Do women with greater trait body dissatisfaction experience body dissatisfaction states differently? An experience sampling study
Introduction
Dissatisfaction with one’s physical appearance (hereafter referred to as body dissatisfaction) is a common experience, particularly in Western cultures and more so for women than men (Frederick, Forbes, Grigorian, & Jarcho, 2007; Swami et al., 2010). Traditionally, body dissatisfaction has been measured as a stable trait-like construct (e.g., with questions such as, ‘In general, how dissatisfied are you with your appearance?’; Thompson, 2004). Recent research efforts have focused on the experience of body dissatisfaction in daily life (e.g., Melnyk, Cash, & Janda, 2004; Rudiger, Cash, Roehrig, & Thompson, 2007). In large part, this recent interest is driven by the awareness that body dissatisfaction may vary within individuals over time, and this time-varying information may be theoretically and clinically informative.
Using the experience sampling method (ESM; alternatively referred to as ecological momentary assessment), researchers have shown that body dissatisfaction is highly variable within and across days (Colautti et al., 2011, Heron and Smyth, 2013a, Lattimore and Hutchinson, 2010; Mills, Fuller-Tyszkiewicz, & Holmes, 2014; Rudiger, Cash, Roehrig, & Thompson, 2007). Importantly, these shifts in body dissatisfaction states have been linked to a range of contextual factors, such as recent exercise (Fuller-Tyszkiewicz, Skouteris, & McCabe, 2013; LePage & Crowther, 2010) and appearance-related comparisons, especially with more attractive comparators (Leahey, Crowther, & Mickelson, 2007; Ridolfi, Myers, Crowther, & Ciesla, 2011). Elevated state body dissatisfaction may also predict onset of binge episodes (Holmes, Fuller-Tyszkiewicz, Skouteris, & Broadbent, 2014), engagement in dietary restraint efforts (Lattimore & Hutchinson, 2010), and reluctance to engage in social interactions (Mills et al., 2014).
Key features of state-based body dissatisfaction data may also be useful for distinguishing individuals with typical levels of body dissatisfaction from those with clinically-relevant levels of trait body dissatisfaction or related conditions, such as eating disorders. Drawing upon Kuppens and colleagues’ work (e.g., Houbens, van den Noortgate, & Kuppens, 2015; Kuppens, Oravecz, & Tuerlinckx, 2010) on emotional states, we argue that at least three features of state-based data may be relevant for body dissatisfaction: (a) average level of state body dissatisfaction, (b) temporal instability in state body dissatisfaction ratings, and (c) level of correlation among body dissatisfaction states over time (often referred to as inertia).
Insofar as trait-like body dissatisfaction reflects continued, regular experience of body dissatisfaction states, we would anticipate that those with higher trait body dissatisfaction (and related constructs, such as eating pathology) would, on average, tend to have higher state body dissatisfaction ratings in daily life. Consistent with this view, Rudiger et al. (2007) and Melnyk, Cash, and Janda (2004) both found moderate to strong correlations between state mean body satisfaction scores and trait-level body image and eating pathology variables. Importantly, Melnyk et al. demonstrated that mean state body satisfaction scores uniquely contributed to prediction of eating pathology, even after controlling for trait body dissatisfaction.
Individuals may also differ in the extent to which their state body dissatisfaction fluctuates in daily life. While some individuals may be stable in their body dissatisfaction levels over time, others may fluctuate, such that they experience periods of relative satisfaction or intensified body dissatisfaction. These changes may be rapid, or sustained over a number of hours. Moreover, for individuals with regular fluctuations in state body dissatisfaction, the correspondence between her/his average state body dissatisfaction rating and trait-like body dissatisfaction may be weakened as it is more difficult for the individual to appraise their body dissatisfaction in general.
A difference in the magnitude of shift across participants may arise, for instance, if one individual is highly reactive or sensitive to predictors of state body dissatisfaction, whereas the other person is not. Conceivably, more regular shifts in state body dissatisfaction may also occur if a person is more attentive to cues in her/his environment that would prompt these shifts. Indeed, there is empirical support for these notions. Individuals with heightened body image disturbances (including individuals with an eating disorder) tend to exhibit greater sensitivity to appearance threat cues (Espeset, Gulliksen, Nordbo, Skarderud, & Holte, 2012) and show elevated rejection sensitivity (Calogero, Park, Rahemtulla, & Williams, 2010). Moreover, in ESM studies, there is some evidence to suggest that the magnitude and direction of state-based relationships involving body dissatisfaction may depend on one’s level of trait body dissatisfaction (e.g., Leahey, Crowther, & Ciesla, 2011; LePage and Crowther, 2010, Mills et al., 2014). Melnyk et al. (2004) and Rudiger et al. (2007) reported that instability in state body satisfaction was associated with trait-level body image disturbances, with Melnyk et al. also finding an association with eating pathology. However, the moderating effect of instability on the association between trait and state-mean body dissatisfaction remains untested.
Finally, inertia refers to the strength of correlation among state-based ratings from one time-point to the next, and is typically modelled as an auto-correlation (e.g., state body dissatisfaction at time t − 1 predicting state body dissatisfaction at time t). An individual with high inertia experiencing state body dissatisfaction may remain in a state of body dissatisfaction for longer (possibly intensifying over time) than for an individual with lower inertia, who – due to the low autocorrelation – may instead experience a subsequent reversal in state body dissatisfaction level. Such a pattern of high inertia is consistent with the notion of ‘getting stuck’ in a negative mood state (Koval, Kuppens, Allen, & Sheeber, 2012; Kuppens, Allen, & Sheeber, 2010).
We are unaware of any attempts to evaluate whether individuals with trait body dissatisfaction exhibit higher levels of inertia for state body dissatisfaction in daily life. However, the concept of inertia appears compatible with cognitive accounts of body image disturbances, which emphasize that individuals with heightened body dissatisfaction may differ from those with healthier body image in the way they engage in appearance-related stimuli in their environment. Individuals with heightened trait body dissatisfaction appear to be more attentive to negative appearance-related information (Mills et al., 2014), struggle to disengage from these negative stimuli once encountered (Ridolfi et al., 2011), and seem to show a bias towards interpreting ambiguous stimuli as a threat to their appearance (Rodgers & DuBois, 2016; Smeets, Roefs, van Furth, & Jansen, 2008). This hyper-vigilance towards, and difficulty disengaging from, negative appearance-related information is thought to perpetuate body dissatisfaction and related symptoms (e.g., disordered eating). Within the context of experience sampling designs, this may manifest as: (a) longer maintenance of elevated state body dissatisfaction level, and/or (b) stronger associations between consecutive state body dissatisfaction ratings over time for individuals with elevated trait body dissatisfaction. These characteristics may be captured using the concept of inertia.
While these findings offer some insights into the nature of the relationship between state- and trait-level body dissatisfaction, this evidence base remains small and gaps clearly remain. To our knowledge, the relationship between trait body dissatisfaction and inertia has yet to be tested empirically despite theoretical reasons to suspect a link. Similarly, although state mean and trait body image have been shown to strongly correlate (Melnyk et al., 2004, Rudiger et al., 2007), it is unclear whether such associations are contingent upon level of instability an individual experiences. Finally, with the exception of Melnyk et al., there has been insufficient consideration of whether state-based properties of body dissatisfaction may have predictive utility beyond that afforded by trait body dissatisfaction. If characteristics such as inertia and instability are different manifestations of negative body image, then it may be anticipated that they uniquely contribute to body image-related outcomes, such as eating pathology.
The primary aim of the present study is to extend prior investigations of the relationship between state and trait body dissatisfaction. In particular, the present study explored whether: (a) trait body dissatisfaction measured at baseline was associated with mean level, instability, and inertia in state body dissatisfaction measured repeatedly across a one-week testing period; and (b) the strength of correspondence between trait body dissatisfaction and average of state body dissatisfaction ratings is moderated by level of instability during the ESM phase. Given that it is unlikely that trait body dissatisfaction will be perfectly related to these characteristics of state body dissatisfaction, a secondary aim was to evaluate the unique contributions of these state-based body dissatisfaction characteristics to predict eating pathology, controlling for trait body dissatisfaction.
We predicted that average state body dissatisfaction, instability, and inertia will be positively correlated with trait body dissatisfaction (Hypothesis 1a–c) and eating pathology (Hypothesis 2a–c). However, it was further predicted that the correlation between trait and state average body dissatisfaction would be lower for individuals with more unstable state body dissatisfaction ratings during the ESM testing phase (Hypothesis 3). Finally, it was hypothesized that the three state body dissatisfaction characteristics and trait body dissatisfaction would each be significant unique predictors of eating pathology when modelled together (Hypothesis 4).
Section snippets
Participants
One hundred and sixty-one women aged 18–40 (Mage = 22.13, SD = 6.73) were recruited via advertising on websites of eating disorder-related organizations, social media, and through the Australian universities where the authors work. Although the sample was recruited for a broader study exploring the influence of body image on social interactions, we limit reporting to body dissatisfaction and disordered eating measures relevant for present hypotheses.
Self-reported body mass indices (BMI) ranged
Compliance rates
The average number of responses completed per participant (out of a possible 42) was 21.82 (SD = 9.21). Bivariate correlations between compliance rates for ESM surveys and baseline measures were small yet significant for age (r = −0.20, p = .013) and eating pathology (r = −0.22, p = .005), and non-significant for BMI (r = −0.05, p = .54), trait body dissatisfaction (r = −0.07, p = .405), state mean body dissatisfaction (r = −0.04, p = .581), inertia (r = 0.02, p = .811), and temporal
Discussion
Despite growing research interest in state-based conceptions of body dissatisfaction (e.g., Fuller-Tyszkiewicz et al., 2013; Heron and Smyth, 2013a, Heron and Smyth, 2013b, Rogers et al., 2017), much of this research has focused on identifying antecedents to, and consequences of, spikes in state body dissatisfaction. This is in spite of early studies in this area focusing on how state and trait body image constructs relate (e.g., Melnyk et al., 2004, Rudiger et al., 2007). Although these
Acknowledgements
The authors of this manuscript have no conflicts of interest to declare, financial or otherwise, related to this submitted work. Further, all contributors to this manuscript are named as authors.
References (52)
- et al.
Predicting excessive body image concerns among British university students: The unique role of appearance-based rejection sensitivity
Body Image
(2010) - et al.
Screening for anorexia and bulimia nervosa in a college population: Problems and limitations
Addictive Behaviors
(1984) - et al.
Accounting for fluctuations in body dissatisfaction
Body Image
(2011) - et al.
Is intensive measurement of body image reactive? A two-study evaluation using ecological momentary assessment suggests not
Body Image
(2013) - et al.
Tests of an extension of the dual pathway model of bulimic symptoms to the state-based level
Eating Behaviors
(2014) - et al.
Perceived calorie intake and state body image satisfaction in women attempting weight loss: A preliminary investigation
Body Image
(2010) - et al.
The effects of exercise on body satisfaction and affect
Body Image
(2010) - et al.
The frequency, nature, and effects of naturally occurring appearance-focused social comparisons
Behavior Therapy
(2007) - et al.
Ecological momentary assessment of the effects of weight and shape social comparisons on women with eating pathology, high body dissatisfaction, and low body dissatisfaction
Behavior Therapy
(2011) - et al.
Body dissatisfaction and body change behaviors in Chile: The role of sociocultural factors
Body Image
(2008)