ARTICLE IN PRESS
Appetite 49 (2007) 92–99
www.elsevier.com/locate/appet
Research Report
Effects of social influence on eating in couples,
friends and strangers
Sarah-Jeanne Salvya,, Denise Jarrinb, Rocco Palucha, Numrah Irfanc, Patricia Plinerd
a
Department of Pediatrics, Division of Behavioral Medicine, State University of New York at Buffalo,
Farber Hall, Room G56, 3435 Main Street, Building #26, Buffalo, NY 14214-3000, USA
b
Concordia University, Montreal, Que., Canada
c
University of Guelph, Ont., Canada
d
University of Toronto at Mississauga, Mississauga, Ont., Canada
Received 11 September 2006; received in revised form 11 December 2006; accepted 15 December 2006
Abstract
Previous research indicates that both males and females eat less in the presence of a stranger of the opposite sex than in the presence of
a same sex. Another literature shows that people tend to model or matching the amount eaten by others. The extent to which people are
eager to inhibit their food consumption or match other’s intake is likely to vary as a function of the characteristics of the co-eater. The
present study examines how males and females adjust their level of eating as a function of their familiarity with and the gender of their
eating companion, using a free-eating paradigm. Findings indicated that both the familiarity between co-eaters and the participants’
gender predicted food consumption. Although unfamiliarity suppressed both men’s and women’s food intakes, the matching effect
operated only when a female co-eater was involved. We conclude that the overarching motive (i.e., producing a positive impression) does
not necessarily vary substantially across the various gender-familiarity combinations, but that the means or strategies (eating lightly and
or matching of intake) by which the person accomplishes it and the strength of the motive vary as a function of the audience. In other
words, in some social contexts self-enhancing motives can be served by restricting intake as well as through ingratiatory strategies such as
attitudinal or behavioral conformity.
r 2007 Elsevier Ltd. All rights reserved.
Keywords: Social influences; Eating; Free-eating paradigm; Impression management
Introduction
Food is obviously used to satisfy the body’s needs, but it
can also serve a communicative function. What and how
much you eat communicate more than your culinary
preferences; it is a self-presentational statement. You are
what you eat! Self-presentation, or impression management, involves regulating one’s own behavior to create a
specific image for an audience (Leary & Kowalski, 1990;
Leary et al., 1994; Schlenker, 1975), and eating behavior
may have utility for impression management (Bock &
Kanarek, 1995; Mori, Chaiken, & Pliner, 1987; Pliner &
Chaiken, 1990). Conveying a positive impression through
Corresponding author.
E-mail address: ssalvy@buffalo.edu (S.-J. Salvy).
0195-6663/$ - see front matter r 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.appet.2006.12.004
eating can manifest itself in eating lightly and through
eating conformity (i.e., matching or modeling the amount
eaten by someone else).
Studies on impression management through eating
lightly show that some eating companions make people
more eager to suppress their intake in order to convey a
good impression (Mori et al., 1987; Pliner & Chaiken,
1990). These studies have looked at the amount of food
people consumed when the characteristics of the eating
companion were manipulated. Mori et al. (1987, Experiment 1) found that both male and female participants ate
less in the presence of a desirable partner of the opposite
sex than in the presence of a less desirable partner. In a
second study, women whose feminine identity was threatened (i.e., because they scored ‘‘masculine’’ on a questionnaire) restricted their intake. Pliner and Chaiken (1990)
ARTICLE IN PRESS
S.-J. Salvy et al. / Appetite 49 (2007) 92–99
replicated Mori et al.,’s (1987) findings and explored the
reported social motives operating in interpersonal situations. They found that behaving in a socially desirable
manner was an important motive for men, while being
socially desirable and appearing feminine were important
for women. Other research shows that women who eat
minimally are viewed more positively than are women who
eat a lot (Bock & Kanarek, 1995; Chaiken & Pliner, 1987).
Further, it is widely assumed that people become overweight because of a lack of self-control around food
(DeJong & Kleck, 1986; Puhl, Schwartz, & Brownell,
2005), and overweight people are stereotyped as lazy, selfindulgent, unattractive, lacking self-esteem, socially inept,
uncooperative, and intellectually slow (Allon, 1982; DeJong, 1993; Harris, 1990; Hebl & Heatherton, 1998; Madey
& Ondrus, 1999). Thus, individuals’ concern about eating
too much in front of others and attempts to manage
impressions by eating lightly are not totally misguided.
The effect of modeling or matching effect on can be seen
most clearly in studies using a conformity paradigm, in
which a naı̈ve participant eats in the presence of a
confederate whose level of eating is predetermined by the
experimenter. These studies have consistently found that
participants eat less in the presence of a confederate who
eats minimally than in the presence of one who eats a large
amount (Conger, Conger, Costanzo, Wright, & Matter,
1980; Goldman, Herman, & Polivy, 1991; Nisbett &
Storms, 1974; Roth, Herman, Polivy, & Pliner, 2001;
Herman, Roth, & Polivy, 2003). One explanation for these
findings is that people believe that by conforming they will
ingratiate themselves to others; behavioral similarity will
lead others to like and accept them (e.g., Sunnafrank, 1991;
Sunnafrank, Ramirez, & Metts, 2004). In the language of
Jones and Pittman’s (1982) model of self-presentation,
matching the intake of the eating companion would have a
similar ingratiating function as attitude conformity. However, most studies on eating conformity have averaged the
participants’ intake and compared these levels of eating to
the consumption of experimental confederates who were
eating predetermined amounts of food. A more appropriate test of the matching hypothesis would involve
assessing the relationship between co-eaters’ intakes (i.e.,
correlational analysis).
Most of the findings in the literatures on impression
management and matching in eating are based on eating
among strangers, so it is not known whether familiarity
mediates or moderates these effects. From a self-presentational perspective, people should be more motivated to
convey a good impression during their initial interactions
with a stranger than with someone who they know well
(Leary et al., 1994). This is consistent with findings
indicating that individuals use different self-presentation
strategies when conversing with different targets, relying on
self-enhancement with strangers but shifting toward
modesty with friends (Tice, Butler, Muraven, & Stillwell,
1995). The familiarity hypothesis is further reinforced by
Clendenen, Herman, and Polivy’s (1994) finding that
93
people ate less dessert when eating with strangers than
with friends, although there was no difference in entrée
consumption (deli food) between friends and strangers.
Presumably, strangers were more concerned with selfenhancement (i.e., appearing socially desirable or feminine)
and limited their intake of ‘‘forbidden’’ dessert. Evidence
for the familiarity hypothesis also comes from de Castro’s
findings that social facilitation of eating (people eat more in
the presence of others than when they eat alone) appears to
be stronger when people eat in the presence of friends and
family than when they eat with strangers (de Castro, 1990,
1994).
The studies reviewed above indicate that people sometimes inhibit their food intake when eating with someone
else (Mori et al., 1987; Pliner & Chaiken, 1990) and
sometimes match their co-eater’s food consumption
(Nisbett & Storms, 1974; Rosenthal & Marx, 1979;
Rosenthal & McSweeney, 1979; Roth et al., 2001).
Although, the evidence for impression management and
matching (eating conformity) has come from separate
literatures, we contend that it is possible for both processes
to operate at the same time and to be studied simultaneously. The idea is that overarching motive (i.e., producing a positive impression) does not necessarily vary
substantially, but that the means or strategies (eating
lightly and or matching of intake) by which the person
accomplishes it and the strength of the motive vary as a
function of the audience. In some social contexts selfenhancing motives can be served by restricting intake while
in others individuals may use ingratiatory strategies such as
behavioral conformity (Jones & Pittman, 1982).
The present study examines how males and females
adjust their level of eating as a function of their familiarity
with and the gender of their eating companion, using a
free-eating paradigm. In previous matching studies, because the confederates’ intake was varied only at two levels,
we can conclude only that participants’ eating is elevated
when the confederate eats a lot compared to when the
confederate eats only a little. We cannot determine how
well the subjects are ‘‘tracking’’ the confederates’ intake. In
contrast, the free-eating paradigm allows the formulation
of predictions regarding not only the amount of food
consumed by the two participants, but also the relationship
between co-eaters’ food intake (i.e., matching effect).
We contend that people strategically vary their level of
eating with different audiences to fulfill a similar overarching motive of a positive self-presentation. Based on the
impression management literature, we predicted that both
males and females would eat less in the presence of a
partner of the opposite sex than in the presence of a same
sex co-eater (Mori et al., 1987; Pliner & Chaiken, 1990).
However, we also predicted that the inhibiting effect of an
opposite sex co-eater would be moderated by familiarity.
Presumably, eating that takes place in the first stages of a
relationship is in marked contrast with the eating that
occur when partners are well acquainted with each
other. Individuals interacting with well-known others are
ARTICLE IN PRESS
94
S.-J. Salvy et al. / Appetite 49 (2007) 92–99
generally assured of their affection and have less need to
use strategic self-presentation to obtain their approval
(Jellison & Gentry, 1978; Leary et al., 1994).
Consistent with a self-presentational account of eating
conformity, we predicted that individuals would match the
amount eaten by their partner, resulting in a positive
correlation between the amounts eaten by the two. We also
predicted a stronger effect of matching among strangers
than among acquainted co-eaters because of the greater
importance of ingratiation when eating with a stranger
than when eating with someone who knows one well (Jones
& Pittman, 1982). Empirically this should translate into
higher correlations between co-eaters’ intakes in dyads
composed of strangers than in the corresponding acquainted dyads. Finally, we predicted that the more
affiliative and supportive nature of women’s interpersonal
interactions would promote greater eating conformity in
dyads involving at least one female (i.e., greater relationship between co-eaters’ intake) (Eagly, 1978; Wood,
Christensen, Hebl, & Rothgerber, 1997). Women may
conform more than men, because they take greater
responsibility for establishing and maintaining positive
interpersonal bonds (Eagly, 1978, 1983, 1987), while men’s
greater drive for independence and distinctiveness
(Baumeister & Sommer, 1997; Cross & Madson, 1997)
lends itself to non-conformity with regard to eating.
Method
Overview and design
The design used in this experiment was a 2 3 factorial
design with familiarity (familiar vs. strangers) and dyads’
gender composition (both females, both males, mixed) as
between-subject factors. The design generated six experimental conditions: opposite-sex romantic partners, opposite-sex strangers, female friends, female strangers, male
friends and male strangers. Under the pretext of a study on
patterns of interaction among friends, couples and
strangers, participants were asked to converse in dyads
about a non-threatening topic (i.e., enjoyable activities) for
ten minutes. With the rationale of making the situation
more relaxed and similar to everyday-life conversations,
participants were casually offered pre-weighed bowls of
snack foods (cookies and crackers). The main variable of
interest was the total amount of snack in grams consumed
by participants.
Participants
Participants were 173 women and 137 men recruited
through the psychology undergraduate subject pool,
through an advertisement posted on a bulletin board
devoted to the recruitment of paid participants in experiments, through word of mouth, or through direct approach
by the experimenters in public places around the campus.
Students recruited through the subject pool received a
course credit for their participation; all other participants
received a 10-$ compensation. At the time of recruitment,
participants were told that the study assayed patterns of
interaction among friends, couples and strangers and were
informed that they would be conversing for 10 min with
their romantic partner, a friend or a stranger, depending on
the condition. All participants were asked to bring a friend
or their romantic partner for the experimental session. The
individuals accompanying the participants (friend, romantic partner) were also compensated (i.e., course credit if
they were students and 10-$ compensation if they were
not). On the day of the experiment, participants were
randomly paired with their friend or partner or with a
stranger. Dyads of strangers were composed of participants
who were paired with the partner/friend of another
participant who had come for the same session. At the
time of recruitment, participants were told that food
consumption could impact on moods and that for this
reason they were required to abstain from eating 2 h before
their experimental session. Adherence to this recommendation was assessed upon their arrival to the laboratory.
Participants who did not comply with this requirement
were rescheduled. All experimental sessions were conducted between 1000 and 1800 h. Table 1 displays the
participants’ mean ages and weights across conditions.
Female friends had known each other for an average of
2.96 years, male friends for 4.16 years and romantic
partners for 2.75 years.
Procedure
After consent procedures, participants were instructed to
discuss activities that they enjoyed doing with either their
Table 1
Mean ages and body weights across conditions (SD)
Unfamiliar
Familiar
Same sex
Age
Weight (lb)
Opposite sex
Same sex
Opposite sex
Female
Male
Female
Male
Age
Weight
Age
Weight
19 (1.11)
132 (26.8)
20 (4.25)
158.6 (31.08)
19 (1.32)
128.44 (28.60)
19 (1.64)
159.29 (27.01)
19 (1.19)
132.27 (24.73)
19 (1.68)
162.11 (30.36)
20 (4.19)
122.13 (21.64)
21 (4.52)
170.12 (28.73)
Note: Participants (males and females separately) did not differ in terms of body weight across conditions.
ARTICLE IN PRESS
S.-J. Salvy et al. / Appetite 49 (2007) 92–99
friends or with their romantic partner depending on the
person they had brought with them. For example,
participants who came with their friends were asked to
talk about activities they enjoyed doing with their friends.
Eating was therefore ‘‘incidental’’ to what was described
and the study did not involve working on a task, which
might have influenced the amount participants eat (e.g., a
forced tasting task).
On the pretext of making the atmosphere more
convivial, participants were casually offered bowls copiously filled with crackers and cookies. So that it would be
possible to assess the intake of each participant independently, each received a separate set of bowls. They were
casually told that they could eat as much or as little as they
wanted while conversing. These pre-weighed bowls of
snack foods were positioned so that participants had access
only to their own bowls without being able to reach their
partner’s bowls. Each participant had access to three small
bowls of snacks (described below). Participants were sitting
side by side at a table but their chairs were positioned to
face each other. A tape recorder was placed on the table
and participants were told that they would be left alone for
the discussion but that their conversation would be
recorded. Post-experiment interviews indicated that participants were naı̈ve to the real purpose of the experiment
and were unaware that their snack consumption would be
recorded.
Since the purpose of the study did not involve analyzing
the participants’ dialogues, the recorded conversations
were erased after each session. Before leaving the room, the
experimenter informed the participants that she would
knock on the door at the end of the 10 min conversation
period. After exactly 10 min, the experimenter returned to
the experimental room and turned off the tape recorder.
The participants were asked to end their conversation, the
snacks were removed and participants were asked to
complete a questionnaire package about their daily habits,
behaviors, preferences and perceptions. Questionnaires
were administered and height and weight measurements
were obtained at the end of the session so that these
measurements would not influence the eating data.
Participants were then debriefed about the real purpose
of the experiment, and the bowls of snacks of each
participant were weighed on a Tanita electronic scaleTM
to the nearest gram. Our dependent variable was the total
amount of snack foods eaten by each participant during
the discussion, as measured by the weight in grams.
Food
Two kinds of cookies and one type of crackers were used
in this study. The cookies were Chips Ahoy Mini
Chocolate Chip cookies and Oreo Sandwich Cookies Mini
Bite Size, and the crackers were Ritz Bits Cracker Cheese
Sandwiches. All three snacks were produced by Nabisco
(East Hanover, NJ) and all had an average weight of 2 g
per cookie/cracker and an average of 10 kcal (7.3 g) per
95
cookie/cracker. Cookies and crackers were stored in
hermetically sealed bags. Uneaten snacks were discarded
after each session. Each participant received pre-weighed
bowls of cookies and crackers (150720 g of snacks per
bowl). The amounts of food provided were intentionally
large so that participants could eat as much as they wanted
without finishing all the food provided.
Results
Data analysis
In this study, we were interested in two aspects of
participants’ food intake: (1) the average levels of food
consumption and (2) the relationship between co-eaters’
food intakes. Before proceeding to the main analyses, we
assessed whether there was any difference of body weight
across conditions, given that such difference would impact
on the results; there were not (p ¼ .28). We also compared
female friends, male friends and couples in terms of how
long the participants in these dyads had known each other.
The results of the ANOVA revealed no significant
difference across these conditions.
Amount of food consumed
When the observations are non-independent, it is
necessary to treat the dyad rather than the individual as
the unit of analysis (Kenny, 1995; Kenny & Judd, 1986).
One participant’s intake is influenced by the partner’s
intake and has to be considered in relation to other
participant’s consumption. This mutual influence violates
the assumption of independence between observations
required for analysis of variance. Random-effects models
(also called multilevel, hierarchical linear or mixed models)
provide a useful approach to account for interdependence
in two-person relationships (Gibbons & Hedeker, 1994;
Hedeker & Gibbons, 1994; Hedeker, Gibbons, & Flay,
1994). These models make it possible to assess the effects
due to the individual while controlling for the variance due
to the other member of the dyad. This is determined by the
covariances among the regression coefficients and can be
characterized by a covariance function (Hedeker &
Gibbons, 1996). The outcomes at the individual level are
modeled taking into consideration the dependence of a
participant’s data on the other individual’s data (Hedeker,
2003). Participants’ snack intakes were analyzed using
SYSTAT Software (Systat, 2004). Based on our hypotheses, familiarity, partner’s gender, participant’s gender and
interactions between these variables were entered as fixed
effects into the models as predictors of participants’ snack
intake.
Fig. 1 illustrates participants’ mean snack consumption
across conditions. A random regression analysis performed
with consumption as the dependent variable revealed a
main effect of familiarity, F(1, 153) ¼ 15.80, po.001, a
main effect of participant gender F(1, 153) ¼ 11.53,
ARTICLE IN PRESS
96
S.-J. Salvy et al. / Appetite 49 (2007) 92–99
90
80
70
Snack Consumption (g)
60
50
40
30
20
10
0
Male
Male
Friends Strangers
(n=36)
(n=40)
Female Female
Friends Strangers
(n=36)
(n=76)
Females
Males
Females
Couples
(n= 74)
Males
Strangers
(n=48)
Fig. 1. Mean snack consumptions and standard errors across conditions.
po.001, and a familiarity by partner gender interaction,
F(1, 153) ¼ 6.44, po.05. Differences of least square means
indicated that male friends (M ¼ 70.14) ate more than male
strangers (M ¼ 30.92), t(153) ¼ 3.73, po.001), female
friends (M ¼ 31.16), t(153) ¼ 4.21, po.001), and female
strangers (M ¼ 23.06), t(153) ¼ 4.36, po.001). Male
friends also ate more than both males (M ¼ 30.83),
t(153) ¼ 3.58, po.001) and females eating with an
opposite sex stranger (M ¼ 16.92), t(153) ¼ 4.85,
po.001) and more than males (M ¼ 43.87), t(153) ¼
2.65, po.01) and females (M ¼ 44.08), t(153) ¼ 2.62,
po.001) eating with their significant other. In other words,
male friends ate more than all other participants. Females
eating with their significant others ate more than females
eating with female strangers, t(153) ¼ 2.12, po.05, and
with male strangers, t(153) ¼ 2.68, po.01.
The proportion of participants, who ate nothing,
referred to as ‘‘non-eaters’’, in the couples group was
10%; whereas as in the opposite sex strangers condition,
30% ate nothing. In the friends conditions, non-eaters
ranged from 1% to 10% for males and females friends,
respectively. In the opposite-sex strangers group, 10% for
males and 40% for the females did not eat. When the noneaters were removed from the analyses, the results were
substantially the same.
Relationship among co-eaters’ intakes
The second question of interest was whether co-eaters’
familiarity and the dyads’ gender composition would
moderate the intake matching reported in previous studies.
Table 2
Correlation between partners’ intake by condition
Familiarity
Male dyads
Female dyads
Mixed dyads
Familiar
Unfamiliar
.23a
.18a
.53**a
.73**a
.56**b
.59*b
* ¼ po.005; ** ¼ po.001.
a
Intraclass correlation.
b
Pearson correlation.
The assessment of relationships between two measures via
Pearson correlations applies only to situations where
members of a dyad are distinguishable (i.e. one unequivocally belongs to the X group and the other unequivocally
belongs to the Y group). When the designation of members
of a pair is arbitrary (i.e., when there is no way to
disentangle variability due to specific individual), intraclass
correlation coefficients (ICCs) provide accurate estimations
of the magnitude of the relationships between variables
(Griffin & Gonzalez, 1995; Shrout & Fleiss, 1979). ICCs
were calculated for undistinguishable dyads (i.e., same-sex
dyads), while Pearson correlations coefficients were calculated for distinguishable dyads (i.e., opposite-sex dyads).
ICCs for dyads are interpreted in the same fashion as
Pearson correlations (Kenny, Kashy, & Cook, 2006).
Therefore, intradyadic similarity was indexed by the
Intraclass and Pearson correlation coefficients. The ICCs
were measured with the SPSS 12.0 ‘‘reliability’’ procedure
(SPSS, 2003), using a one-way random model (Howell,
2000).
ARTICLE IN PRESS
S.-J. Salvy et al. / Appetite 49 (2007) 92–99
Table 2 displays the correlation coefficients between the
co-eaters’ consumption. The relationships between the coeaters’ intakes were used as indexes of social influence. The
degree of intake matching was high in all dyads with the
exception of the relationships found in dyads composed of
male strangers and male friends. All intradyadic correlations of intake were statistically significant with the
exception of the ICCs in the two types of male-only dyads
(male friends and male strangers). Comparison of all the
ICCs generated indicated that the dyads composed of male
strangers significantly differed from all the other dyads
with the exception of the correlation of intake in pairs of
male friends. Male friends’ matching of intake significantly
differed from the ICC observed in dyads composed of
female strangers (Z ¼ 2.78, po.01) and marginally differed
from the correlation found in couples (Z ¼ 1.91,
p ¼ .056).
Discussion
This study examined how males and females adjust their
level of eating as a function of their familiarity with their
co-eater and the dyads’ gender composition, using a freeeating paradigm. The two variables of interest were the
amount of snack-food the participants ate and the degree
of intake matching among eating companions.
The results pertaining to the amount of food consumed
indicated that both male and female participants who were
eating with familiar co-eaters ate more than male and
female participants eating in the presence of unfamiliar
eating companions. However, these results were qualified
by the gender of the partner. Females in the couple
condition ate more than females in all other conditions,
while male friends ate significantly more than the males in
all other conditions. The effect of familiarity on food
consumption is consistent with findings indicating that
social facilitation of eating may be stronger among friends
and relatives than among strangers and stronger among
men than among women (de Castro, 1994).
As expected, participants’ gender had an effect on the
level of matching. The matching data indicated that the
overall degree of intake relationship was high in all dyads
involving at least one female. In contrast, the ICCs in
dyads involving only males (i.e., male strangers and male
friends) did not reach statistical significance. This is
consistent with the studies on sex-related differences in
social interactions and conformity described earlier (Boyatzis, Chazan, & Ting, 1993; Davis & Franzoi, 1991;
DePaulo, Epstein, & Wyer, 1993; Exline, 1972) on which
we based our prediction. These studies have found that
women are more likely than men to consider their partner’s
perspective in order to anticipate the partner’s needs and
reactions and are better at adjusting to situational factors
to foster affiliation (Pollak Levine & Feldman, 1997).
Assuming that self-presentation through eating operates in
the same way, females may have adjusted their level of
intake to ingratiate themselves, appear likeable, or cement
97
a bond of acquaintance. In this view matching of eating
would have the social function of fostering group cohesiveness and enhancing amicability. Conversely, men’s
drive for distinctiveness (Baumeister & Sommer, 1997;
Cross & Madson, 1997) may have lent itself to nonconformity with regard to eating. In other words, men
might have monitored accurately how much their eating
companions were eating, but used this information to
distinguish or separate themselves from others rather than
to foster similarity.
Contrary to our hypothesis, the partners’ familiarity did
not significantly influence matching. The matching of
intake in dyads of strangers did not differ from the
matching in dyads of acquainted co-eaters. Our failure to
obtain a familiarity effect in the matching data is puzzling
given that we did obtain a familiarity effect in amounts
eaten. We are unable to explain this at the moment and can
only suggest that it is possible that the depth of the
relationship influences the amount of matching on eating.
In this study, this variable was not assessed beyond
examining the duration of the relationships between friends
and between partners. Although these dyads did not
differ significantly in terms of how long the partners
had known each other, it is possible that differences in
terms of the depth of the relationships moderate the
effects of social influences on snack consumption and their
intake matching.
Since social conditions often act by increasing the
amount of time at meals (de Castro, 1994; Pliner, Bell,
Hirsch, & Kinchla, 2006), the restriction that the session
was always 10 min long probably affected the results. This
time limitation was useful for experimental control, but it
limited the interpretation of the findings. Similarly,
although there were significant differences in food intake
as a function of familiarity and participant gender, the
overall food intake was rather low (e.g. females eating with
male strangers). The time constraint imposed on eating
probably explains, at least partly, low food consumption.
Future studies might benefit from using a similar paradigm
to look at meal, rather than snack intake.
Future research ought to explore the social motives
accounting for differences between men and women with
regard to matching of eating. Even if women are somewhat
more willing to match others’ intake when it seems that
conformity is useful to develop or cement a bond (Eagly,
1978), there might be a threshold that they are not ready to
cross because their own personal values and desires conflict
with their inclination to adapt to the situation. And further
research should examine the moderating/mediating effects
of depth and/or duration of relationship on the kinds of
behaviors studied here.
References
Allon, N. (1982). The stigma of overweight in everyday life. In B. B.
Wolman (Ed.), Psychological aspects of obesity: A handbook
(pp. 130–174). New York: Van Nostrand Reinhold.
ARTICLE IN PRESS
98
S.-J. Salvy et al. / Appetite 49 (2007) 92–99
Baumeister, R. F., & Sommer, K. L. (1997). What do men want? Gender
differences and two spheres of belongingness. Psychological Bulletin,
122, 38–44.
Bock, B. C., & Kanarek, R. B. (1995). Women and men are what they eat:
The effects of gender and reported meal size on perceived characteristics. Sex Roles, 33, 109–119.
Boyatzis, C. J., Chazan, E., & Ting, C. Z. (1993). Preschool children’s
decoding of facial emotions. Journal of Genetic Psychology, 154,
375–382.
Chaiken, S., & Pliner, P. (1987). Women, but not men, are what they eat:
The effect of meal size and gender on perceived femininity and
masculinity. Personality and Social Psychology Bulletin, 13, 166–176.
Clendenen, V. I., Herman, C. P., & Polivy, J. (1994). Social facilitation of
eating among friends and strangers. Appetite, 23, 1–13.
Conger, J. C., Conger, A. J., Costanzo, P. R., Wright, K. L., & Matter, J.
A. (1980). The effect of social cues on the eating behavior of obese and
normal subjects. Journal of Personality, 48, 258–271.
Cross, S. E., & Madson, L. (1997). Models of the self: Self-construals and
gender. Psychological Bulletin, 122, 5–37.
Davis, M. H., & Franzoi, S. L. (1991). Stability and change in adolescent
self-consciousness and empathy. Journal of Research in Personality, 25,
70–87.
de Castro, J. M. (1990). Social facilitation of duration and size but not rate
of the spontaneous meal intake of humans. Physiology & Behavior, 47,
1129–1135.
de Castro, J. M. (1994). Family and friends produce greater social
facilitation of food intake than other companions. Physiology &
Behavior, 56, 445–455.
DeJong, W. (1993). Obesity as a characterological stigma: The issue of
responsibility and judgements of task performance. Psychological
Reports, 73, 963–970.
DeJong, W., Kleck, R. E. (1986). The social psychological effects of
overweight. In Paper presented at the physical appearance, stigma, and
social behavior. The Ontario symposium, Hillsdale, NJ.
DePaulo, B., Epstein, J., & Wyer, M. (1993). Sex differences in lying: How
women and men deal with dilemma of deceit. In M. Lewis, & C. Saarni
(Eds.), Lying and deception in everyday life (pp. 126–147). New York,
NY: Guilford.
Eagly, A. H. (1978). Sex differences in influenceability. Psychological
Bulletin, 85(1), 86–116.
Eagly, A. H. (1983). Gender and social influence: A social psychological
analysis. American Psychologist, 38, 971–981.
Eagly, A. H. (1987). Sex differences in social behavior: A social-role
interpretation, Vol. xii. NJ, England: Lawrence Erlbaum Associates,
Inc.
Exline, R.V. (1972). Visual interaction: The glances of power and
preference. In Paper presented at the Nebraska symposium on
motivation, Lincoln, NE.
Gibbons, R. D., & Hedeker, D. (1994). Application of random-effects
probit regression models. Journal of Consulting and Clinical Psychology, 62, 285–296.
Goldman, S. J., Herman, C. P., & Polivy, J. (1991). Is the effect of a social
model on eating attenuated by hunger? Appetite, 17, 129–140.
Griffin, D., & Gonzalez, R. (1995). Correlational analysis of dyad-level
data in the exchangeable case. Psychological Bulletin, 118,
430–439.
Harris, M. (1990). Is love seen as different for the obese? Journal of
Applied Social Psychology, 20, 1209–1224.
Hebl, M. R., & Heatherton, T. F. (1998). The stigma of obesity in women:
The difference is black and white. Personality and Social Psychology
Bulletin, 24, 417–426.
Hedeker, D. (2003). A mixed-effects multinomial logistic regression
model. Statistics in Medicine, 22, 1433–1446.
Hedeker, D., & Gibbons, R. D. (1994). A random-effects ordinal
regression model for multilevel analysis. Biometrics, 50, 933–944.
Hedeker, D., & Gibbons, R. D. (1996). MIXREG: A computer program
for mixed-effects regression analysis with autocorrelated errors.
Computer Methods and Programs in Biomedicine, 49, 229–252.
Hedeker, D., Gibbons, R. D., & Flay, B. R. (1994). Random-effects
regression models for clustered data with an example from smoking
prevention research. Journal of Consulting and Clinical Psychology, 62,
757–765.
Herman, C. P., Roth, D. A., & Polivy, J. (2003). Effects of the presence of
others on food intake: A normative interpretation. Psychological
Bulletin, 129, 873–886.
Howell, D. C. (2000, August 26). Intraclass correlation: For unordered
pairs. Retrieved July 26, 2006, from /http://www.uvm.edu/dhowell/
StatPages/More_Stuff/icc/icc.html##Shrout&FleisS.
Jellison, J. M., & Gentry, K. W. (1978). A self-presentation interpretation
of the seeking of social approval. Personality and Social Psychology
Bulletin, 4, 227–230.
Jones, E. E., & Pittman, T. S. (1982). Toward a general theory of strategic
self-presentation. In J. Suls (Ed.), Psychological perspectives on the self,
Vol. 1 (pp. 231–262). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.
Kenny, D. A. (1995). The effect of nonindependence on significance
testing in dyadic research. Personal Relationships, 2, 67–75.
Kenny, D. A., & Judd, C. A. (1986). Consequences of violating the
independence assumption in analysis of variance. Psychological
Bulletin, 99, 422–431.
Kenny, D. A., Kashy, D. A., & Cook, W. L. (2006). The analysis of dyadic
data. New York: Guilford Press.
Leary, M. R., & Kowalski, R. M. (1990). Impression management: A
literature review and two-component model. Psychological Bulletin,
107, 34–47.
Leary, M. R., Nezlek, J. B., Downs, D., Radford-Davenport, J., et al.
(1994). Self-presentation in everyday interactions: Effects of target
familiarity and gender composition. Journal of Personality and Social
Psychology, 67, 664–673.
Madey, S. F., & Ondrus, S. A. (1999). Illusory correlations in perceptions
of obese and hypertensive patients’ noncooperative behaviors. Journal
of Applied Social Psychology, 29(6), 1200–1217.
Mori, D., Chaiken, S., & Pliner, P. (1987). ‘‘Eating lightly’’ and the selfpresentation of femininity. Journal of Personality and Social Psychology, 53, 693–702.
Nisbett, R. E., & Storms, M. D. (1974). Cognitive and social determinants
of food intake. In H. London, & R. E. Nisbett (Eds.), Thought and
feeling: Cognitive alternation of feeling states (pp. 190–208). Chicago:
Aldine.
Pliner, P., Bell, R., Hirsch, E. S., & Kinchla, M. (2006). Meal duration
mediates the effect of ‘‘social facilitation’’ on eating in humans.
Appetite, 46, 189–198.
Pliner, P., & Chaiken, S. (1990). Eating, social motives, and selfpresentation in women and men. Journal of Experimental Social
Psychology, 26, 240–254.
Pollak Levine, S., & Feldman, R. S. (1997). Self-presentational goals, selfmonitoring, and nonverbal behavior. Basic and Applied Social
Psychology, 19, 505–518.
Puhl, R. M., Schwartz, M. B., & Brownell, K. D. (2005). Impact of
perceived consensus on stereotypes about obese people: A new
approach for reducing bias. Health Psychology, 24, 517–525.
Rosenthal, B., & Marx, R. D. (1979). Modeling influences on the eating
behavior of successful and unsuccessful dieters and untreated normal
weight individuals. Addictive Behaviors, 4, 215–221.
Rosenthal, B., & McSweeney, F. K. (1979). Modeling influences on eating
behavior. Addictive Behaviors, 4, 205–214.
Roth, D. A., Herman, C. P., Polivy, J., & Pliner, P. (2001). Selfpresentational conflict in social eating situations: A normative
perspective. Appetite, 36, 165–171.
Schlenker, B. R. (1975). Self-presentation: Managing the impression of
consistency when reality interferes with self-enhancement. Journal of
Personality and Social Psychology, 32(6), 1030–1037.
Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlations:
Uses in assessing rater reliability. Psychological Bulletin, 2,
420–428.
SPSS. (2003). SPSS base 12.0.1 for Windows (Version 12.0.1.). Chicago,
IL: SPSS, Inc.
ARTICLE IN PRESS
S.-J. Salvy et al. / Appetite 49 (2007) 92–99
Sunnafrank, M. (1991). Interpersonal attraction and attitude similarity: A
communication-based assessment. In J. A. Anderson (Ed.), Communication yearbook, Vol. xiv (pp. 451–497). CA, USA: Sage Publications, Inc.
Sunnafrank, M., Ramirez, A., Jr., & Metts, S. (2004). At first sight:
Persistent relational effects of get-acquainted conversations. Journal of
Social and Personal Relationships, 21, 361–379.
Systat. (2004). SYSTAT Software (Version 110). Richmond, CA:
SYSTAT Software, Inc.
99
Tice, D. M., Butler, J. L., Muraven, M. B., & Stillwell, A. M. (1995).
When modesty prevails: Differential favorability of self-presentation to
friends and strangers. Journal of Personality and Social Psychology, 69,
1120–1138.
Wood, W., Christensen, P., Hebl, M. R., & Rothgerber, H. (1997).
Conformity to sex-typed norms, affect, and the self-concept. Journal of
Personality and Social Psychology, 73, 523–535.