Body Image 22 (2017) 65–71

Contents lists available at ScienceDirect

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

Idealised media images: The effect of fitspiration imagery on body

satisfaction and exercise behaviour
Lily Robinson a , Ivanka Prichard b,d,∗ , Alyssa Nikolaidis a , Claire Drummond b,d ,
Murray Drummond c,d , Marika Tiggemann a
a
School of Psychology, Flinders University, Australia
b
Discipline of Health & Exercise Sciences, School of Health Sciences, Flinders University, Australia
c
School of Education, Flinders University, Australia
d
SHAPE Research Centre, Flinders University, Australia

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

Article history: Recent studies have documented a shift in the cultural ideal of physical attractiveness, with women
Received 17 February 2017 subscribing to a visibly toned ideal that emphasises health and fitness. The present study experimentally
Received in revised form 2 June 2017 investigated the impact of athletic and muscular fitness-idealised images compared to traditional thin
Accepted 2 June 2017
ideal images on women’s body dissatisfaction and exercise behaviour, under the framework of Social
Comparison Theory. Participants were 106 female undergraduate students randomly assigned to view
Keywords:
one of three sets of images (thin ideal, athletic ideal, or muscular ideal) followed by a bout of exercise.
Media
Acute exposure to athletic ideal and thin ideal images led to increased body dissatisfaction, but exposure
Exercise
Social comparison
to muscular ideal images did not. Relative to thin ideal images, fitness-idealised images did not motivate
Fitspiration participants to engage in higher levels of exercise suggesting that this type of fitness inspiration might
Muscularity not motivate actual exercise behaviour.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction image in women (Levine & Murnen, 2009). However, an ‘athletic

ideal’ characterised by both thinness and visible toning, including
A large body of research has demonstrated that mass media a more muscular upper body, toned abdomen and firmer lower
images have a significant influence on sociocultural body ideals and body, has become popular across different forms of media (Boepple,
are strongly influential on female body image perceptions (Groesz, Ata, Rum, & Thompson, 2016; Schaefer et al., 2015; Tiggemann &
Levine, & Murnen, 2002; Perloff, 2014; Stice, Spangler, & Agras, Zaccardo, 2016). A small amount of research has demonstrated
2001; Thompson & Heinberg, 1999). Body image is defined as an that media exposure to images that are toned as well as thin do
individual’s perception of their body appearance, as well as the promote negative body image, similar to the thin ideal (Homan,
thoughts and feelings that result from this perception (Cash, 2004; 2010). This appears to be the case for print media (Homan, 2010),
Tiggemann, 2004). Due to their emphasis on appearance and the as well as social media images presented through the Internet (Jong
importance of attractiveness in Western societies, idealised media & Drummond, 2016a, 2016b; Tiggemann & Zaccardo, 2015).
images place pressure on women to achieve a similar aesthetic pro- This shift in the popularity of the athletic ideal has, in part,
file and can have significant negative implications for body image been due to the global social media movement known as ‘fitspi-
(Levine & Murnen, 2009). ration’ (an amalgamation of the words ‘fitness’ and ‘inspiration’), a
The thin ideal represents the female form as a slender, femi- social media source which many women now use for health-related
nine physique with a small waist and little body fat (Grogan, 2008; information and inspiration related to diet and exercise (Jong &
Schaefer et al., 2015). Past experimental studies have demonstrated Drummond, 2016a; Vaterlaus, Patten, Roche, & Young, 2015). “Fit-
a link between exposure to thin ideal traditional mass media, such spiration” consists of images and inspirational quotations aimed
as television, magazines and advertisements, and negative body typically at women to inspire healthy living by providing tips
on exercise and diet to improve health and fitness (Boepple &
Thompson, 2016; Jong & Drummond, 2016a). A search (December,
∗ Corresponding author at: School of Health Sciences, Flinders University, GPO
2016) of the #fitspiration hashtag on Instagram returned close to
Box 2100, Adelaide, South Australia 5001, Australia.
10 million pictures. A perusal of the images under the fitspira-
E-mail address: ivanka.prichard@flinders.edu.au (I. Prichard). tion hashtag shows two distinct ideal body types: an athletic ideal

http://dx.doi.org/10.1016/j.bodyim.2017.06.001
1740-1445/© 2017 Elsevier Ltd. All rights reserved.
66 L. Robinson et al. / Body Image 22 (2017) 65–71

and a muscular ideal. The athletic ideal refers to a lean body, little types of ideal images (e.g., Benton & Karazsia, 2015) has not exam-
body fat, and an athletic (i.e., fit) and toned/muscular appearance ined the effect of these images on exercise intentions or actual
(Schaefer et al., 2015). In contrast, the muscular ideal, whilst still exercise behaviour. Given the physical and psychological benefits of
generally reasonably thin, is overtly muscular in appearance with exercise (Buckworth, Dishman, O’Connor, & Tomporowski, 2013),
well-defined leg, arm, and abdominal muscles. it is important to determine whether certain types of idealised
While athletic ideal images may be viewed as healthier (due to images actually motivate engagement in exercise and/or whether
their focus on promoting fitness) than thin ideal images, they may their tendency to promote negative body image deters women from
be quite difficult to attain and are actually somewhat paradoxical engaging in exercise. As fitspiration images have been shown to
in nature. Realistically, attainment of an athletic ideal necessi- increase exercise inspiration (Tiggemann & Zaccardo, 2015), it is
tates strict dieting practice, as well as substantial and consistent anticipated that these types of images are likely to also promote
weight training (Ramme, Donovan, & Bell, 2016). For the major- actual exercise behaviour. By doing so, strategies can be devised to
ity of women, this lifestyle is unrealistic and unobtainable (Curioni better motivate engagement in physical activity.
& Lourenco, 2005). Recent content analyses of fitspiration media The present study aimed to investigate the effects of fitness-
have also warned that components of these images, including their idealised images in comparison to traditional thin ideal images,
overall focus on appearance, the presence of objectifying features, on body dissatisfaction and exercise behaviour (distance travelled
the emphasis on appearance-related reasons for exercise, and the on a treadmill over a 10-minute period), under the framework of
focus on weight management behaviours, are likely to have dam- Social Comparison Theory. This was done using an experimental
aging effects for female body image (Boepple et al., 2016; Carrotte, design examining the impact of three different types of images
Prichard, & Lim, 2017; Simpson & Mazzeo, 2016; Tiggemann & (thin ideal, athletic ideal, and muscular ideal). It was predicted that
Zaccardo, 2016). exposure to both thin ideal and athletic ideal images emphasising
Research suggests that exposure to images featuring a toned thinness, would lead to higher levels of body dissatisfaction and
and fit body only produces negative body image when paired with state social comparison than exposure to muscular ideal images. It
thinness (Groesz et al., 2002; Homan, McHugh, Wells, Watson, & was expected that this effect would be moderated by trait appear-
King, 2012). For example, Homan et al. (2012) found that women ance comparison. Given their focus on promoting fitness, it was also
exposed to athletic images of thin women experienced increased predicted that exposure to ‘fitspiration’ images (athletic ideal and
body dissatisfaction compared to those who viewed athletic images muscular ideal) would lead to a greater level of exercise (measured
of women who were not thin. Relatedly, Benton and Karazsia (2015) as distanced travelled on a treadmill over a 10-minute period) in
demonstrated that an athletic ideal (thin and toned) had a similar comparison to exposure to thin ideal images that do not focus on
significant negative impact on women’s body satisfaction as the fitness.
thin ideal, in a way that a more muscular (and less thin) ideal did
not. It is possible that women see the muscular ideal as too mas- 2. Method
culine and therefore not attractive (Dworkin, 2001). This notion is
supported by the common lack of support within the general popu- 2.1. Participants
lation for the level of muscularity obtained by female body builders
(Grogan, Evans, Wright, & Hunter, 2004; Shilling & Bunsell, 2009). Participants were 106 female university students recruited from
One framework that is useful for examining the potential impact Flinders University in South Australia. On average they were 20.60
of idealised images on body image is Social Comparison Theory years old (SD = 1.85) with a mean body mass index (BMI) of 22.65
(Festinger, 1954). This theory suggests that individuals frequently (SD = 3.93), which falls within the ‘healthy weight’ range.
assess their success and levels of ability by comparing them-
selves to others (Festinger, 1954). This comparison also extends to 2.2. Design
other dimensions of the self, such as physical appearance (Wheeler
& Miyake, 1992). Social comparison can be examined as a trait The study employed a between-subjects experimental design
variable, in that there exist relatively stable individual differ- with image type (thin ideal, athletic ideal, muscular ideal) as the
ences in social comparison engagement (Thompson, Heinberg, & independent variable. The major dependent variables were state
Tantleff, 1991), or as a state variable, in response to specific sit- body dissatisfaction (measured pre and post exposure), state social
uations (Tiggemann & McGill, 2004). Trait social comparison has comparison, and exercise behaviour. Trait social comparison was
been found to moderate the impact of idealised media images on tested as a moderator.
body image, such that women with higher trait social comparison
tendencies experience greater body dissatisfaction after viewing 2.3. Materials
idealised images than those with lower social comparison tenden-
cies (Dittmar & Howard, 2004). State appearance comparison is 2.3.1. Experimental manipulation: image type. Three sets
considered a key mechanism by which the media negatively affect of visual stimuli were created for the study, each containing 15
women’s body image (Tiggemann & McGill, 2004; Tiggemann & images. Images were sourced from Google Images and Instagram,
Zaccardo, 2015). According to Social Comparison Theory (Festinger, forms of media commonly accessed by the target population and
1954), humans have an innate desire to compare themselves to oth- available for public use (Sensis, 2016). To choose images, the key-
ers on relevant dimensions. When a woman compares herself to an words entered were ‘thinspiration’ for thin ideal images; ‘athletic
idealised media image, she almost invariably falls short, resulting fitspiration’ for athletic ideal; and ‘muscular fitspiration’ for mus-
in dissatisfaction with her own body. Experimentally, state appear- cular ideal images. The image sets contained women of different
ance comparison has been shown to at least partially mediate racial backgrounds and captured a variety of activities to reflect
the effect of exposure to idealised images on body dissatisfaction the types of images commonly seen on social media (Tiggemann &
(Tiggemann & McGill, 2004; Tiggemann & Zaccardo, 2015). Zaccardo, 2016). The final sets of images (15 per condition) were
Health and fitness-related images such as fitspiration aim to chosen from a larger set of 85 images that were rated equivalently
engage and motivate women to partake in healthy behaviours such on levels of visual quality and physical attractiveness (1 = not at all
as exercise. Recently, Tiggemann and Zaccardo (2015) established a to 7 = very much) by 10 independent female raters aged 17–25 years.
link between acute viewing of fitspiration images and a single-item Thin ideal images were rated high on thinness (M = 5.85; SD = 0.08);
measure of inspiration to exercise. To date, research on different medium on tone (M = 4.40; SD = 0.10); and low on athleticism and
 L. Robinson et al. / Body Image 22 (2017) 65–71 67

muscularity (M = 3.30; SD = 0.11; M = 3.06; SD = 0.11 respectively). weight concern, and internalisation of appearance ideals). Partici-
Athletic ideal images were rated medium-high on thinness; tone; pants were asked to indicate on the 5-point Likert scale (1 = never
and athleticism (M = 5.37; SD = 0.08; M = 5.42; SD = 1.00; M = 5.13; to 5 = always), how often they made physical appearance compar-
SD = 0.10 respectively); but only as moderately muscular (M = 4.63; isons across a range of social situations. A mean trait appearance
SD = 0.10). Muscular ideal images were rated high on tone; ath- comparison score was then calculated by averaging all the scores
leticism; and muscularity (M = 6.25; SD = 0.12; M = 6.34; SD = 0.13; (range 1–5). Higher scores implied a greater tendency to engage in
M = 6.45; SD = 0.13 respectively); but only medium on thinness appearance comparison. The current study demonstrated accept-
(M = 4.36; SD = 0.10). Images were presented to participants on an able internal consistency (˛ = .91).
Apple Mac 13-inch computer screen in a fixed order and each image
was displayed for approximately 15 s. 2.4.6. Fitspirational goals. Following Tiggemann and
Zaccardo (2015), two items were included to test the inspira-
2.4. Measures tional effects of the images. On a 7-point Likert scale participants
were asked how inspired they felt when viewing the images to
2.4.1. Social media use. To gauge general social media usage, improve their fitness and to be physically active (1 = not at all
participants were asked to indicate which social networking sites inspired, to 7 = very inspired). A mean fitness inspiration score was
they use and how much time they spend on these sites (less than created by averaging the two items.
10 min, 10–30 min, 31–60 min, over 1 h).
2.4.7. Exercise behaviour. Exercise behaviour was measured
as the distance travelled on a treadmill (in kilometres) by partici-
2.4.2. Regular physical activity levels. As an index of usual
pants in the 10-min period.
activity levels, all participants were asked to indicate their cur-
rent exercise habits (frequency and duration of any physical
2.4.8. Demographic information. Participants were asked
activity/sport they engage in). Following Prichard and Tiggemann
their age and ethnicity. Objective measures of current weight and
(2008), total amount of exercise was calculated by summing each
height were taken to calculate body mass index (BMI; kg/m2 ).
activity’s frequency by its duration.

2.5. Procedure
2.4.3. State body dissatisfaction. Body dissatisfaction was
measured before and after viewing the images. Following previ- Following ethics approval from the University’s Social and
ous research (Heinberg & Thompson, 1995; Prichard & Tiggemann, Behavioural Research Ethics committee, participants were
2012), this was done using four visual analogue scales (VAS; fat, recruited through the School of Psychology’s research manage-
physically attractive, satisfied with my body shape and satisfied ment system for a study on social media, exercise and well-being.
with body size). The scales consisted of 100-mm horizontal lines, Potential participants were informed that the study involved
with poles labelled not at all (0) and very much (100) that are mea- answering a series of questionnaires about their weight, exercise
sured to the nearest millimetre. Participants indicated how they behaviours, and attitudes as well as viewing some social media
currently felt by marking a spot along the line for each dimension. images and participating in a bout of physical activity. They
Positive items were reverse coded and the mean of the four items received either course credit for their participation (n = 22) or
was calculated to create a measure of state body dissatisfaction at were reimbursed $20 for 45–60 min their time (n = 86), and were
each time point. VAS are reliable measures, able to detect small debriefed upon the conclusion of the study. Testing sessions
changes in body dissatisfaction (Heinberg & Thompson, 1995). lasted approximately 45–60 min. Upon arrival at the research
In the current study, the body dissatisfaction scale demonstrated laboratory, after providing informed consent, participants com-
acceptable internal consistency at pre-exposure (˛ = .82), and post- pleted measures of their social media use and regular physical
exposure (˛ = .82). activity levels, and then completed the baseline state measure of
body dissatisfaction. Participants were then randomly assigned
2.4.4. State social comparison. The State Appearance Com- by Qualtrics to one of three conditions (thin ideal, athletic ideal,
parison Scale (Tiggemann & McGill, 2004) was used to determine or muscular ideal images). Participants were asked to rate each
the amount of appearance comparison participants engaged in after image on its visual quality to ensure participants paid attention
viewing the images. Using three 7-point Likert scales, participants to each image. All participants then completed follow-up state
rated the extent to which they thought about their appearance measures of body dissatisfaction. Participants were then asked to
when viewing images (1 = no thought about appearance, 7 = a lot of exercise on a treadmill for 10 min at any pace. After the exercise
thought about appearance), and the extent to which they compared component of the study, participants completed the measures
their body/specific parts with those seen in the images they viewed of state appearance comparison and fitness inspiration. The final
(1 = no comparison, 7 = a lot of comparison). State appearance com- questionnaire measure was trait social comparison. Participant
parison was calculated by averaging the three items, with higher height and weight were measured objectively at the end of the
scores indicating greater state appearance comparison. Previous session.
research has shown that items on this scale are highly correlated
(r = .71 to .82; Tiggemann & McGill, 2004). The current study had 3. Results
acceptable internal reliability (˛ = .85).
3.1. Sample characteristics
2.4.5. Trait social comparison. The Physical Appearance Com-
parison Scale-Revised (PACS-R) was used to measure trait Participants (N = 106) had a mean age of 22.60 years (SD = 1.85)
appearance comparison (Schaefer & Thompson, 2014). This scale and a mean BMI of 22.65 (SD = 3.93). A series of one-way anal-
consists of eleven items that assess an individual’s tendency to com- yses of variance indicated that across the three conditions (thin
pare their own appearance to the appearance of others. The PACS-R ideal, athletic ideal, muscular ideal) there was no statistically signif-
is recognised for its good inter-item reliability (r = .72; Schaefer & icant difference between groups on age, F(2, 103) = 0.013, p = .987,
Thompson, 2014) as well as good convergent validity with theo- BMI, F(2, 103) = 2.222, p = .114, weekly media use, F(2, 105) = 0.050,
retically related variables (e.g., body satisfaction, dietary restraint, p = .952, image quality ratings, F(2, 103) = 0.364, p = .696, typical
68 L. Robinson et al. / Body Image 22 (2017) 65–71

Table 1
Mean (SD) scores for age, BMI, exercise, image quality, weekly media use, trait social comparison, and baseline body dissatisfaction.

Image type

Overall (N = 106) Thin ideal (n = 36) Athletic ideal (n = 34) Muscular ideal (n = 36)

Age 20.60 (1.85) 20.58 (2.01) 20.65 (1.94) 20.58 (1.65)

BMI 22.65 (3.93) 22.68 (4.03) 23.65 (4.13) 21.69 (3.47)
Exercise (min weekly) 287.29 (206.57) 241.95 (182.26) 304.97 (207.62) 317.73 (230.30)
Image quality 3.41 (0.48) 3.42 (0.50) 3.46 (0.48) 3.36 (0.47)
Media use 2.48 (0.48) 2.46 (0.49) 2.50 (0.46) 2.48 (0.49)
Trait social comparison 3.32 (0.76) 3.26 (0.72) 3.36 (0.81) 3.36 (0.77)
State social comparison 5.15 (1.50) 5.17 (1.64) 5.56 (1.37) 4.78 (1.43)
Baseline body dissatisfaction 54.02 (17.94) 59.82 (21.09) 51.84 (16.00) 50.29 (15.02)

Scale end points: image quality, 1–5; media use, 1–4; trait social comparison, 1–5; state social comparison, 1–7; body dissatisfaction, 0–100.

amount of exercise (minutes per week), F(2, 103) = 1.368, p = .259, 3.4. Moderation by trait appearance comparison
or baseline body dissatisfaction, F(2, 103) = 3.022, p = .053 (means
in Table 1). Importantly, there was no difference between the three A hierarchical multiple regression analysis tested whether the
groups on trait social comparison, F(2, 103) = 0.202, p = .817. relationship between image type and body dissatisfaction was
moderated by trait appearance comparison. Following Aiken and
3.2. The effect of image type on body dissatisfaction West’s (1991) recommendation, trait social comparison scores
were centred to reduce multicollinearity. Two computed dummy
Given the specified predictions regarding group differences, an variables: dummy 1 (1 if in the thin ideal condition, 0 otherwise),
ANCOVA with orthogonal planned contrasts via LMATRIX was used and dummy 2 (1 if in the athletic ideal image condition, 0 oth-
to test the effect of image type on post-exposure body dissatis- erwise) were used for this analysis to represent the independent
faction. Pre-exposure body dissatisfaction was controlled for as a variable image type. Two product term variables were obtained by
covariate, and adjusted means are presented in Table 2. Specifi- multiplying both dummy variables by the centered trait appear-
cally, the planned comparisons tested the impact of viewing the ance comparison scores. Pre-exposure body dissatisfaction was
thin images (thin ideal and athletic ideal combined) against view- entered in Step 1, and then the two dummy variables and centred
ing the muscular images (contrast: −1, −1, +2), and viewing the trait appearance comparison in Step 2. The two two-way product
thin against the athletic ideal images (contrast: −1, +1, 0). terms were entered in Step 3. The total variance explained by the
The overall ANCOVA was significant, F(2, 102) = 5.35, p = .006, model as a whole was 77.5%, F(6, 101) = 57.85, p < .001. However, the
partial 2 = .095. The first planned comparison which compared inclusion of the product terms (Step 3) did not explain significant
the thin images (thin and athletic ideals) against the muscular additional variance in body dissatisfaction, R2 change = .004, Fchange (2,
images was significant, F(1, 102) = 4.98, p = .028, partial 2 = .047, 101) = 0.924, p = .400, and thus there was no significant interaction.
whereby women exposed to the thin and athletic images reported
significantly greater body dissatisfaction than women exposed to 3.5. The effect of image type on exercise inspiration and
the muscular images. The planned comparison between thin ideal behaviour
images and athletic images was also significant, F(1, 102) = 5.61,
p = .020, partial 2 = .052. Women exposed to the athletic ideal An examination of whether the different images inspired and
images reported significantly greater body dissatisfaction than motivated fitness indicated that there was a statistically signifi-
women exposed to the thin ideal images. Posthoc analysis indi- cant difference between the image type groups on inspiration to
cated that there was no difference (p > .05) between the thin and improve fitness and physical activity, F(2, 103) = 3.657, p = .029,
muscular ideal conditions. partial 2 = 0.066. Athletic ideal (M = 5.43, SD = 0.28) images were
viewed as significantly more inspirational than thin ideal images
3.3. Effect of image type on state appearance comparison (M = 4.47, SD = 0.27, p = .046), but were not significantly more inspi-
rational than viewing muscular ideal images (M = 4.57, SD = 0.27,
A one-way ANOVA with planned comparisons via LMATRIX p = .086).
was conducted to compare state appearance comparison resulting For exercise behaviour, an ANCOVA was conducted to deter-
from viewing the thin and athletic images versus the muscu- mine the effect of image type on distance travelled on the treadmill
lar images, and the athletic images versus the thin ideal images after controlling for usual exercise behaviour (mins per week),
(means displayed in Table 1). The overall ANOVA was not signif- which was entered as a covariate. Results showed there was no
icant, F(2, 101) = 2.18, p = .118, partial 2 = .041, and neither were difference in distance travelled on the treadmill across the image
the planned comparisons; thin and athletic versus muscular: F(1, type groups, F(2, 100) = 0.230, p = .795, partial 2 = .005. Participants
101) = 3.47, p = .066, partial 2 = .033, and thin versus athletic ideal: who viewed the athletic ideal (M = 0.97, SD = 0.31) and muscular
F(1, 101) = 0.96, p = .329, partial 2 = 0.009. ideal images (M = 0.98, SD = 0.29) did not travel significantly further
Given that overall scores on state social comparison were rel- on the treadmill than participants who viewed thin ideal images
atively high (M = 5.15 on a scale of 1–7) and all experimental (M = 0.92, SD = 0.34).
conditions contained images of people, the impact of state social
comparison on body dissatisfaction, irrespective of condition, was 4. Discussion
tested using hierarchical multiple regression. Pre-exposure body
dissatisfaction was entered in Step 1, and state social comparison The present study sought to determine how exposure to ide-
was entered in Step 2. At Step 2, state social comparison accounted alised fitness images (‘fitspiration’) affected women’s state body
for unique variance in body dissatisfaction, beyond that provided dissatisfaction relative to that of the traditional thin ideal, and
by baseline body dissatisfaction, R2 change = .014, F change = 5.538, to determine whether these images inspire exercise and promote
p = .021. Thus, state social comparison predicted increased body actual engagement in exercise. Consistent with previous research,
dissatisfaction across conditions. exposure to idealised media images that emphasised thinness, that
 L. Robinson et al. / Body Image 22 (2017) 65–71 69

Table 2
Mean (SD) scores for body dissatisfaction by image type.

Image type Pre-exposure Post-exposure Adjusted post-exposure

Thin ideal 59.82 (21.09) 62.85 (21.02) 57.74 (1.57)a

Athletic ideal 51.84 (16.00) 61.15 (16.97) 63.08 (1.59)b
Muscular ideal 50.29 (15.02) 52.84 (15.60) 56.13 (1.56)a

Scale end points: body dissatisfaction, 0–100.

Means with different superscripts denote significant differences (p < .05).

is, both the athletic and thin ideal images, produced a negative ‘healthy lifestyle’, they may not accomplish their intended purpose,
effect on body dissatisfaction, relative to muscular ideal images instead leading to increased body dissatisfaction.
(Benton & Karazsia, 2015; Homan et al., 2012). In addition, par- To date, research has established a preliminary link between
ticipants who viewed athletic ideal images reported greater body viewing fitspirational images and a greater inspiration to exercise,
dissatisfaction than participants who viewed traditional thin ideal compared to participants who viewed travel images (Tiggemann
images, suggesting that thin and fit images may be more damaging & Zaccardo, 2015). In the present study, greater reported inspira-
for women’s body image. tion to exercise was also evident among participants who viewed
The finding that the athletic ideal images led to the most body athletic ideal images compared to participants who viewed images
dissatisfaction is perhaps not surprising. The athletic ideal promises depicting the thin ideal. However, this inspiration did not trans-
that, if an individual invests enough time and effort into maintain- late into actual behaviour. No differences were found between the
ing a healthy lifestyle (e.g., regular exercise and eating a healthy participants who viewed fitness-idealised images and participants
diet), they too can achieve this physique. Problematically, these who viewed thin ideal images on exercise behaviour (distance trav-
images may oversell the attainability of this ‘ideal’ body, poten- elled). While deemed inspirational, in the present study exposure
tially setting many women up for failure. The saturation of athletic to fitness imagery did not have an effect on actual exercise engage-
ideal images further makes this ideal body seem both attainable and ment. This is perhaps not surprising given that previous research
normative, desensitising women from their generally unobtainable with young people has shown that intentions to do more exercise
nature (Sabiston & Chandler, 2009). In addition, while athletic ideal do not always translate into actual behaviour (Poobalan, Aucott,
images are deemed inspirational, it remains unknown what effect Clarke, & Smith, 2012). These findings raise concern given that
they may have on women’s reasons for exercise. Content analyses exposure to the same imagery also results in increased body dis-
suggest that fitspiration images focus heavily on the appearance- satisfaction, and indicate the need for future research in this area.
related benefits of engaging in exercise and dieting rather than the One possible explanation as to why there were no emergent
health and wellbeing benefits (e.g., Carrotte et al., 2017; Tiggemann group differences on exercise behaviour is that the current study
& Zaccardo, 2016). Exercising for appearance reasons rather than was conducted in a laboratory environment and as such, the
for health and wellbeing reasons has been linked to higher body exercise component may have been considered artificial. Future
dissatisfaction and disordered eating, and low self-esteem among research would benefit from considering the real-world effects
women (Gonçalves & Gomes, 2012; Prichard & Tiggemann, 2008). of exposure to different types of idealised images on exercise
As such, the value these images place on exercising to improve behaviour. In a real world setting, the impact of viewing fitness-
appearance could have a widespread detrimental effect on female idealised images on exercise behaviour may be reliant on a variety
body image and should be examined further. of other factors such as cost, time, travel, and social support. Future
In the present study, images were examined in the context exploration of idealised media images and exercise behaviour
of Social Comparison Theory (Festinger, 1954). Western societies should also reconsider the way in which exercise is measured. In
offer women opportunities to engage in appearance comparisons the current study, participants travelled at a pace of their choosing
virtually on a daily basis (Jung & Lee, 2006), thus increasing the on the treadmill for a brief bout of 10 min. Perhaps exercising for a
opportunities for women to experience increased body dissatisfac- greater period or being offered a choice of whether to exercise or
tion. The athletic ideal is visible in a variety of contexts, including not would have elicited differences between the groups.
fitness centres, health magazines, and on social media in the form The study offers a variety of practical implications and con-
of fitspiration (Boepple et al., 2016; Tiggemann & Zaccardo, 2016). sequently possible avenues for future research. Contemporary
Overall, participants in each condition reported levels of state social women are aware that viewing and subscribing to the thin ideal can
comparison similar to that noted in previous research with fitspi- contribute to a number of negative health outcomes (e.g., anorexia,
ration images (Tiggemann & Zaccardo, 2015), and there were no bulimia, and low self-esteem; Smolak & Levine, 2015). Widespread
significant differences between conditions. It appears that partici- mass media coverage of these potential effects has frequented
pants compared their appearance to the women presented in each the news headlines in the past decade (e.g., collective banning of
of the image conditions. Further, irrespective of condition, state proanorexia and #thinspiration social media pages by the big social
appearance comparison predicted increased body dissatisfaction, networking sites; Duca, 2015). Similar to exposure to the thin ideal,
highlighting its important role in relation to elevated body image results from the present study show that athletic ideal images in
concern. In line with Tiggemann and Zaccardo (2015), although the form of ‘fitspiration’ also lead to a significant increase in body
contrary to Dittmar and Howard (2004), trait appearance compari- dissatisfaction. Further, exposure to these images, in comparison
son was not found to moderate the relationship between exposure to thin ideal images or muscular images, did not increase engage-
to the different images and body dissatisfaction. This finding sug- ment in exercise in the present study (as measured in a laboratory
gests that everyday state appearance comparisons may play a more setting). This suggests that these images, while potentially inspi-
important role than trait appearance comparison. The findings also rational, may not fulfil their intended purpose. This is particularly
provide insight into how the change in current sociocultural body important given the current popularity of fitspiration media and
image ‘ideal’ could have many implications for women’s psycholog- the general perception that because of its focus on health and fit-
ical functioning and behaviour. Although fitness-idealised images ness, it is better than traditional thin ideal images. Therefore, media
aim to inspire positive body image and encourage women to live a outlets should be careful in their use of athletic ideal images, partic-
ularly as a source of appearance-related motivation. This may prove
70 L. Robinson et al. / Body Image 22 (2017) 65–71

a difficult challenge, particularly in western societies where the use Funding

of social media websites, including the relatively new online trend
‘fitspiration’, has increased dramatically. This work was supported by the Faculty of Medicine, Nursing &
These findings also present specific implications for the way Health Sciences at Flinders University.
in which health care professionals provide support for women
struggling with body image concerns. To date, some intervention References
programs have helped women to effectively confront their beliefs
about the importance and appeal of thinness (Mitchell, Mazzeo, Benton, C., & Karazsia, B. T. (2015). The effect of thin and muscular images on
Rausch, & Cooke, 2007). In the present study, the muscular (but women’s body satisfaction. Body Image, 13, 22–27. http://dx.doi.org/10.1016/
j.bodyim.2014.11.001
not thin) ideal was somewhat protective of body dissatisfaction. By
Boepple, L., Ata, R. N., Rum, R., & Thompson, J. K. (2016). Strong is the new skinny:
emphasising the importance of health and fitness, women may be A content analysis of fitspiration websites. Body Image, 17, 132–135. http://dx.
encouraged to see fitness, rather than achieving the ‘ideal body’, doi.org/10.1016/j.bodyim.2016.03.001
Boepple, L., & Thompson, J. K. (2016). A content analytic comparison of fitspiration
as their aim. Accordingly, incorporating the importance of strength
and thinspiration websites. International Journal of Eating Disorders, 49, 98–101.
and fitness into these interventions might lead to positive body http://dx.doi.org/10.1002/eat.22403
image outcomes for women. Future research could examine the Boyd, D. (2014). It’s complicated: The social lives of networked teens. New Haven, CT:
impact that messages of strength and fitness have on the other- Yale University Press.
Buckworth, J., Dishman, R. K., O’Connor, P. J., & Tomporowski, P. D. (2013). Exercise
wise negative health outcomes associated with viewing idealised psychology (2nd ed.). Champaign, IL: Human Kinetics.
images. Carrotte, R. E., Prichard, I., & Lim, C. M. S. (2017). Fitspiration on social media: A
The findings should be interpreted in light of a number of lim- content analysis of gendered images. Journal of Medical Internet Research, 19,
e95. http://dx.doi.org/10.2196/jmir.6368
itations. First, the study was conducted in a laboratory setting. Cash, T. F. (2004). Body image: Past, present, and future. Body Image, 1, 1–5. http://
This limits the generalisability of the findings. Second, partici- dx.doi.org/10.1016/S1740-1445(03)00011-1
pants were asked to rate the quality of each image they viewed. Curioni, C. C., & Lourenco, P. M. (2005). Long-term weight loss after diet and exercise:
A systematic review. International Journal of Obesity, 29, 1168–1174. http://dx.
This was done to ensure that they attended to the images, but is doi.org/10.1038/sj.ijo.0803015
not typical of usual engagement with images in different forms Dittmar, H., & Howard, S. (2004). Thin-ideal internalization and social comparison
of media. Third, as a neutral control condition (with no images tendency as moderators of media models’ impact on women’s body-focused
anxiety. Journal of Social and Clinical Psychology, 23, 768–791. http://dx.doi.org/
of people) was not utilised, it remains unknown whether athletic
10.1521/jscp.23.6.768.54799
ideal images may increase exercise engagement over and above Duca, L. (2015). Can thinspiration really be #Banned from Instagram? The
exposure to neutral non-fitness focused images. Fourth, despite Huffington Post. Retrieved from http://www.huffingtonpost.com/laurenduca/
thinspiration-banned-from-instagram b 3829155.html?ir=Australia
being the first study to examine the impact of fitness images on
Dworkin, S. L. (2001). Holding back: Negotiating a glass ceiling on women’s muscular
exercise behaviour, our measurement of exercise behaviour was strength. Sociological Perspectives, 44, 333–350. http://dx.doi.org/10.1525/sop.
limited to a 10-min bout on a treadmill and may not have accurately 2001.44.3.333
captured exercise behaviour. Future research should consider Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7,
117–140. http://dx.doi.org/10.1177/001872675400700202
other methods of measuring exertion and exercise behaviour Gonçalves, S. F., & Gomes, A. R. (2012). Exercising for weight and shape reasons
such as heart rate monitors, and more naturalistic measurement vs. health control reasons: The impact on eating disturbance and psychological
(e.g., fitness trackers). In addition, it would be useful to explore functioning. Eating Behaviors, 13, 127–130. http://dx.doi.org/10.1016/j.eatbeh.
2011.11.011
the mechanisms by which fitspiration images might inspire fit- Groesz, L. M., Levine, M. P., & Murnen, S. K. (2002). The effect of experimental pre-
ness (e.g., appearance-related reasons for exercise, focus on body sentation of thin media images on body satisfaction: A meta-analytic review.
functionality). Lastly, our sample was predominantly young Cau- International Journal of Eating Disorders, 31, 1–16. http://dx.doi.org/10.1002/eat.
10005
casian Australian university students. Although the majority of the Grogan, S. (2008). Body image: Understanding body dissatisfaction in men, women, and
research on body image tends to employ a similar sample, future children (2nd ed.). London and New York: Routledge.
research should consider examining larger and more diverse pop- Grogan, S., Evans, R., Wright, S., & Hunter, G. (2004). Feminity and muscularity:
Accounts of seven women body builders. Journal of Gender Studies, 13, 49–61.
ulations. In particular, the study would benefit from considering
Heinberg, L. J., & Thompson, J. K. (1995). Body image and televised images of thinness
the impact of these images on female adolescents, as this age and attractiveness: A controlled laboratory investigation. Journal of Social and
group is the largest user of online social media websites (Boyd, Clinical Psychology, 14, 325–338. http://dx.doi.org/10.1521/jscp.1995.14.4.325
Homan, K. (2010). Athletic-ideal and thin-ideal internalization as prospective pre-
2014).
dictors of body dissatisfaction, dieting, and compulsive exercise. Body Image, 7,
In conclusion, notwithstanding these limitations, the study 240–245. http://dx.doi.org/10.1016/j.bodyim.2010.02.004
offers a significant contribution to the existing body of research Homan, K., McHugh, E., Wells, D., Watson, C., & King, C. (2012). The effect of viewing
on the psychological and behavioural outcomes associated with ultra-fit images on college women’s body dissatisfaction. Body Image, 9, 50–56.
http://dx.doi.org/10.1016/j.bodyim.2011.07.006
viewing idealised media images for women. The current study Jong, S. T., & Drummond, M. J. N. (2016a). Exploring online fitness culture and young
extends previous research (e.g., Benton & Karazsia, 2015; Homan females. Leisure Studies, 35, 758–770. http://dx.doi.org/10.1080/02614367.2016.
et al., 2012) by experimentally testing the effects of idealised fit- 1182202
Jong, S. T., & Drummond, M. J. N. (2016b). Hurry up and ‘like’ me: Immediate feedback
ness images compared to thin ideal images on body image, under on social networking sites and the impact on adolescent girls. Asia-Pacific Journal
the framework of Social Comparison Theory. Furthermore, the cur- of Health, Sport and Physical Education, 7, 251–267. http://dx.doi.org/10.1080/
rent study was the first to investigate the impact of these images 18377122.2016.1222647
Jung, J., & Lee, S.-H. (2006). Cross-cultural comparisons of appearance self-schema,
on actual exercise behaviour. The results add to the growing evi- body image, self-esteem, and dieting behavior between Korean and U.S. women.
dence that athletic ideal media images can be just as detrimental Family and Consumer Sciences Research Journal, 34, 350–365. http://dx.doi.org/
as the traditional thin ideal (Benton & Karazsia, 2015; Homan 10.1177/1077727x06286419
Levine, M. P., & Murnen, S. K. (2009). Everybody knows that mass media are/are not
et al., 2012), if not more so. In addition, despite the notion that
[pick one] a cause of eating disorders: A critical review of evidence for a causal
these images should inspire fitness, no evidence was found in link between media, negative body image, and disordered eating in females.
the present study for any impact on exercise behaviour. As a Journal of Social and Clinical Psychology, 28, 9–42. http://dx.doi.org/10.1521/jscp.
2009.28.1.9
result, the findings of this study advise against excessive expo-
Mitchell, K. S., Mazzeo, S. E., Rausch, S. M., & Cooke, K. L. (2007). Innovative
sure to fitspiration images that promote the athletic ideal. Future interventions for disordered eating: Evaluating dissonance-based and yoga
research is now needed to examine different possible mechanisms interventions. International Journal of Eating Disorders, 40, 120–128. http://dx.
by which fitspiration images impact upon body image and exercise doi.org/10.1002/eat.20282
Perloff, R. M. (2014). Social media effects on young women’s body image concerns:
behaviour. Theoretical perspectives and an agenda for research. Sex Roles, 71, 363–377.
http://dx.doi.org/10.1007/s11199-014-0384-6
 L. Robinson et al. / Body Image 22 (2017) 65–71 71

Poobalan, A. S., Aucott, L. S., Clarke, A., & Smith, W. C. (2012). Physical activity atti- Smolak, L., & Levine, M. P. (2015). Body image, disordered eating, and eating disor-
tudes, intentions and behaviour among 18–25 year olds: A mixed method study. ders. In L. Smolak & M. P. Levine (Eds.), The Wiley handbook of eating disorders
BMC Public Health, 12, 640. http://dx.doi.org/10.1186/1471-2458-12-640 (pp. 3–10). Chichester, UK: John Wiley & Sons, Ltd. http://dx.doi.org/10.1002/
Prichard, I., & Tiggemann, M. (2008). Relations among exercise type, self- 9781118574089.ch1
objectification, and body image in the fitness centre environment: The role of Stice, E., Spangler, D., & Agras, W. S. (2001). Exposure to media-portrayed thin-ideal
reasons for exercise. Psychology of Sport and Exercise, 9, 855–866. http://dx.doi. images adversely affects vulnerable girls: A longitudinal experiment. Journal of
org/10.1016/j.psychsport.2007.10.005 Social and Clinical Psychology, 20, 270–288. http://dx.doi.org/10.1521/jscp.20.3.
Prichard, I., & Tiggemann, M. (2012). The effect of simultaneous exercise and expo- 270.22309
sure to thin-ideal music videos on women’s state self-objectification, mood Thompson, J. K., & Heinberg, L. J. (1999). The media’s influence on body image distur-
and body satisfaction. Sex Roles, 67, 201–210. http://dx.doi.org/10.1007/s11199- bance and eating disorders: We’ve reviled them, now can we rehabilitate them?
012-0167-x Journal of Social Issues, 55, 339–353. http://dx.doi.org/10.1111/0022-4537.00119
Ramme, R. A., Donovan, C. L., & Bell, H. S. (2016). A test of athletic internalisation Thompson, J. K., Heinberg, L. J., & Tantleff, S. (1991). The Physical Appearance Com-
as a mediator in the relationship between sociocultural influences and body parison Scale (PACS). The Behavior Therapist, 14, 174.
dissatisfaction in women. Body Image, 16, 126–132. http://dx.doi.org/10.1016/j. Tiggemann, M. (2004). Body image across the adult life span: Stability and change.
bodyim.2016.01.002 Body Image, 1, 29–41. http://dx.doi.org/10.1016/S1740-1445(03)00002-0
Sabiston, C. M., & Chandler, K. (2009). Effects of fitness advertising on weight and Tiggemann, M., & McGill, B. (2004). The role of social comparison in the effect of
body shape dissatisfaction, social physique anxiety, and exercise motives in a magazine advertisements on women’s mood and body dissatisfaction. Journal
sample of healthy-weight females. Journal of Applied Biobehavioral Research, 14, of Social and Clinical Psychology, 23, 23–44. http://dx.doi.org/10.1521/jscp.23.1.
165–180. http://dx.doi.org/10.1111/j.1751-9861.2010.00047.x 23.26991
Schaefer, L. M., Burke, N. L., Thompson, J. K., Dedrick, R. F., Heinberg, L. J., Calogero, Tiggemann, M., & Zaccardo, M. (2015). “Exercise to be fit, not skinny”: The effect of
R. M., . . . & Swami, V. (2015). Development and validation of the Sociocul- fitspiration imagery on women’s body image. Body Image, 15, 61–67. http://dx.
tural Attitudes Towards Appearance Questionnaire-4 (SATAQ-4). Psychological doi.org/10.1016/j.bodyim.2015.06.003
Assessment, 27, 54–67. http://dx.doi.org/10.1037/a0037917 Tiggemann, M., & Zaccardo, M. (2016). ‘Strong is the new skinny’: A content analysis
Schaefer, L. M., & Thompson, J. K. (2014). The development and validation of the of #fitspiration images on instagram. Journal of Health Psychology, http://dx.doi.
Physical Appearance Comparison Scale-Revised (PACS-R). Eating Behaviors, 15, org/10.1177/1359105316639436
209–217. http://dx.doi.org/10.1016/j.eatbeh.2014.01.001 Vaterlaus, J. M., Patten, E. V., Roche, C., & Young, J. A. (2015). #gettinghealthy: The
Sensis. (2016). Sensis Social Media Report 2016: How Australian people and businesses perceived influence of social media on young adult health behaviors. Computers
are using social media.. Retrieved from https://www.sensis.com.au/about/our- in Human Behavior, 45, 151–157. http://dx.doi.org/10.1016/j.chb.2014.12.013
reports/sensis-social-media-report/sensis-social-media-report-archive Wheeler, L., & Miyake, K. (1992). Social comparison in everyday life. Journal of Person-
Shilling, C., & Bunsell, T. (2009). The female bodybuilder as a gender outlaw. Qual- ality and Social Psychology, 62, 760–773. http://dx.doi.org/10.1037/0022-3514.
itative Research in Sport and Exercise, 1, 141–159. http://dx.doi.org/10.1080/ 62.5.760
19398440902909009
Simpson, C. C., & Mazzeo, S. E. (2016). Skinny is not enough: A content analysis of
fitspiration on pinterest. Health Communication, 1–8. http://dx.doi.org/10.1080/
10410236.2016.1140273