Body Image 38 (2021) 49–62

Contents lists available at ScienceDirect

Body Image
journal homepage: www.elsevier.com/locate/bodyimage

Young women’s body image following upwards comparison to

Instagram models: The role of physical appearance perfectionism
and cognitive emotion regulation
Sarah E. McComb ∗ , Jennifer S. Mills
Department of Psychology, York University, 4700 Keele St, Toronto, ON, M3J 1P3, Canada

a r t i c l e i n f o a b s t r a c t

Article history: The present study examined whether trait physical appearance perfectionism moderates young women’s
Received 31 March 2020 body image following upwards appearance comparison to idealized body images on social media, and
Received in revised form 14 March 2021 whether cognitive coping mediates the relationship between physical appearance perfectionism and
Accepted 17 March 2021
resulting body image from social comparison processes. Female undergraduate students (N = 142) were
randomly assigned to either 1) compare the size of their body parts to the body parts of attractive Insta-
Keywords:
gram models, or 2) an appearance-neutral control condition. All participants completed measures of trait
Social comparison
physical appearance perfectionism, pre and post measures of state body image, and state cognitive coping
Body image
Perfectionism
processes. Appearance comparison to the models resulted in lowered confidence and increased appear-
Body dissatisfaction ance and weight dissatisfaction. High trait physical appearance perfectionism predicted lower confidence
Social media and higher weight dissatisfaction and appearance dissatisfaction, and these relationships were mediated
by engagement in rumination and catastrophizing. Clinical implications of the findings are discussed.
© 2021 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).

1. Introduction mon way that young women are exposed to images of thin ideal
bodies (Mills, Shannon, & Hogue, 2017).
Despite the potential benefits of social networking sites, such as Recent research has shown that exposure to thin idealized
maintaining contact with physically distant friends, sharing infor- images on Instagram results in increased negative mood and
mation, and networking (Ellison, Steinfield, & Lampe, 2007), there increased body dissatisfaction among young women, and this is
are well-documented adverse effects of social media use on psycho- found whether the target woman featured was a peer known
logical well-being. Research over several decades has shown that to participants (Hogue & Mills, 2019), a same-aged female peer
women who are exposed to images of idealized bodies internalize that was unknown to participants, a female celebrity (Brown
the thin ideal and strive for an unrealistic standard of beauty, which & Tiggemann, 2016), or a popular female Instagram influencer
can result in feelings of shame, body dissatisfaction, and low mood (Fardouly & Holland, 2018). In other words, it is not just expo-
when they cannot achieve the same body type (Tiggemann & Slater, sure to idealized images of strangers, models, and celebrities in the
2013). Thin ideal imagery is characterized by images of impossibly media, but also exposure to idealized images of peers that triggers
thin and attractive female beauty that are circulated by the mass appearance-based social comparison.
media as the standard of Western beauty (Hargreaves & Tiggemann,
2003). Often these images depict women with unattainable pro- 1.1. Social comparison on social media and body image
portions, flawless skin, thin waists, long and slender legs, and
well-developed breasts (Groesz, Levine, & Murnen, 2002). These The ubiquity of idealized body images on Instagram allows
effects, first demonstrated in traditional forms of media like maga- ample opportunity for young women to engage in social compari-
zines and television, have now been demonstrated in social media, son to peers, celebrities, and models (Verduyn, Gugushvili, Massar,
which have relatively quickly and recently become the most com- Taht, & Kross, 2020). Such images elicit upwards appearance-based
comparisons, which are characterized by comparing oneself to
someone deemed as more attractive or superior in some way
∗ Corresponding author. (Morrison, Kalin, & Morrison, 2004). Social comparison theory
E-mail addresses: [email protected] (S.E. McComb), [email protected] argues that upwards comparisons to others results in a per-
(J.S. Mills). ceived negative contrast between oneself and the idealized model’s

https://doi.org/10.1016/j.bodyim.2021.03.012
1740-1445/© 2021 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
S.E. McComb, J.S. Mills Body Image 38 (2021) 49–62

appearance and, in turn, feelings of dissatisfaction because the measure two domains of physical appearance perfectionism. The
difference between a superior other and oneself becomes salient first is called ‘worry about imperfection’, and reflects deep dissatis-
(Festinger, 1954). The tripartite theory has identified social com- faction with one’s appearance and concerns that one’s appearance
parison as a mediating process between sociocultural pressure will never be good enough. The second domain is ‘hope for per-
from the media about beauty ideals and body dissatisfaction fection’, and reflects hope that others admire one’s appearance and
(Rodgers, McLean, & Paxton, 2015). Research has supported both striving for an attractive appearance. Worry about imperfection has
social comparison theory and the tripartite theory of body dissatis- been shown to be more maladaptive than hope for perfection, due
faction in this context. For instance, state appearance comparison to stronger relations with a host of negative body-related variables
has been found to be a mediator between thin ideal image exposure such as eating disorder symptoms, social anxiety related to appear-
and resulting body image concerns (Brown & Tiggemann, 2016; ance, appearance satisfaction, body shape satisfaction, unhealthy
Fardouly & Holland, 2018; Hogue & Mills, 2019). Appearance com- weight control behaviours, and negative associations with appear-
parisons made on social media have been found to result in greater ance self-esteem and body area satisfaction (Stoeber & Yang, 2015;
negative mood and poorer body image than appearance compar- Yang & Stoeber, 2012). Conversely, hope for perfection has been
isons made in-person or to traditional media (Fardouly, Pinkus, & found to be more closely associated with striving for high goals,
Vartanian, 2017). This may be because of the vast amount of time impression management, and perfectionistic self-promotion (Yang
that young people spend on social media, the widespread use of & Stoeber, 2012). Women who are high in physical appearance
filters and photoshop that can be used to alter images on social perfectionism, especially those with worry about imperfection, are
media, as well as the fact that images women see on Instagram highly invested in their appearance and ought to be especially sen-
are often especially relevant comparison targets, such as same-age sitive to events, such as social comparison, that trigger thoughts
peers (Hogue & Mills, 2019). While past research widely supports that they do not measure up to the appearance ideal. Therefore, it
the link between social comparison to thin ideal imagery and nega- could be expected that the body image of women with high levels
tive body image, there is still little theoretical or empirical research of physical appearance perfectionism, in particular high levels of
that articulates how individual difference factors and cognitive pro- worry about imperfection, would be especially negatively affected
cesses interact with social comparison events. To address this gap by exposure and comparison to thin ideal images on Instagram, as
in the literature, the current study will test a comprehensive expla- this would be threatening to their sense of self. Whereas the social
nation of who is most vulnerable to their adverse effects on body comparison framework is applicable for individuals in general, the
image and why, through examining the impact of perfectionism psychological processes by which women who are high on physi-
and cognitive coping on body image. cal appearance perfectionism come to feel particularly bad about
themselves following upward appearance comparisons is not well
1.2. Perfectionism and body image understood.

While research generally confirms that exposure and compari- 1.3. Cognitive models of body image
son to attractive others on Instagram results in negative changes to
body image in young women, it remains unclear which individuals Cognitive models of body dissatisfaction have been proposed
are most vulnerable to these effects. Not all women feel worse about (Lewis-Smith, Diedrichs, & Halliwell, 2019) that can pull together
themselves and their bodies following exposure to the thin ideal much of the research on social media and body image and extend
(Mills, Polivy, Herman, & Tiggemann, 2002). A meta-analysis of over it in useful ways. A key variable in them is how individuals process
200 studies by Ferguson (2013) revealed that the effects of exposure distressing events that make them cognizant of the discrepancy
to media-portrayed thin ideal images in magazines, TV, and music between their body and the ideal body. Traditional social compar-
videos, on body dissatisfaction are not as widespread as previously ison theory treats the negative contrast that occurs following an
thought, and that effects sizes are negligible for the average woman. upwards comparison as an automatic response. However, this neg-
Instead, there may be certain personality subgroups of women who ative contrast, and how the person comes to feel after an upward
are more vulnerable to the adverse effects of exposure to idealized comparison, is mediated by how a person thinks about or cogni-
body images than others (Ferguson, 2013). Past research on Insta- tively appraises the contrast between themselves and the target
gram exposure has mainly been concerned with examining how of their appearance (e.g., “I look huge compared to her”) (Lewis-
thin ideal imagery impacts women in general, and has failed to Smith et al., 2019). Further, there are individual differences that
examine how personality traits make certain subgroups of women determine how important or relevant body image is to someone.
more vulnerable to the adverse effects of Instagram use than others. Fairburn, Cooper, and Shafran (2003) identify perfectionism as one
Women who are high in the personality trait of perfection- of four processes that may serve to maintain poor body image and
ism may be especially vulnerable to the adverse effects of social distorted cognitions about the body that interfere with cognitive
media exposure on self-evaluation. Shafran, Cooper, and Fairburn behaviour treatment for body image and eating disorders. Fur-
(2002) conceptualize clinical perfectionism as being characterized ther, Cash, Santos, and Fleming Williams (2005) speculated that
by self-evaluation that is overly dependent on the pursuit of per- appearance-schematic individuals, such as those high on physi-
sonally demanding and self-imposed standards in at least one cal appearance perfectionism, have a greater investment in how
domain, despite adverse consequences that may come from those they look and will therefore be more attuned to appearance related
trying to achieve these standards, such as extreme self-criticism stimuli. They reasoned then that those who are highly invested
and negative self-evaluation. Perfectionism has been consistently in their appearance may be especially vulnerable to negative cog-
found to be associated with body dissatisfaction (Donovan, Chew, nitive distortions or appraisals following social comparison to an
& Penny, 2014; Wade & Tiggemann, 2013; Welch, Miller, Ghaderi, attractive other because they are likely to feel threatened by this
& Vaillancourt, 2009), and has also been shown to be predictive negative contrast. Those with negative body image attitudes and
of social comparison to idealized others in everyday life (Pokrajac- schemas have distorted and dysfunctional cognitions about their
Bulian, Ambrosi-Randic, & Kukic, 2008) as well as on Instagram and appearance, such as magnifying appearance defects and making
Facebook (Padoa, Berle, & Roberts, 2018). biased social comparison (Cash et al., 2005; Lewis-Smith et al.,
Physical appearance is one domain in which individuals can be 2019) Therefore, they may be more prone than others to engage
perfectionistic (Stoeber & Stober, 2009). Yang and Stoeber (2012) in negative cognitive coping strategies to help manage their body-
developed the Physical Appearance Perfectionism Scale (PAPS) to related distress following biased social comparisons (Cash et al.,

50
S.E. McComb, J.S. Mills Body Image 38 (2021) 49–62

2005). This needs to be empirically tested in the context of social tive coping will improve our understanding of which women are
media and thin ideal images. most at risk of adverse effects of certain types of social media and
Cognitive coping, or cognitive emotional regulation, can be where prevention and intervention should focus.
defined as the cognitions or thoughts that manage and regu-
late one’s emotions, and represent the mental processes used by
1.4. The current study
an individual to cope with a distressing event (Rudolph, Flett, &
Hewitt, 2007). In this context, attention to thin ideal images trig-
The aim of the current study was to extend the literature on
gers for some individuals thoughts, beliefs, and coping processes,
social comparison, body image, and social media in two ways.
some of which may cause a shift towards a more negative self-
First, we sought to investigate the effect of personality on social
evaluation than others. For this reason, social comparison to thin
comparison processes. The current study examined whether trait
ideal images may or may not elicit a more negative body image,
physical appearance perfectionism moderates young women’s
depending on the coping and self-regulatory strategies on which
body image following upwards appearance comparison to idealized
an individual relies. The Cognitive Emotion Regulation Question-
body images on social media. Physical appearance perfection-
naire (CERQ; Garnefski, Kraaij, & Spinhoven, 2001) recognizes four
ism, as reviewed above, is reliably associated with negative body
maladaptive (rumination, catastrophizing, self-blame, and other-
image; we sought to examine whether young women who hold
blame) and five adaptive (acceptance, positive refocusing, putting
extremely high standards for their appearance are especially
it in perspective, positive reappraisal, and refocus on planning)
adversely impacted by comparing themselves to thin and attrac-
cognitive coping strategies. Perfectionism has been consistently
tive women on social media in terms of how they evaluate their
found to be positively related to maladaptive cognitive coping and
own appearance. Based on past research we hypothesized that
to a deficit in adaptive cognitive coping (Castro, Soares, Pereira,
physical appearance perfectionism would moderate the relation-
& Macedo, 2017; Rudolph et al., 2007). While, in general, there
ship between exposure to Instagram images and body image,
has been scarce research on how cognitive coping is related to
such that among women who compared themselves to idealized
body image, some reliable trends have emerged regarding the use
body images, those who were moderate or high on worry about
of rumination and acceptance. Several researchers have experi-
imperfection physical appearance perfectionism subscale would
mentally induced women to feel body dissatisfied after comparing
experience a larger increase in appearance and weight dissatis-
themselves to images of thin fashion models and then instructed
faction and decrease in confidence than those low on physical
participants to engage in acceptance or rumination to cope with
appearance perfectionism. Second, we sought to investigate the
the feelings evoked by the images. Researchers have consistently
role of cognitive coping in social comparison processes. Specifically,
found that engaging in rumination increased body dissatisfaction
we were interested in investigating how women high in physical
following exposure to thin images, while engaging in acceptance
appearance perfectionism (in particular worry about imperfection)
decreased body dissatisfaction (Atkinson & Wade, 2012; Margolis
cognitively cope when they engage in social comparison impacts
& Orsillo, 2016; Naumann, Tuschen-Caffier, Voderholzer, Schafer, &
the way in which mood and body image are affected. The cur-
Svaldi, 2016; Wade, George, & Atkinson, 2009). Comparatively less
rent study examined whether maladaptive (i.e. rumination and
research has been done on how other cognitive coping strategies
catastrophizing) and adaptive (i.e. acceptance and positive refocus-
relate to body image, but some findings have linked catastrophiz-
ing) cognitive coping strategies mediate the potential relationship
ing to poorer body image, and positive refocusing to healthier body
between physical appearance perfectionism and changes in body
image. More specifically, past research has found that individuals
image (i.e. weight and appearance dissatisfaction and confidence)
with an eating disorder, and those without an eating disorder but
caused by comparison to ideal images on social media. Based on
who have high body dissatisfaction, reported greater engagement
past research we hypothesized that trait worry about imperfection
in catastrophizing than healthy controls with low body dissatis-
physical appearance perfectionism would be indirectly related to
faction (Cohen & Petrie, 2005; Moller & Bothma, 2001). Conversely,
greater state weight and appearance dissatisfaction, and less confi-
engagement in positive refocusing, where one intentionally focuses
dence, through the effects of rumination and catastrophizing, and
their attention away from distressing symptoms towards more
also indirectly related to less weight and appearance dissatisfac-
pleasant thoughts, has been linked to increases in weight and
tion, and greater confidence, through the effects of acceptance and
appearance satisfaction after exposure to thin ideal imagery (Wade
positive refocusing.
et al., 2009), and is associated with less binge eating behaviour
(Kelly, Lydecker, & Mazzeo, 2012). Therefore, it may be that the
way in which viewers think about and cognitively process idealized 2. Method
images of others on social media determines whether or not they
feel worse or better about themselves. In other words, maladap- 2.1. Participants
tive cognitive coping may mediate exposure to attractive others
on Instagram and increases in negative body image, while adap- Participants were 142 female psychology undergraduate stu-
tive cognitive coping may mediate exposure to attractive others dents recruited through a research participant pool at York
on Instagram and decreases in negative body image. Based on past University in Toronto, Canada. Inclusion criteria included identify-
research, that links perfectionism with poor cognitive coping and ing as female and being between the ages of 18–25 years; this age
negative body image, it could be expected that those high in phys- range of women was chosen not only for convenience, but because
ical appearance perfectionism would be especially likely to engage women in this age range are the heaviest users of Instagram (Pew
in maladaptive cognitive coping strategies following a threaten- Research Center, 2018). Participant ages ranged from 18 to 24 years
ing upwards appearance comparison, which would increase their (M = 19.06, SD = 1.34). The self-reported ethnic distribution of the
feelings of body dissatisfaction and negative mood. sample was 26.8 % Caucasian, 22.5 % South-Asian, 15.5 % East-Asian,
Putting it all together, there are compelling theoretical and 13.4 % Middle Eastern, 7 % Black/African-Canadian, 6.3 % Other, 4.2
practical reasons to study concurrently the influence of person- % Caribbean, 1.4 % Pacific Islander, .7 % Native, .7 % West Asian, and
ality (i.e., perfectionism) and cognitive processes (i.e., coping and .7 % Hispanic/Latino. Participants’ objective body mass index (BMI
self-regulatory processes) in the context of the effect of social com- = kg/m2 ) scores ranged from 12.3 to 46.5 (M = 21.83, SD = 4.97),
parison to thin ideal images and body image. A comprehensive with no extreme outliers. The mean, median, and mode for BMI
model that looks at the combination of perfectionism and cogni- scores all fell within the “normal” weight range (18.5–24.9) (World

51
S.E. McComb, J.S. Mills Body Image 38 (2021) 49–62

Health Organization, 2019). BMI data was collected for descriptive to stressful events of interest to the researchers, such as interper-
purposes and in order to compare the current study to other study sonal conflict, exam stress, the death of a pet, and negative impacts
samples. to one’s livelihood (Garnefski, Baan, & Kraaij, 2005; Green, Kangas,
& Fairholm, 2018; Sakakibara & Endo, 2016); ours was the first
2.2. Measures study to use the CERQ to investigate how young women respond
to the stress of appearance comparison. For the current study only
2.2.1. Perfectionism four of the nine CERQ strategies are reported on. We used the CERQ
The Physical Appearance Perfectionism Scale (PAPS; Yang & to measure thoughts and feelings following engagement in com-
Stoeber, 2012) was used to measure trait physical appearance per- parison to Instagram models to examine active engagement in the
fectionism. The PAPS is composed of two subscales: worry about following coping strategies: rumination (e.g., I am dwelling upon
imperfection (7 items; e.g., “I worry that my appearance is not the feelings comparing my body to the models has evoked in me; ␣ = .
good enough”) and hope for perfection (5 items; e.g., “I hope my 83), catastrophizing (e.g., I am thinking that how I feel about my body
body shape is perfect”). Participants respond to each item on a five- after comparison to the models is much worse than what others have
point scale from 1-strongly disagree to 5-strongly agree, with higher experienced; ␣ = . 71), acceptance (e.g., I think that I have to accept
scores indicating greater perfectionism. For the current study we my body the way it is; ␣ = . 63), and positive refocusing (e.g., I think
elected to use the worry about imperfection subscale, as this sub- about pleasant experiences that have nothing to do with my body; ␣
scale has been shown to be a more maladaptive type of appearance = . 73). These four subscales were chosen because they seemed the
perfectionism closely related to poor body image, while the hope most relevant to how one would cope with body dissatisfaction,
for perfection subscale was less meaningfully related to body image and seemed like the most reasonable targets for a cognitive coping
(Stoeber & Yang, 2015; Yang & Stoeber, 2012). Internal consistency intervention for body dissatisfaction. In the current study, partici-
in the current study was good (␣ = . 87) pants were asked to indicate how often they actually experienced
each type of thought while looking at the Instagram images. Higher
2.2.2. State body image scores indicate greater usage of that cognitive coping strategy.
To measure state body image, we adapted the procedure by
Tiggemann and McGill (2004), and used three visual analogue 2.3. Materials
scales (VAS) to measure state body image and confidence before
and after viewing the Instagram images. Each scale consisted of Participants in the experimental condition were asked to view
a 100 mm horizontal line with endpoints that were labelled as 26 thin ideal images sourced from two public Instagram accounts
‘none’ and ‘very much’. Participants were asked to report how they of highly attractive thin women. Instagram models were selected
feel “right now” in regard to feelings of appearance dissatisfaction, by five members of the experimenters’ research lab based on how
weight dissatisfaction, and overall feelings of confidence by plac- closely they represented the thin ideal. The images chosen were of
ing a vertical tick mark at the point on the line that best depicts Caucasian women, as this was thought to be most representative
how they are feeling in that moment. These three state items have of the Western beauty ideal circulated in the media according to
commonly been measured using VAS in body image research (e.g. past research (Yan & Bissell, 2014). Both of the models chosen had
Mills, Musto, Williams, & Tiggemann, 2018; Tiggemann & McGill, a large number of followers on Instagram (each had over 11 million
2004). Participants also completed a number of filler VAS items followers at the time of selection), and therefore we reasoned that
to help distract from the study’s aims regarding changes to body the photos were representative of the types of female body image
image. Responses to the scales were measured with a ruler from the that are popular on Instagram. Specific photographs were selected
‘none’ endpoint to the closest millimeter, and therefore responses based on which photos had the most ‘likes’ on the models’ Insta-
could range from 0 to 100, with higher scores indicating stronger gram accounts, and which also clearly showed the bodies and faces
feelings. Scales of this type are commonly used in experimental of the models. For each model, three photos showed the model in
research to reliably assess pre- and post- fluctuations in psycholog- a bikini, four showed the model fully clothed, and four showed a
ical states (Heinberg & Thompson, 1995). Visual analogue scales are close up the woman’s face and shoulders. Both models were pic-
advantageous because they can be completed quickly and previous tured wearing tight and revealing clothing that showcased their
responses are difficult to remember since the line is not numbered, figures.
which minimizes response bias and improves sensitivity to changes We included a control condition where participants were asked
(Frey, 2018). to view landscape painting images sourced from a popular public
Instagram art account. This was included to control for a number
2.2.3. Cognitive emotion regulation of factors. First, the landscape control condition was included to
The Cognitive Emotion Regulation Questionnaire (CERQ; control for the effect of viewing any imagery on Instagram, so that
Garnefski et al., 2001) is a 36 item self-report questionnaire that we could speak to the effect of thin ideal imagery in particular, and
assesses the types of thoughts participants use to regulate their not just time spent on Instagram in general, that results in changes
emotions following a stressful event. The questionnaire has nine to body image. Second, we wanted to control for social compari-
subscales, which are composed of four items each, which represent son processes across conditions so that we could test the effects of
nine different cognitive emotion regulation strategies (rumina- social comparison to thin ideal imagery on changes to body image.
tion, catastrophizing, self-blame, other-blame, acceptance, positive Participants in the landscape (control) condition were instructed
refocusing, positive reappraisal, putting it in perspective, and refo- to engage in a comparison task so as to control for mental engage-
cus on planning). The CERQ is designed to be applicable to any ment and time, but were unable to socially compare their bodies
negative situation and is meant to be adapted for use in a spe- to any target featured in the presented images. We excluded any
cific situation. The instructions specify that you insert the negative body-related imagery in the control condition in case participants
or threatening situation of interest into the instructions and ask automatically engaged in social comparison, without any explicit
participants to indicate how they cognitively responded to the situ- instruction to do so, which would obscure group differences. Land-
ation. We added the situation of interest (i.e. comparing one’s body scape images that had the highest number of likes were chosen
to Instagram models’ bodies) within each item to keep the situa- from the profile as stimuli for the current study. Landscape paint-
tion of reference front-of-mind for participants. Other studies have ings were chosen as a control stimulus because they were deemed
used the CERQ to investigate how participants cognitively respond by the research team to be colourful and attractive, but featured no

52
S.E. McComb, J.S. Mills Body Image 38 (2021) 49–62

people or body images. Pictures of landscapes have been used as ily interested in examining how women cognitively cope with the
control condition stimuli in several past experiments that exam- feelings elicited by social comparison to thin ideal media and the
ined the impact of imagery on body image (Irving, 1990; Michaels, effect on body image and mood, we wanted to control for the poten-
Parent, & Moradi, 2013; Mulgrew, Johnson, Lane, & Katsikitis, 2014). tial confounding effect of whether women were engaging in social
comparison or fantasy thinking, and did so by asking all participants
2.4. Procedure to engage in social comparison. In this way we were able to increase
internal validity by making the experience of women in the exper-
Ethics approval was granted by the York University Research imental condition as similar as possible, except for how women
Ethics Review Board. Eligible students could see and sign up for the chose to cognitively process the thin ideal images presented. This
study through an online experiment management system. Inter- forced comparison allowed us to comment on how cognitive coping
ested participants volunteered to take part in an advertised study following social comparison impacts body image and mood.
on “personality and social media” for partial course credit. The Those in the control condition viewed the images of landscape
study was completed in two parts, one week apart. For Part 1 of the paintings on Instagram for exactly 10 min. While participants
study participants were sent a link and completed online informed browsed the images, they were asked to complete an art critique
consent forms and online measures of trait physical appearance task that required participants to compare the images of the paint-
perfectionism, as well as demographic information such as age and ings on various artistic aspects.
ethnicity. This step was completed in advance of the in-person por- After participants in each condition finished viewing the images
tion of the study in order to reduce demand characteristics and the and had completed their individual task all participants imme-
possibility of priming perfectionism among participants. diately completed another VAS for body image. All participants
For Part 2 of the study, done in-person, participants were ran- then completed the CERQ. While completing the CERQ those in the
domly assigned to either the control (n = 70) or experimental (n = experimental condition were instructed to recall the comparison
72) condition prior to arriving at the lab. Participants were individ- task and to report what thoughts they experienced when dealing
ually tested in a private room and were asked to leave their bags with feelings evoked from comparing themselves to the Instagram
and any electronic devices outside of the testing area to avoid dis- models. To control for time spent on questionnaires and order
traction. All participants read and signed a second informed consent effects of completing questionnaires, we also had the control condi-
form and completed a baseline VAS for body image. Following com- tion complete the CERQ. However, the control condition was asked
pletion of the VAS, participants in the experimental condition were to recall a time when they had gotten in a fight with family member
instructed to browse unaltered Instagram images of the Instagram and to report what thoughts they usually experienced when fight-
models on a desktop computer for exactly 10 min. Participants were ing with a family member. This situation was chosen as the CERQ
asked to browse through the photos sequentially first, and were reference event because everyone could be expected to relate to
told they could then return to look at any image they wanted if the this scenario and because it was body-neutral. Upon completing
ten minutes had not expired. While participants viewed the images the questionnaires all participants were thanked for their time and
of each model they were asked to complete a body comparison task debriefed. All participants had their height and weight measured
that required participants to compare the size of various parts of by the experimenter backwards on a balance beam scale prior to
their own bodies to the size of the Instagram model’s body parts. leaving the lab; this information was used to calculate participants’
Participants were instructed to look at the models’ photos and to BMI.
rate if their thighs, arms, butt, waist, hips, biceps, breasts, legs, and
stomach were much smaller, smaller, about the same size, larger, or 2.5. Data analysis
much larger than the corresponding body parts of the model. They
were also asked to indicate if their face and overall physical appear- Statistical analyses were conducted using SPSS version 25. As
ance was much more attractive, slightly more attractive, about the a manipulation check to ensure that the images selected for the
same level of attractiveness, less attractive, or much less attractive experimental condition were more threatening to body image than
than the Instagram models’. Participants completed this task for the images selected for the control condition, a mixed model uni-
each of the two models, which took exactly 10 min (five minutes variate analysis of variance (ANOVA) was used to test whether
for each model). Forced comparison of this type has been used in pre and post scores for weight dissatisfaction, appearance dis-
previous studies that have investigated the impact of comparison satisfaction, and confidence were dependent upon experimental
to thin ideal imagery on body image (Cattarin, Thompson, Thomas, condition. Time (Time 1- Time 2) was the within-subject factor,
& Williams, 2000; Mills et al., 2002; Tiggemann & McGill, 2004; and experimental condition (experimental or control group) was
Tiggemann & Polivy, 2010; Tiggemann, Polivy, & Hargreaves, 2009; the between-subjects factor. Statistically significant interactions
Tiggemann & Slater, 2004). between time and condition were followed up with paired samples
A forced comparison task was included in the current study t-tests with Bonferroni correction (p< 0.016).
for several reasons. First, this task was intended to induce active To test whether physical appearance perfectionism-worry
appearance-based comparison to the Instagram models, so that about imperfection moderated the effect of condition on post-
participants were actively engaging with the photos, rather than exposure scores moderation analyses were conducted using the
passively viewing them. Second, we used forced comparison so PROCESS SPSS macro version 3.2 (Hayes, 2017). We tested model
that we could later examine how women cognitively cope (mea- 1, which includes one outcome variable, one predictor, one moder-
sured with the CERQ, see below) with body dissatisfaction when ator, and room for covariates. We used condition as the predictor
they perceive themselves as less attractive than models featured in variable, physical appearance perfectionism worry about imper-
thin ideal imagery. Third, we wanted to control as best we could fection scores as the moderator, post-exposure confidence, weight
what kind of psychological mechanisms participants were engag- dissatisfaction, or appearance dissatisfaction scores as the out-
ing in when viewing the images. Past research has shown that when come variable, and controlled for the relevant pre-exposure score.
exposed to media images some women engage in social comparison As recommended by Aiken and West (1991) physical appearance
which most often results in negative body image and mood, while comparison scores were centered around the mean to reduce multi-
other women engage in fantasy thinking (versus social compari- collinearity. Significant interactions were followed up with simple
son) which can result in positive effects to body image and mood slopes analysis. The simple slopes analyses tested whether exper-
(Mills et al., 2002; Tiggemann et al., 2009). Because we were primar- imental condition had an effect on post-exposure body image

53
S.E. McComb, J.S. Mills Body Image 38 (2021) 49–62

Note. Intercorrelations below the diagonal reflect scores for the experimental condition, and intercorrelations above the diagonal represent scores for the control condition; PAPS-WAI = Physical appearance perfectionism
24.78

28.97
25.71
30.15

19.80

20.28
5.57
3.28

3.21
3.02
3.02
SD
Control

43.53
45.76
53.47
38.91

53.93
23.53
12.61

12.24
40.76

8.67

8.84
M

21.99

25.35
23.22
29.30
23.60

30.12

5.44
3.12
2.64
2.64
2.77
SD
Experimental

35.57
36.64

46.63
45.67
21.17

13.95
11.14
54.50
47.03

11.04
8.06
M
Fig. 1. Proposed relationships between physical appearance perfectionism, CERQ

−.187

−.133

−.153
−.029

−.110
subscales, and body image outcome variables.

−.010
.132

.127

.114
.060
11

—
outcomes at low (1 SD below the mean), medium (mean), and high
(1 SD above the mean) levels of physical appearance perfectionism.

.498**
.385**
−.072

−.130

.257*
.138
.048

.093
.105

.108
To test whether cognitive coping mediated the effects of

10

—
appearance comparison to Instagram models, mediation regression
analyses were conducted on experimental condition data only (n =

−.406**
−.242*
.581**
72), using the PROCESS SPSS macro version 3.2 (Hayes, 2017). We

−.159

−.184
.172
.181

.212
.104
.087
tested model 4, which includes one outcome variable, one predic-

—
9
tor, and room for mediators. As shown in Fig. 1, we used physical
appearance perfectionism-worry about imperfection as the predic-

−.315**

−.332**

−.482**
.329**

.701**
tor, rumination, catastrophizing, acceptance, or positive refocusing

.213

.161
.107

.092

.007
as the mediator, and weight dissatisfaction, appearance dissatis-

—
8
faction, or confidence as the outcome variable. Significant indirect
effects were determined by examining 95 % confidence intervals to

−.413**
−.670**
.445**

.353**
.421**

.477**
.510**

.430**

−.215
−.031
ensure the interval did not contain the value zero; intervals that do
not contain the value zero would indicate that some indirect effect

—
7
between appearance perfectionism and body image exists that is
different from zero. An a priori power analysis was conducted using

−.382**

−.518**
−.510**
−.291*
−.297*

−.287*
−.300*
.742**

.478**
G*Power (Faul, Erdfelder, Lang, & Buchner, 2007) to estimate the

.077
—
sample size needed for the mediation analyses (conducted within
6

the experimental group) to achieve adequate power. Using an alpha

level of .01, a medium effect size, and power estimate of .80, the
−.424**

−.455**

−.319**
.553**
.898**

.669**

.561**
.486**
.528**
−.134
Intercorrelations for Physical Appearance Perfectionism, CERQ Strategy, and Body Image Outcomes.

recommended sample size was 58 participants, which we achieved.

—
5

3. Results
−.326**

−.346**

−.281*
−.251*
.613**

.686**

.469**
.424**
.890**

.509**

3.1. Preliminary analyses

—
4

Inspection of histograms, skewness, and kurtosis suggested that

−.391**
−.485**

−.315**
−.279**

−.305**

all of the variables were normally distributed. There were no statis-

.704**

−.182
−.190

.291*
.234

tically significant outliers (±3 SD) among the predictor, mediator,

—
3

or outcome variables.
A series of one-way ANOVAs showed that those in the exper-
−.454**

−.398**

imental and control conditions did not differ in regard to age, F

.657**

.691**

.364**
.580**

.306**
.402**
−.164
−.138

scale-Worry About Imperfection score. * p<0.05; **p < 0.01.

(1,140) = .89, p = .35, BMI, F (1,138) = .69, p = .41, or baseline mea-

—
2

sures of confidence, F (1,140) = .09, p = .77, or weight dissatisfaction,

F (1,140) = 2.55, p = .11. They did however differ in baseline levels
−.333***

of appearance dissatisfaction, F (1,140) = 5.04, p = .03, with those in

.695**

.743**
.436**
−.188

−.221
−.095
.217

.224
.083

the control condition reporting higher appearance dissatisfaction

—
1

than those in the experimental condition (see Table 1). Together,

these results suggest that randomization to condition resulted in
5. T2 Appearance Dissatisfaction
2. T1 Appearance dissatisfaction

reasonably equivalent groups on all but one of the participant

1. T1 Weight dissatisfaction

4. T2 Weight dissatisfaction

characteristics and baseline states. However, those in the control

group started the experiment feeling slightly worse about their
11. Positive refocusing

appearance than did those in the experimental group. Despite there

9. Catastrophizing

being baseline differences in appearance dissatisfaction between

3. T1 Confidence

6. T2 Confidence

10. Acceptance
8. Rumination

the experimental groups, mixed ANOVAs were still conducted to

7. PAPS-WAI

examine changes in weight dissatisfaction, appearance dissatisfac-

Variable

tion, and confidence over time between the conditions. However,

Table 1

because there were baseline differences in appearance dissatis-

faction between the experimental conditions, we decided to also

54
S.E. McComb, J.S. Mills Body Image 38 (2021) 49–62

Table 2
Moderated regression results and simple slopes analysis at low, medium, and high levels of physical appearance perfectionism for post-exposure weight dissatisfaction,
appearance dissatisfaction and confidence.

B SE t p

State weight dissatisfaction

Condition 16.23 2.76 5.87 .000
Physical appearance perfectionism .84 .26 3.21 .001
Condition x physical appearance perfectionism 1.73 .50 3.48 .000
Low physical appearance perfectionism 6.53 3.91 1.67 .097
Moderate physical appearance perfectionism 16.23 2.76 5.87 .000
High physical appearance perfectionism 25.93 3.94 6.58 .000

State appearance dissatisfaction

Condition 14.77 2.54 5.82 .000
Physical appearance perfectionism .80 .25 3.19 .001
Condition x physical appearance perfectionism 1.20 .45 2.66 .008
Low physical appearance perfectionism 8.01 3.58 2.24 .026
Moderate physical appearance perfectionism 14.77 2.54 5.82 .000
High physical appearance perfectionism 21.53 3.60 5.97 .000

State confidence
Condition −9.93 2.53 −3.93 .000
Physical appearance perfectionism −.40 .25 −1.61 .109
Condition x physical appearance perfectionism −1.33 .46 −2.87 .004
Low physical appearance perfectionism −2.46 3.66 −.67 .502
Moderate physical appearance perfectionism −9.93 2.53 −3.93 .000
High physical appearance perfectionism −17.40 3.60 −4.84 .000

run ANCOVAs to test for between group differences in Time 2 26.15, p < .001, ␩p 2 = .158, and confidence scores, F (1, 139) = 12.56,
appearance dissatisfaction, weight dissatisfaction, and confidence, p = .001, ␩p 2 = .083, after controlling for the relevant pre-exposure
while controlling for Time 1 scores as covariates, to ensure our scores. The effects of weight and appearance dissatisfaction repre-
findings were not biased by baseline differences. No adjustments sented large effect sizes, while the effect of confidence represented
were made to the mediation analyses involving appearance dis- a medium effect. Pairwise comparisons revealed that those in the
satisfaction since they involved only the experimental group at experimental group felt significantly more weight dissatisfaction
post-manipulation. (p < .001, mean difference = 8.11) and appearance dissatisfaction (p
Bivariate correlations, means, and standard deviations were < .001, mean difference = 5.87), and less confidence (p < .001, mean
computed between physical appearance perfectionism scores, difference = 8.26), than those in the control group after comparison
CERQ subscales, and post-induction measures of the VAS outcomes to the Instagram images. Therefore, the manipulation check indi-
for the experimental group, and can be found in Table 1. In sum, cates that exposure to the experimental condition thin ideal images
the variables were related to each other in the expected directions was more threatening to body image than exposure to the control
and in theoretically meaningful ways. condition landscape images.

3.3. Physical appearance perfectionism as a moderator

3.2. Changes to body image as a function of condition

3.3.1. Weight dissatisfaction

Results revealed that significant 2-way interactions (condition
As shown in Table 2, we tested whether levels of physical
x time) were found for confidence, F(1, 140) = 11.96, p = .001, ␩p 2 =
appearance perfectionism-worry about imperfection moderated
.079, weight dissatisfaction, F(1, 140) = 28.11, p < .001, ␩p 2 = .167,
the relationship between experimental condition and post-
and appearance dissatisfaction, F(1, 140) = 31.50, p < .001, ␩p 2 =
exposure weight dissatisfaction. Main effects of both condition
.184, meaning that the two experimental conditions were not equal
and physical appearance perfectionism on post-exposure weight
with respect to changes in those outcome variables from Time 1 to
dissatisfaction were observed, however these main effects were
Time 2. These represented large effects for appearance and weight
qualified by a significant interaction. As shown in Fig. 2, phys-
dissatisfaction and a medium effect for confidence.
ical appearance perfectionism had no impact on post-exposure
Significant interactions were followed up with paired samples
weight dissatisfaction scores in the control condition. However, as
t-tests. Results showed that those in the experimental condition
predicted, within the experimental condition those with medium
experienced increases in both weight dissatisfaction, t (71) = 4.56,
or high levels of physical appearance perfectionism experienced
p < .001, Cohen’s d = .537 (mean difference of 11.46), and appear-
greater amounts of weight dissatisfaction after exposure to thin
ance dissatisfaction, t (71) = 4.39, p < .001, Cohen’s d = .518 (mean
ideal images, than those who were low on physical appearance
difference of 9.98) from Time 1 to Time 2, while those in the control
perfectionism.
condition experienced decreases in weight dissatisfaction, t (69) =
2.77, p = .007, Cohen’s d = -.333 (mean difference of -4.62), and
appearance dissatisfaction, t (69) = 3.67, p < .001, Cohen’s d = .438 3.3.2. Appearance dissatisfaction
(mean difference of -5.00). Further, those in the experimental con- As shown in Table 2, we tested whether levels of physical
dition experienced significant decreases in confidence, t (71) = 4.30, appearance perfectionism-worry about imperfection moderated
p < .001, Cohen’s d = .507 (mean difference of -8.83) from Time the relationship between experimental condition and post-
1 to Time 2, while those in the control condition experienced no exposure appearance dissatisfaction. Main effects of both condition
statistically significant changes in confidence. and physical appearance perfectionism on post-exposure appear-
Further, the ANCOVAs revealed that there were significant group ance dissatisfaction were observed, however these main effects
differences in post-exposure weight dissatisfaction, F (1, 139) = were qualified by a significant interaction. As shown in Fig. 3, phys-
24.94, p < .001, ␩p 2 = .152, appearance dissatisfaction, F (1, 139) = ical appearance perfectionism had no impact on post-exposure

55
S.E. McComb, J.S. Mills Body Image 38 (2021) 49–62

Fig. 2. Weight dissatisfaction as a function of condition and level of physical appearance perfectionism.

Fig. 3. Appearance dissatisfaction as a function of condition and level of physical appearance perfectionism.

appearance dissatisfaction scores in the control condition. How- sure to thin ideal images, than those who were low on physical
ever, as predicted, within the experimental condition those with appearance perfectionism.
medium or high levels of physical appearance perfectionism
experienced greater amounts of appearance dissatisfaction after
3.4. Mediation analyses
exposure to thin ideal images, than those who were low on physical
appearance perfectionism.
3.4.1. Physical appearance perfectionism, cognitive coping, and
weight dissatisfaction
As shown in Table 3, we tested whether physical appearance
3.3.3. Confidence perfectionism was directly related to weight dissatisfaction, and
As shown in Table 2, we tested whether levels of physical also indirectly related through the effect of rumination, catastro-
appearance perfectionism-worry about imperfection moderated phizing, acceptance, or positive refocusing1 . Because we tested
the relationship between experimental condition and post- several mediation models the alpha level was set to a conservative
exposure confidence. There was not a significant main effect of .01 to control for type I error.
physical appearance perfectionism on state confidence. There was
a significant main effect of condition, however this was qualified
by a significant interaction. As shown in Fig. 4, physical appearance
1
perfectionism had no impact on post-exposure confidence scores The other five subscales of the CERQ (self-blame, other-blame, putting it in
perspective, positive reappraisal, and refocus on planning) were also all tested as
in the control condition. However, as predicted, within the exper- possible mediators between physical appearance perfectionism and weight and
imental condition those with medium or high levels of physical appearance dissatisfaction and confidence, and none were statistically significant
appearance perfectionism experienced less confidence after expo- mediators.

56
S.E. McComb, J.S. Mills Body Image 38 (2021) 49–62

Table 3
Mediation Analyses Direct and Indirect Effects.

Weight Dissatisfaction

Regression paths b 95 % CI

Rumination
Mediation a path (PAPS-WAI on rumination) .27*** [.15, .40]
Mediation b path (Rumination on weight dissatisfaction) 2.49* [.25, 4.74]
Direct effect, c’ (PAPS-WAI on weight dissatisfaction) 1.92** [.62, 3.21]
Indirect effect with 95 % CI .68 [.01, 1.40]
Catastrophizing
Mediation a path (PAPS-WAI on catastrophizing) .21*** [.10, .31]
Mediation b path (Catastrophizing on weight dissatisfaction) 4.31*** [1.83, 6.78]
Direct effect, c’ (PAPS-WAI on weight dissatisfaction) 1.70** [.50, 2.90]
Indirect effect with 95 % CI .89 [.21, 1.74]
Acceptance
Mediation a path (PAPS-WAI on acceptance) −.02 [−.13, .10]
Mediation b path (Acceptance on weight dissatisfaction) −3.04** [−5.35, −.73]
Direct effect, c’ (PAPS-WAI on weight dissatisfaction) 2.56*** [1.44, 3.67]
Indirect effect with 95 % CI .05 [−.31, .50]
Positive refocusing
Mediation a path (PAPS-WAI on positive refocusing) −.11 [−.23, .01]
Mediation b path (Positive refocusing on weight dissatisfaction) −.1.71 [−4.04, .61]
Direct effect, c’ (PAPS-WAI on weight dissatisfaction) 2.41*** [1.23, 3.60]
Indirect effect with 95 % CI .19 [−.06, .72]

Appearance Dissatisfaction

Regression paths b 95 % CI

Rumination
Mediation a path (PAPS-WAI on rumination) .27*** [.15, .40]
Mediation b path (Rumination on appearance dissatisfaction) 2.29** [.54, 4.04]
Direct effect, c’ (PAPS-WAI on appearance dissatisfaction) 1.99*** [.98, 3.00]
Indirect effect with 95 % CI .63 [.09, 1.04]
Catastrophizing
Mediation a path (PAPS-WAI on catastrophizing) .21*** [.10, .31]
Mediation b path (Catastrophizing on appearance dissatisfaction) 3.38*** [1.43, 5.33]
Direct effect, c’ (PAPS-WAI on appearance dissatisfaction) 1.91*** [.97, 2.86]
Indirect effect with 95 % CI .70 [.19, 1.56]
Acceptance
Mediation a path (PAPS-WAI on acceptance) −.02 [−.11, .06]
Mediation b path (Acceptance on appearance dissatisfaction) −1.12 [−3.01, .77]
Direct effect, c’ (PAPS-WAI on appearance dissatisfaction) 2.60*** [1.68, 3.52]
Indirect effect with 95 % CI .02 [−.16, .23]
Positive refocusing
Mediation a path (PAPS-WAI on positive refocusing) −.11 [−.23, .01]
Mediation b path (Positive refocusing on appearance dissatisfaction) −1.90* [−3.71, −.10]
Direct effect, c’ (PAPS-WAI on appearance dissatisfaction) 2.41*** [1.49, 3.33]
Indirect effect with 95 % CI .21 [−.06, .56]

Confidence

Regression paths b 95 % CI

Rumination
Mediation a path (PAPS-WAI on rumination) .27*** [.15, .40]
Mediation b path (Rumination on confidence) −3.15*** [−4.87, −1.44]
Direct effect, c’ (PAPS-WAI on confidence) −.76 [−1.75, .22]
Indirect effect with 95 % CI −.87 [−1.56, −.34]
Catastrophizing
Mediation a path (PAPS-WAI on catastrophizing) .21*** [.10, .31]
Mediation b path (Catastrophizing on confidence) −3.82*** [−5.78, −1.86]
Direct effect, c’ (PAPS-WAI on confidence) −.83 [−1.78, .12]
Indirect effect with 95 % CI −.83 [−1.45, −.26]
Acceptance
Mediation a path (PAPS-WAI on acceptance) −.02 [−.13, .10]
Mediation b path (Acceptance on confidence) .58 [−1.37, 2.53]
Direct effect, c’ (PAPS-WAI on confidence) −1.62** [−2.57, −.67]
Indirect effect with 95 % CI −.01 [−.21, .10]
Positive refocusing
Mediation a path (PAPS-WAI on positive refocusing) −.11 [−.23, .01]
Mediation b path (Positive refocusing on confidence) 3.48*** [1.76, 5.19]
Direct effect, c’ (PAPS-WAI on confidence) −1.25** [−2.12, −.38]
Indirect effect with 95 % CI −.38 [−2.12, −.38]

Note. PAPS-WAI = physical appearance perfectionism scale-worry about imperfection. *p < .05. **p < .01. ***p < .001.

57
S.E. McComb, J.S. Mills Body Image 38 (2021) 49–62

Fig. 4. Confidence as a function of condition and level of physical appearance perfectionism.

Physical appearance perfectionism was found to be directly 3.4.3. Physical appearance perfectionism, cognitive coping, and
and positively related to weight dissatisfaction in all four models confidence
that were tested. As predicted, physical appearance perfection- As shown in Table 3 we tested whether physical appearance per-
ism was indirectly associated with weight dissatisfaction through fectionism was directly related to confidence, and also indirectly
the effects of both rumination and catastrophizing. Contrary to related through the effect of rumination, catastrophizing, accep-
hypotheses, physical appearance perfectionism was not indirectly tance, or positive refocusing. Because we tested several mediation
associated with weight dissatisfaction through the effects of either models the alpha level was set to a conservative .01 to control for
acceptance or positive refocusing. Additionally, physical appear- type I error.
ance perfectionism was found to significantly and positively predict Physical appearance perfectionism was, for the most part, not
rumination and catastrophizing, but did not significantly predict found to be directly related to confidence; it was only directly
acceptance or positive refocusing. Furthermore, catastrophizing and negatively related to confidence in one of the four models
positively predicted weight dissatisfaction, while acceptance neg- tested. However, as predicted, physical appearance perfectionism
atively predicted weight dissatisfaction. Rumination marginally was indirectly and positively associated with confidence through
predicted weight dissatisfaction (p = .03) based on our conserva- the effects of rumination and catastrophizing, and indirectly and
tive alpha level. Positive refocusing was not significantly related to negatively associated with confidence through the effect of positive
weight dissatisfaction. refocusing. Contrary to hypotheses, physical appearance perfec-
tionism was not indirectly associated with confidence through the
effect of acceptance. Additionally, physical appearance perfection-
ism was found to significantly and positively predict rumination
3.4.2. Physical appearance perfectionism, cognitive coping, and and catastrophizing, but did not significantly predict acceptance or
appearance dissatisfaction positive refocusing. Furthermore, rumination and catastrophizing
As shown in Table 3 we tested whether physical appearance negatively predicted confidence, while positive refocusing posi-
perfectionism was directly related to appearance dissatisfaction, tively predicted confidence.
and also indirectly related through the effect of rumination, catas-
trophizing, acceptance, or positive refocusing. Because we tested 4. Discussion
several mediation models the alpha level was set to a conservative
.01 to control for type I error. The purpose of the current study was to extend the litera-
Physical appearance perfectionism was found to be directly ture on social comparison and body image in the context of social
and positively related to appearance dissatisfaction in all four media in two important ways. Guided by cognitive models of body
models that were tested. As predicted, physical appearance perfec- image, we aimed to investigate the path between personality and
tionism was indirectly associated with appearance dissatisfaction body dissatisfaction following exposure to thin ideal images. More
through the effects of both rumination and catastrophizing. Con- specifically, we were interested in investigating whether trait phys-
trary to hypotheses, physical appearance perfectionism was not ical appearance perfectionism moderates young women’s body
indirectly associated with appearance dissatisfaction through the image following upwards comparison to thin ideal images on social
effect of acceptance or positive refocusing. Additionally, phys- media, so as to determine whether young women who hold high
ical appearance perfectionism was found to significantly and standards for their appearance are more adversely impacted by
positively predict rumination and catastrophizing, but did not appearance comparisons made on social media. Second, we aimed
significantly predict acceptance or positive refocusing. Further- to investigate whether the way women think about the negative
more, rumination and catastrophizing both positively predicted contrast that results from the comparison to idealized social media
appearance dissatisfaction. Positive refocusing marginally nega- images can explain why women who are perfectionistic about their
tively predicted appearance dissatisfaction (p = .03). Acceptance appearance may experience worsened body image after the com-
was not significantly associated with appearance dissatisfac- parison. In particular, we were interested in examining whether
tion. the cognitive processes used to cope with a perceived negative

58
S.E. McComb, J.S. Mills Body Image 38 (2021) 49–62

contrast following social comparison to idealized images could greater weight and appearance dissatisfaction and less confidence
explain why women who were highly perfectionistic about their following the comparison to the Instagram models, relative to those
appearance experience more negative body image following the who were relatively unconcerned about having a “flawed” appear-
comparison. This was an important extension of social compari- ance. It seems that the comparison to idealized body types, and
son theory (Festinger, 1954), which assumes that dissatisfaction the contrast between how the models’ bodies were perceived rel-
because of a negative contrast following social comparison is an ative to how young women perceived their own bodies, was more
automatic response, rather than a process that is mediated by how detrimental for those who prize having a “perfect” appearance.
one cognitively appraises that negative contrast. The findings gen- These results are consistent with past literature that has found that
erally supported our hypotheses and the major findings are clear. physical appearance perfectionism is positively associated with
Our first major finding supported social comparison theory appearance and body shape disturbance, body area dissatisfaction,
(Festinger, 1954) and the tripartite model of body dissatisfaction and poor appearance self-esteem (Yang & Stoeber, 2012; Yang et al.,
(Rodgers et al., 2015) and showed that upwards appearance com- 2017). These findings contribute novel information in identifying
parison to thin deal images in a social media environment resulted vulnerable subgroups of women who stand to be the most nega-
in increased weight and appearance dissatisfaction and less confi- tively impacted by comparison to thin ideal images on Instagram,
dence, relative to baseline levels and to the control condition. The which has been very sparsely done in the existing research.
second major finding was that personality played a significant role The third major finding was that the way in which young
in how the effects of the upwards comparison were experienced, women cognitively cope with the negative contrast that results
with those who were moderate to high in physical appearance from comparison to idealized body types can explain how phys-
perfectionism experiencing greater weight and appearance dissat- ical appearance perfectionism is related to resulting body image.
isfaction and less confidence than those who were low on physical We found that engagement in rumination and catastrophizing
appearance perfectionism. The third major finding was that how explained why physical appearance perfectionism was associated
young women cognitively cope with or appraise the negative con- with greater weight and appearance dissatisfaction and lower con-
trast that results from the comparison can explain how physical fidence. It appears that women who are perfectionistic about their
appearance perfectionism impacts body image. More specifically, appearance have a greater tendency to continually dwell upon
engagement in the strategies of rumination and catastrophizing the shape and appearance of their body and how it feels to not
following appearance comparison to thin ideal images mediated have a “perfect” body, and to also catastrophize about how horri-
the relationship between physical appearance perfectionism and ble that experience and feeling are, which then results in lowered
weight dissatisfaction, appearance dissatisfaction, and confidence. confidence and greater weight and appearance dissatisfaction. Our
Further, engagement in the adaptive coping strategy, positive refo- results are consistent with other research which has found that the
cusing, mediated the relationship between physical appearance relationships between general perfectionism and eating disorder
perfectionism and confidence. Engagement in acceptance did not symptoms has been mediated by rumination and catastrophizing
mediate the relationship between physical appearance perfection- (Riviere & Douilliez, 2017), but further extend the literature to show
ism and any of the body image outcome variables. that similar relationships also exist between physical appearance
The first major finding of the current study supports social com- perfectionism and body dissatisfaction and confidence. In particu-
parison theory and the tripartite model and demonstrates that lar, participants who ruminated after comparing themselves to the
comparison to idealized body types on Instagram that are perceived models endorsed thoughts such as, “I dwell upon the feelings compar-
to be superior in appearance, leads to greater weight and appear- ing my body to the models’ has evoked in me”, and “I am preoccupied
ance dissatisfaction and lowered confidence. The results suggest with what I think and feel about my body after comparing it to the
that once young women compare the size and appearance of their models’”. Alternatively, participants who catastrophized after the
body to idealized others, women perceive their own bodies to be comparison to the models, endorsed thoughts such as, “I keep think-
inferior. Perception of a discrepancy between their body and the ing about how terrible my body is compared to the models’” and “”I
models’ bodies invokes adverse feelings, in the form of increased am thinking that how I feel after comparing myself to the models is the
weight and appearance dissatisfaction and lowered feelings of con- worst feeling a person can experience”, and “I continually think about
fidence. These results are a replication of past studies, which have how horrible the comparison of my body to the models’ has been.”
found that exposure and social comparison to idealized body types In sum, the current study found that young women high in phys-
on Instagram is associated with poorer body image and confidence ical appearance perfectionism dwell upon or become preoccupied
amongst young women (Brown & Tiggemann, 2016; Fardouly & with the negative feelings that are evoked by comparing their bod-
Holland, 2018; Hogue & Mills, 2019). Interestingly, during debrief- ies to the models’ bodies, which in turn causes them to report more
ing at the end of the experiment, the majority of participants in weight and appearance dissatisfaction and less self-confidence.
the experimental condition indicated that they did not believe that Our results also showed that engagement in positive refocusing
they had been negatively affected by the Instagram images that mediated the relationship between physical appearance perfec-
they saw, and stated that the images had not made them feel tionism and confidence. We found that women who are higher
worse about themselves because they were used to seeing these in physical appearance perfectionism are less likely to engage in
types of images every day on their own Instagram feeds. It may be positive refocusing, which results in lower confidence. It appears
so normalized in Western society to see idealized images in vari- that women who are perfectionistic about their appearance are
ous forms of the media and to routinely engage in comparison to less likely to distract themselves from negative body related feel-
those images, that young women may lack insight into how these ings by thinking about something pleasant that has nothing to
images negatively affect their self-image. Participants’ feedback to do with their body, and this then negatively impacts their confi-
the experimenter may represent defensiveness or conscious min- dence. Perhaps this is because women who are perfectionistic about
imization of the effects of social media exposure, or may reveal their appearance are too overwhelmed by ruminative and catas-
a true lack of insight that young women have about the negative trophizing thoughts, as our results suggest, to be able to distract
effects of such exposure. themselves from these negative thoughts. However, our results
The second major finding was that personality played a signifi- suggest that when women were able to engage in positive refo-
cant role in how comparison to an idealized other was experienced. cusing, this coping strategy was associated with less appearance
More specifically, we found that young women who were moder- dissatisfaction and higher confidence. Positive refocusing appears
ately to highly perfectionistic about their appearance experienced to be a useful cognitive coping strategy to help young women cope

59
S.E. McComb, J.S. Mills Body Image 38 (2021) 49–62

with negative feelings about their body, at least in the short term. ing and comparing oneself to attractive models on social media
This finding is in line with past research which has found that active resulted in lower confidence and increased weight and appear-
engagement in positive refocusing distraction techniques is associ- ance dissatisfaction amongst young women, especially if women
ated with increases in weight and appearance satisfaction following were moderate-high on physical appearance perfectionism and if
exposure to thin ideal imagery (Wade et al., 2009). they engaged in rumination or catastrophizing. This is concerning
The final major finding was that, contrary to our predic- given that 60 % of Instagram users visit the platform every day
tions, engagement in acceptance did not mediate the relationship and Instagram use is strongly associated with social comparison
between physical appearance perfectionism and weight dissatis- (Stapleton, Luiz, & Chatwin, 2017). However, our results suggest
faction, appearance dissatisfaction, or confidence. Past research has that changing cognitive coping patterns could change the relation-
found that active engagement in acceptance following exposure ship between physical appearance perfectionism and body image.
to thin ideal imagery was linked to decreased body dissatisfac- Clinical interventions should target reducing these negative cogni-
tion (Atkinson & Wade, 2012; Margolis & Orsillo, 2016; Naumann tive coping patterns to prevent negative body image made worse
et al., 2016; Wade et al., 2009), which was not found in the current by upwards appearance comparisons on Instagram. Clinical inter-
study. However, the previous studies that found positive effects of ventions could also focus on teaching young women how to engage
acceptance on body image actively trained participants to engage in in positive refocusing, given that this coping strategy was associ-
metacognitive or experiential acceptance, which emphasize expe- ated with lower levels of appearance dissatisfaction and greater
riencing without judgement and acknowledging, allowing, and confidence. For example, teaching young women how to positively
releasing experience (Atkinson & Wade, 2012; Wade et al., 2009). refocus would entail encouraging women to distract themselves
Whereas, the type of acceptance measured by the CERQ refers to from negative body-related thoughts by instead thinking of pleas-
thoughts of resignation to what has happened (e.g. I think that I have ant things that have nothing to do with their bodies. Teaching these
to accept [my body the way it is]; I think that I cannot change any- cognitive coping strategies could be relatively easy to do, and past
thing about [my body]), which is quite different from experiential research has shown that teaching adaptive cognitive coping strate-
acceptance. The current study did not find this type of resignation- gies is associated with less body dissatisfaction (Atkinson & Wade,
based acceptance to be a helpful strategy for overcoming feelings 2012; Naumann et al., 2016; Wade et al., 2009).
of appearance dissatisfaction or for boosting confidence following
comparison to idealized body images. 4.2. Strengths, limitations, and future research
Taken together, the current findings suggest that young women
who are high in physical appearance perfectionism are especially It was a strength of the current study that we experimentally
sensitive to upwards appearance comparisons to idealized body demonstrated that social comparison to idealized body types on
images on Instagram, and are likely to find those comparisons dis- Instagram causes negative changes in body image, rather than
tressing because of the ways in which they cognitively process or investigating this relationship through correlational methods. An
cope with them. This was the first study to examine how physi- experimental design allowed us to test the causal relationship
cal appearance perfectionism is related to cognitive coping and to between social comparison to thin ideal imagery and resulting body
changes in body image amongst young women following appear- image, rather than just an association between variables, and to
ance comparison to idealized body images on Instagram. This is minimize the influence of myriad confounds (i.e. age, gender, social
also the first study to examine cognitive processes that explain why media platform, likes and comments on thin ideal imagery) that
appearance comparison to idealized body images on Instagram is cannot be controlled for in a correlational design. An experimental
associated with worsened body image. These findings add to our design also allowed us to assess for changes to body image in the
knowledge about the effects of personality on body image follow- moment by inducing social comparison so as to avoid a reliance
ing comparison to idealized body images on Instagram, which has on retroactive experiences of social comparison and recall biases
been sparsely researched. These findings also extend past research of state body image, which is a limitation of many correlational
by identifying the specific types of thoughts that women experience research designs. This was also the first study to investigate how
during social comparison to idealized body images on Instagram, physical appearance perfectionism is related to cognitive coping,
which heighten their feelings of body dissatisfaction. These find- which contributes to the sparse research on physical appearance
ings are an important, and long overdue, contribution to the social perfectionism. It was also a strength that the study had a relatively
comparison literature. Our findings demonstrate that awareness of large and ethnically diverse sample.
a negative contrast following an upwards comparison do not just Despite these strengths, there were several limitations that
automatically result in poor body image. Instead our findings high- should be noted and that can guide future research. First, our sam-
light that it is how one cognitively copes with or appraises this ple only included young women between the ages of 18–25 years,
negative contrast, whether in an adaptive or maladaptive way, that and therefore results cannot be generalized to women of other
determines resulting levels of weight and appearance dissatisfac- ages or to men. Second, the study took place in a laboratory set-
tion and confidence. Further our results highlight that there are ting and therefore participants may have attended to the images
individual differences in how negative contrasts from social com- differently than they would have in their own environment. Third,
parisons are cognitively coped with, and that some individuals are cognitive coping was assessed retrospectively, albeit very briefly,
more vulnerable to the negative effects of social comparisons than and therefore it is somewhat possible responses were influenced
others, such as those high on physical appearance perfectionism. by memory biases. It is also possible that women may not con-
These results support the cognitive model of body dissatisfaction, sciously process images in this way in the real world, but only
which theorizes that certain personality traits, such as perfection- became cognizant of their coping processes when their attention
ism, and activating events both influence cognitive processing and was explicitly drawn to them. Also, the acceptance subscale of
internal dialogues of that event, which then influence resulting the CERQ demonstrated somewhat poor internal reliability in the
body image emotions (Lewis-Smith et al., 2019). current study. Our use of the CERQ to measure thoughts about com-
paring oneself to idealized body image on social media was novel,
4.1. Implications therefore more research needs to be done to support the use of
the measure for this purpose. Fourth, it was also a limitation that
These findings have clinical implications for the treatment and the images presented were of Caucasian women only. The current
prevention of body dissatisfaction. Our findings reflected that view- study had an ethnically diverse sample, and so many participants

60
S.E. McComb, J.S. Mills Body Image 38 (2021) 49–62

viewed images of, and made social comparisons to, a woman of CRediT authorship contribution statement
a different race from them. This may have increased the level of
discrepancy between racialized women and their target of com- Sarah E. McComb: Conceptualization, Methodology, Formal
parison in the current study in a way that was not experienced analysis, Investigation, Data curation, Writing - original draft, Visu-
by Caucasian participants. However, past research has shown that alization, Project administration, Funding acquisition. Jennifer S.
racialized women still experience increases in body dissatisfac- Mills: Methodology, Resources, Writing - review & editing, Funding
tion when exposed to media images of thin-ideal Caucasian images acquisition.
(Thomas & Kleyman, 2020). Because past research has shown that
images of Caucasian women are most often circulated in the media Declaration of Competing Interest
as representations of the Western beauty ideal, we chose to use
images of Caucasian women as the stimuli in the current study, and The authors report no declarations of interest
reasoned that these images would be threatening to body image
across racial groups. Finally, the results may be subject to demand Acknowledgement
effects and a stronger deceptive cover story could have been used.
However, due to participants’ feedback about feeling unaffected by This research was supported by an Insight Grant to the second
the images, and the relatively subtle changes on the VAS scales, we author from the Social Sciences and Humanities Research Council
feel it is unlikely that our findings were due to biased responding of Canada.
by participants.
It has been previously well documented that seemingly upwards References
comparison to peers, celebrities, and influencers with idealized
bodies on social media is associated with negative body image. Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting
However, future research should determine the effect of other types interactions. Thousand Oaks, CA: Sage Publications.
Atkinson, M. J., & Wade, T. D. (2012). Impact of metacognitive acceptance on body
of comparisons on social media on young women’s body image. dissatisfaction and negative affect: Engagement and efficacy. Journal of
For example, it would be useful to examine whether comparison to Consulting and Clinical Psychology, 80, 416–425. https://doi.org/10.1037/
curvier body types or more average body types is associated with a0028263
Brown, Z., & Tiggemann, M. (2016). Attractive celebrity and peer images on
improvements to state body image. Future research might also con- Instagram: Effect on women’s mood and body image. Body Image, 19, 37–43.
tinue to expand on how physical appearance perfectionism impacts https://doi.org/10.1016/j.bodyim.2016.08.007
comparison to idealized body types on Instagram by examining if Cash, T. F., Santos, M. T., & Fleming Williams, E. (2005). Coping with body-image
threats and challenges: Validation of the body image coping strategies
the two subscales of the PAPS (worry about perfection and hope
inventory. Journal of Psychosomatic Research, 58, 191–199. https://doi.org/10.
for perfection) are differentially related to body image and cogni- 1016/j.jpsychores.2004.07.008
tive coping. Future research should also examine other personality Castro, J., Soares, M. J., Pereira, A. T., & Macedo, A. (2017). Perfectionism and
moderators, besides physical appearance perfectionism, that may negative/positive affect associations: The role of cognitive emotion regulation
and perceived distress/coping. Trends in Psychiatry and Psychotherapy, 39,
impact the effect of comparison to thin ideal images on body image, 77–87. https://doi.org/10.1590/2237-6089-2016-0042
in order to identify vulnerable subgroups of women. Studies should Cattarin, J. A., Thompson, J. K., Thomas, C., & Williams, R. (2000). Body image,
also be conducted that include men and relevant outcomes such as mood, and televised images of attractiveness: The role of social comparison.
Journal of Social and Clinical Psychology, 19, 220–239. https://doi.org/10.1521/
drive for muscularity, to see how upwards comparison on Insta- jscp.2000.19.2.220
gram influences men’s mental health. Cohen, D. L., & Petrie, T. A. (2005). An examination of psychosocial correlates of
disordered eating among undergraduate women. Sex Roles, 52, 29–42. https://
doi.org/10.1007/s11199-005-1191-x
Donovan, C. L., Chew, D., & Penny, R. (2014). Perfecting weight restriction: The
moderating influence of body dissatisfaction on the relationship between
5. Conclusions perfectionism and weight control practices. Behaviour Change, 31, 189–204.
https://doi.org/10.1017/bec.2014.11
Experimentally-induced appearance comparisons to idealized Ellison, N. B., Steinfield, C., & Lampe, C. (2007). The benefits of Facebook “friends;”
Social capital and college students’ use of online social network sites. Journal of
body images on Instagram resulted in decreases in state confi-
Computer-Mediated Communication, 12, 1143–1168. https://doi.org/10.1111/j.
dence and increases in state weight dissatisfaction and appearance 1083-6101.2007.00367.x
dissatisfaction in a sample of young women. Young women who Fairburn, C. G., Cooper, Z., & Shafran, R. (2003). Cognitive behavior therapy for
are highly perfectionistic about their appearance are relatively eating disorders: A “transdiagnostic” theory and treatment. Behavior Research
and Therapy, 41, 509–528. https://doi.org/10.1016/S0005-7967(02)00088-8
more vulnerable to lower confidence, and more appearance dis- Fardouly, J., & Holland, E. (2018). Social media is not real life: The effect of
satisfaction and weight dissatisfaction after comparing themselves attaching disclaimer-type labels to idealized social media images on women’s
to idealized Instagram models. This is due to their tendency to body image and mood. New Media & Society, 20, 4311–4328. https://doi.org/10.
1177/1461444818771083
cope with these comparisons by ruminating and catastrophizing Fardouly, J., Pinkus, R. T., & Vartanian, L. R. (2017). The impact of appearance
about how discrepant their bodies are from the idealized bodies comparisons made through social media, traditional media, and in person in
of the models. These results indicate that personality and ways of women’s everyday lives. Body Image, 20, 31–39. https://doi.org/10.1016/j.
bodyim.2016.11.002
coping both play a significant role in how young women’s body Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power 3: A flexible
image are affected by comparison to idealized body types on social statistical power analysis program for the social, behavioral, and biomedical
media. Clinical interventions for body dissatisfaction worsened by sciences. Behavior Research Methods, 39, 175–191. https://doi.org/10.3758/
BF03193146
upwards appearance-based comparisons should target reducing Ferguson, C. J. (2013). In the eye of the beholder: Thin-ideal media affects some,
unhelpful cognitive coping strategies and increasing the use of but not most, viewers in a meta-analytic review of body dissatisfaction in
helpful coping strategies. women and men. Psychology of Popular Media Culture, 2, 20–37. https://doi.org/
10.1037/a0030766
Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7,
117–140. https://doi.org/10.1177/001872675400700202
Frey, B. (2018).. The SAGE encyclopedia of educational research, measurement, and
Funding statement evaluation (Vols. 1–4) Thousand Oaks, CA: SAGE Publications, Inc. https://doi.
org/10.4135/9781506326139
The authors would like to acknowledge funding that was Garnefski, N., Baan, N., & Kraaij, V. (2005). Psychological distress and cognitive
emotion regulation strategies among farmers who fell victim to the foot-
awarded to the both the first and second author from the Social and-mouth crisis. Personality and Individual Differences, 38, 1317–1327. https://
Sciences and Humanities Research Council. doi.org/10.1016/j.paid.2004.08.014

61
S.E. McComb, J.S. Mills Body Image 38 (2021) 49–62

Garnefski, N., Kraaij, V., & Spinhoven, P. (2001). Negative life events, cognitive Riviere, J., & Douilliez, C. (2017). Perfectionism, rumination and gender are related
emotion regulation and emotional problems. Personality and Individual to symptoms of eating disorders: A moderated mediation model. Personality
Differences, 30, 1311–1327. https://doi.org/10.1016/S0191-8869(00)00113-6 and Individual Differences, 116, 63–68. https://doi.org/10.1016/j.paid.2017.04.
Green, C., Kangas, M., & Fairholm, I. (2018). Investigating the emotion regulation 041
strategies implemented by adults grieving the death of a pet in Australia and Rodgers, R. F., McLean, S. A., & Paxton, S. J. (2015). Longitudinal relationships
the UK. Journal of Loss & Trauma, 23, 484–501. https://doi.org/10.1080/ among internalization of the media ideal, peer social comparison, and body
15325024.2018.1478934 dissatisfaction: Implications for the tripartite influence model. Developmental
Groesz, L. M., Levine, M. P., & Murnen, S. K. (2002). The effect of experimental Psychology, 51, 706–713. https://doi.org/10.1037/dev0000013
presentation of thin media images on body satisfaction: A meta-analytic Rudolph, S. G., Flett, G. L., & Hewitt, P. L. (2007). Perfectionism and deficits in
review. The International Journal of Eating Disorders, 31, 1–16. https://doi.org/ cognitive emotion regulation. Journal of Rational-Emotive & Cognitive-Behavior
10.1002/eat.10005 Therapy, 25, 343–357. https://doi.org/10.1007/s10942-007-0056-3
Hargreaves, D. A., & Tiggemann, M. (2003). Female “thin ideal” media images and Sakakibara, R., & Endo, T. (2016). Cognitive appraisal as a predictor of cognitive
boys’ attitudes towards girls. Sex Roles, 49, 539–544. https://doi.org/10.1023/ emotion regulation. The Japanese Psychological Research, 58, 175–185. DOI:
A:1025841008820 10.1111/jpr.12098.
Hayes, A. F. (2017). Introduction to mediation, moderation, and conditional process Shafran, R., Cooper, Z., & Fairburn, C. G. (2002). Clinical perfectionism: A
analysis: A regression-based approach (2nd ed.). New York, NY: The Guilford cognitive-behavioural analysis. Behaviour Research and Therapy, 40, 773–791.
Press. https://doi.org/10.1016/s0005-7967(01)00059-6
Heinberg, L. H., & Thompson, J. K. (1995). Body image and televised images of Stapleton, P., Luiz, G., & Chatwin, H. (2017). Generation validation: The role of
thinness and attractiveness: A controlled laboratory investigation. Journal of social comparison in use of Instagram among emerging adults.
Social and Clinical Psychology, 14, 325–338. https://doi.org/10.1051/jscp.1995. Cyberpsychology, Behavior, and Social Networking, 20, 142–149. https://doi.org/
14.4.325 10.1089/cyber.2016.0444
Hogue, J. V., & Mills, J. S. (2019). The effects of active social media engagement with Stoeber, J., & Stober, F. S. (2009). Domains of perfectionism: Prevalence and
peers on body image in young women. Body Image, 28, 1–5. https://doi.org/10. relationships with perfectionism, gender, age, and satisfaction with life.
1016/j.bodyim.2018.11.002 Personality and Individual Differences, 46, 530–535. https://doi.org/10.1016/j.
Irving, L. M. (1990). Mirror image: Effects of the standard of beauty on the self- and paid.2008.12.006
body-esteem of women exhibiting varying levels of bulimic symptoms. Journal Stoeber, J., & Yang, H. (2015). Physical appearance perfectionism explains variance
of Social and Clinical Psychology, 9, 230–242. https://doi.org/10.1521/jscp.1990. in eating disorder symptoms above general perfectionism. Personality and
9.2.230 Individual Differences, 86, 303–307. https://doi.org/10.1016/j.paid.2015.06.032
Kelly, N. R., Lydecker, J. A., & Mazzeo, S. E. (2012). Positive cognitive coping Thomas, F. N., & Kleyman, K. S. (2020). The influence of Western body ideals on
strategies and binge eating in college women. Eating Behaviors, 13, 289–292. Kenyan, Kenyan American, and African Americans’ body image. Journal of
https://doi.org/10.1016/j.eatbeh.2012.03.012 Prevention & Intervention in the Community, 48, 312–328. https://doi.org/10.
Lewis-Smith, H., Diedrichs, P. C., & Halliwell, E. (2019). Cognitive-behavioral roots 1080/10852352.2019.1627084
of body image therapy and prevention. Body Image, 31, 309–320. DOI: Tiggemann, M., & McGill, B. (2004). The role of social comparison in the effect of
0.1016/j.bodyim.2019.08.009. magazine advertisements on women’s mood and body dissatisfaction. Journal
Margolis, S. E., & Orsillo, S. M. (2016). Acceptance and body dissatisfaction: of Social and Clinical Psychology, 23(1), 23–44. https://doi.org/10.1521/jscp.23.
Examining the efficacy of a brief acceptance based intervention for body 1.23.26991
dissatisfaction in college women. Behavioural and Cognitive Psychotherapy, 44, Tiggemann, M., & Polivy, J. (2010). Upward and downward: Social comparison
482–492. https://doi.org/10.1017/S1352465816000072 processing of thin idealized media images. Psychology of Women Quarterly, 34,
Michaels, M. S., Parent, M. C., & Moradi, B. (2013). Does exposure to 356–364. https://doi.org/10.1111/j.1471-6402.2010.01581.x
muscularity-idealizing images have self-objectification consequences for Tiggemann, M., & Slater, A. (2004). Thin ideals in music television: A source of
heterosexual and sexual minority men? Psychology of Men & Masculinity, 14, social comparison and body dissatisfaction. The International Journal of Eating
175–183. https://doi.org/10.1037/a0027259 Disorders, 35, 48–58. https://doi.org/10.1002/eat.10214
Mills, J., Musto, S., Williams, L., & Tiggemann, M. (2018). “Selfie” harm: Effects on Tiggemann, M., & Slater, A. (2013). Netgirls: The Internet, Facebook, and body
mood and body image in young women. Body Image, 27, 86–92. https://doi.org/ image concern in adolescent girls. The International Journal of Eating Disorders,
10.1016/j.bodyim.2018.08.007 46, 630–633. https://doi.org/10.1002/eat.22141
Mills, J. S., Polivy, J., Herman, C. P., & Tiggemann, M. (2002). Effects of exposure to Tiggemann, M., Polivy, J., & Hargreaves, D. (2009). The processing of thin ideals in
thin media images: Evidence of self-enhancement among restrained eaters. fashion magazines: A source of social comparison or fantasy? Journal of Social
Personality & Social Psychology Bulletin, 28, 1687–1699. https://doi.org/10. and Clinical Psychology, 28, 73–93. https://doi.org/10.1521/jscp.2009.28.1.73
1177/014616702237650 Verduyn, P., Gugushvili, N., Massar, K., Taht, K., & Kross, E. (2020). Social
Mills, J. S., Shannon, A., & Hogue, J. (2017). Beauty, body image, and the media. In comparison on social networking sites. Current Opinion in Psychology, 36,
Perception of beauty. (chapter 8). Retrieved from. https://www.intechopen. 32–37. https://doi.org/10.1016/j.copsyc.2020.04.002
com/books/perception-of-beauty/beauty-body-image-and-the-media Wade, T., George, W. M., & Atkinson, M. (2009). A randomized controlled trial of
Moller, A. T., & Bothma, M. E. (2001). Body dissatisfaction and irrational beliefs. brief interventions for body dissatisfaction. Journal of Consulting and Clinical
Psychological Reports, 88, 423–430. https://doi.org/10.2466/pr0.2001.88.2.423 Psychology, 77, 845–854.
Morrison, T. G., Kalin, R., & Morrison, M. A. (2004). Body-image evaluation and Wade, T. D., & Tiggemann, M. (2013). The role of perfectionism in body
body-image investment among adolescents: A test of sociocultural and social dissatisfaction. Journal of Eating Disorders, 1, 1–6. https://doi.org/10.1186/
comparison theories. Adolescence, 39, 571–592. 2050-2974-1-2
Mulgrew, K. E., Johnson, L. M., Lane, B. R., & Katsikitis, M. (2014). The effect of Welch, E., Miller, J., Ghaderi, A., & Vaillancourt, T. (2009). Does perfectionism
aesthetic versus process images on men’s body satisfaction. Psychology of Men mediate or moderate the relation between body dissatisfaction and disordered
& Masculinity, 15, 452–459. https://doi.org/10.1037/a0034684 eating attitudes and behaviors? Eating Behavior, 10, 168–175. https://doi.org/
Naumann, E., Tuschen-Caffier, B., Voderholzer, U., Schafer, J., & Svaldi, J. (2016). 10.1016/j.eatbeh.2009.05.002
Effects of emotional acceptance and rumination on media-induced body World Health Organization. (2019). BMI classification Retrieved from. http://apps.
dissatisfaction in anorexia and bulimia nervosa. Journal of Psychiatric Research, who.int/bmi/index.jsp?introPage=intro 3.html
82, 119–125. https://doi.org/10.1016/j.jpsychires.2016.07.021 Yan, Y., & Bissell, K. (2014). The globalization of beauty: How is ideal beauty
Padoa, T., Berle, D., & Roberts, L. (2018). Comparative social media use and the influenced by globally published fashion and beauty magazines? Journal of
mental health of mothers with high levels of perfectionism. Journal of Social and Intercultural Communication Research, 43, 194–214. https://doi.org/10.1080/
Clinical Psychology, 37, 514–535. https://doi.org/10.1521/jscp.2018.37.7.514 17475759.2014.917432
Pew Research Center. (2018). Social media use in 2018 Retrieved from. http://www. Yang, H., & Stoeber, J. (2012). The physical appearance perfectionism scale:
pewinternet.org/2018/03/01/social-media-use-in-2018/ Development and preliminary validation. Journal of Psychopathology and
Pokrajac-Bulian, A., Ambrosi-Randic, N., & Kukic, M. (2008). Thin-ideal Behavioral Assessment, 34, 69–83. https://doi.org/10.1007/s10862-011-9260-7
internalization and comparison process as mediators of social influence and Yang, H., Yang, Y., Xu, L., Wu, Q., Xu, J., Weng, E., . . . & Cai, S. (2017). The relation of
psychological functioning in the development of disturbed eating habits in physical appearance perfectionism with body dissatisfaction among school
Croatian college females. Psychological Topics, 17, 221–245. students 9-18 year of age. Personality and Individual Difference, 116, 399–404.
https://doi.org/10.1016/j.paid.2017.05.005

62