz (Ed.), Advances
in experimental social psychology (Vol. 6). New York:
Academic Press, 1972. Pp. 109-148.
P. The influ-
ence of a communicator's body position on opinion change
in others. Journal
Personality and Social Psychology,
1975, 31, 486-490.
P. Effects of
body position and attitude similarity on interpersonal
attraction and opinion change . Psychological Reports ,
1978, 42, 127-138.
1. In their 1975 study, McGinley et al. had nearly equal
smiles in their open and closed body position condi-
tions, seven and six respectively . The faces were inked
the 1978 study .
2. The Whittaker equipment was borrowed from the Navy
Personnel Development and Research Center, San Diego, Cali-
fornia. Our thanks to Ed Aiken and Tom Duffy for the loan of
the equipment. Also, we wish to thank Jean Mandler and the
psychology department at the University of California at San
Diego for allowing us the use of research facilities.
(Receivedfor publication March 6,1978.)