Introduction

Background

In the last few years, increasing attention has been paid to the study of positive body image, a multifaceted construct defined has the detention of appreciative, accepting and protective attitudes and behaviours with respect to one’s own body characteristics [1, 2]. This shift in body image literature, to a greater focus on factors that support psychological health and well-being, was moved by the need to increase knowledge about predictors, consequences, and methods to promote positive body image, which hopefully may result in a more effective prevention of body image-related disturbances [3]. Body appreciation is one of the most explored key components of positive body image, and it is defined by Avalos and colleagues [1] as the deliberate (a) acceptance of one’s body, including the recognition of flaws and imperfections, (b) respect and care for one’s body’s needs, and (c) rejection of unrealistic socially-prescribed beauty standards. Recent findings on positive body image have documented that body appreciation is connected to both interpersonal factors, such as caregiver attitudes, adult attachment anxiety and avoidance, and peer criticism; and intrapersonal abilities, some of which are self-compassion, perfectionism, and motivation for physical activity [4, 5].

The female body constitutes an elementary object of social evaluation [6], thus, it seems important to analyse social context factors when studying body appreciation. On this subject, a recent study [7] corroborated the social importance of appearance for emerging adult women and found that these women are more likely than men to feel that being more attractive is linked with higher success, to perceive media and peer messages about the pertinence of being thin, and to receive more negative messages about appearance. Indeed, literature suggests that, in women, body dissatisfaction may constitute a warning signal of social insecurity, and that drive for thinness and associated dieting behaviours may emerge as strategies to achieve a favourable social rank position (that is the perception that one is accepted and valued within a certain social group [6]). In this line, several interpersonal factors have been linked to body dissatisfaction as well as with its consequent body image-concealing behaviours. For instance, a recent study by Ferreira et al. [8] suggested that the effect of the absence of early warmth and safeness experiences with close others is associated to disordered eating behaviours, via higher perfectionist self-presentation of body image. Although literature has not widely covered the role of early memories of peer relationships (i.e., friends and colleagues), its lack has been recently associated to symptoms of eating disorders, via an increased perception that one and others hold unfavourable opinions about the self [9]. Additionally, Vangeel et al. [10] conducted a longitudinal study which found that the internalization of appearance ideals in adolescence is linked to long-term self-objectification and body surveillance behaviours, that is still evident in the age group of emerging adults. Regarding positive body image, the role of others has been reviewed, but only regarding body image-specific influence from parents and peers. For instance, a study by Avalos and Tylka [11] revealed that, in young adult females, both general and body unconditional acceptance from others predicted body appreciation and intuitive eating [12]. Furthermore, Wood-Barcalow, Tylka, and Augustus-Horvath [13] found that positive body image among college women is highly promoted by messages of unconditional acceptance from significant others, namely family members, friends and partners.

Self-compassion, although proved to be a powerful antidote for negative body image, and for disordered eating behaviours [14], was only recently proposed as an important construct for the study of positive body image. In fact, fresh accounts have shown that self-compassion associates positively with body appreciation and with intuitive eating [15]. Moreover, data on self-compassion have also underlined its role as an enhancer of body appreciation in the presence of body image-related threats [16], and as a buffer of the harmful effect of general feelings of shame in body appreciation levels [17]. This ability to be mindfully kind to oneself in times of suffering, and to tolerate and understand pain and suffering as common and shared experiences [18] is embedded in an emotion regulation system defined as the safeness-soothing system [19, 20]. This system is triggered by signals of care and compassion from others, and thus has evolutionarily evolved to register social safeness and promote feelings of affiliation, safeness and connectedness [19, 20]. Social safeness, the experience of the social world as warm, calming and reassuring in response to caring behaviours from others [21] is, in fact, considered to be an effective protector against psychosocial suffering [22]. Recently, Kelly and Dupasquier [23] reported that social safeness mediates the relationship between memories of parental warmth and the capacity for self-compassion.

Aims

Taking together the extant information on the matters of recalled peer warmth, self-compassion and social safeness, and also the need to explore roots and correlates of the facets of positive body image, the present study aims at analysing whether, in Portuguese emerging adult women, the capacity of recalling early memories of warmth and safeness with peers associates with higher self-compassion abilities and with higher sense of social safeness. Moreover, the main aim of the study is to explore how these variables associate with the detention of appreciative and protective attitudes toward one’s body characteristics, as well as with the display of symptoms of eating disorders. It is expected that the recall of early memories of warmth and safeness with peers associates with higher self-compassion and social safeness, and that these three variables predict higher body appreciation. In addition, it is hypothesized that the interaction of all these variables may predict a lower display of disordered eating. In the studied model, the effect of Body Mass Index (BMI) was controlled for, given that it is expected that it holds a significant impact on variables directly associated to body image, i.e., body appreciation and disordered eating.

Method

Participants and procedure

The Ethics Committee of the Faculty of Psychology and Educational Sciences of the University of Coimbra, as well as boards of the institutions involved, were fully informed about the purposes of the study, and assured the absolute compliance with ethical requirements, providing their approval. Aims of the study were presented to potential participants (women from several distinct organizations such as higher education institutions, private companies, retail services), who were invited to volunteer as participants. A total of 738 women agreed to participate in the study and were familiarized with the nature and rationale of the study, as well as the voluntary character of their participation, and data’s confidentiality. Women who agreed to take part in the study provided their written informed consent, and only then were given a set of self-report questionnaires to be responded to during a break (authorized by the institution’s boards) of approximately 15 min. According to the main purpose of the study, data-cleaning procedures were conducted to exclude participants who were younger than 18 and older than 25 years old (to achieve a sample only composed of emerging adult women), and the cases in which 15% or more of the responses were missing from a questionnaire, which resulted in a final sample of 387 women.

Women, in the final sample, were aged between 18 and 25 years old (M = 21.64; SD = 1.70). Mean BMI was of 21.58 (SD = 3.26). BMI values of the sample were in accordance with those typically found in the female Portuguese population of the same age group [24], i.e., 12.9% (n = 50) were underweight, 74.2% (n = 287) presented normal BMI, 10.8% (n = 42) were overweight, and 2,1% (n = 8) fell within the obese category, according to WHO [25].

Measures

BMI was calculated by dividing participants’ reported body weight by the square of their self-reported height (kg/m2).

Early Memories of Warmth and Safeness Scale—Peers Version (EMWSS_Peers; [26]) assesses positive memories of early caring relationships with peers (i.e., with friends and colleagues). It comprises 12 items (such as “I felt happy when I was with my friends”, and “I could easily be soothed by my friends when I was unhappy”) rated on a Likert-type scale, ranging from 0 (“No, never”) to 4 (“Yes, most of the time”). Participants are asked to rate the extent to which each statement reflects their childhood experiences. The original study of the EMWSS_Peers reported its excellent internal consistency (α = 0.97) and construct, concurrent, and divergent validities.

Self-Compassion Scale (SCS; [27, 28]) is a 26-item questionnaire which assesses self-compassion, as individuals’ attitudes and actions towards themselves in difficult times. It is composed of six subscales: three positive (self-kindness, common humanity, and mindfulness) and three negative (self-judgment, isolation, and over-identification). The items are rated on a Likert-type scale, ranging from 1 (“Almost never”) to 5 (“Almost always”). According to the purposes of the present study, only the composite measure of the three positive subscales of the SCS was used and defined as “self-compassion” (SCS_SC). This dimension comprises items such as “When things are going badly for me, I see the difficulties as part of life that everyone goes through”, “I try to be loving towards myself when I’m feeling emotional pain” and “When something painful happens I try to take a balanced view of the situation”. The “self-compassion” dimension is proved as a valid measure of self-compassion abilities (with a reported Cronbach’s alpha value of 0.91; [26]).

Social Safeness and Pleasure Scale (SSPS; [21]) aims at evaluating positive emotional experiences (e.g., belonging, warmth from others) which occur within a soothing social environment, that is social safeness. The SSPS is composed of 11 items (e.g., “I feel a sense of warmth in my relationships with people”, “I feel easily soothed by those around me”) to be rated on a 5-point scale from 0 (“Almost never”) to 4 (“Almost all the time”). The SSPS has been proved as a psychometrically reliable measure (α = 0.92).

Body Appreciation Scale-2 (BAS; [29, 30]) evaluates a positive body image-related construct—body appreciation—defined as a posture of acceptance and protection toward one’s own body’s features, despite the assumption of a certain degree of body dissatisfaction. It comprises 10 items (e.g., “I respect my body”, and “I feel that my body has at least some good qualities”) to be rated for frequency of occurrence (Likert-style scale, ranging between 1—“Never” and 5—“Always”). The BAS has proved to be a psychometrically sound measure (α = 0.97, original study; α = 0.95, Portuguese adaptation study).

Eating Disorder Examination Questionnaire (EDE-Q; [31, 32]) is a 36-item (such as “Have you tried to exclude from your diet any foods that you like to influence your shape or weight (whether or not you have succeeded)?”, and “Have you felt fat?”) instrument designed to evaluate the display of disordered eating attitudes and behaviours. The EDE-Q comprises four subscales: restraint, eating concern, shape concern, and weight concern. Respondents are asked to rate the items for frequency of occurrence or for severity of key eating psychopathology symptoms, within a 28-day time frame. In the present study, only the global EDE-Q score was used, which consists of the mean value of scores for all four subscales. Previous studies have shown EDE-Q’s good psychometric properties (α = 0.94).

Analytic strategy

To explore the characteristics of the sample in terms of the variables in study, descriptive statistics (e.g., means and standard deviations) were used. Also, to analyse the associations between early positive memories with peers (EMWSS_Peers scale), BMI, self-compassion (SCS_SC composite measure), social safeness (SSPS scale), body appreciation (BAS scale), and disordered eating (EDE-Q scale), product-moment Pearson correlation analyses were conducted. These analyses were computed by resorting to the software IBM SPSS (v.22; SPSS Inc., Chicago, IL, USA).

A series of path analyses inspecting assumed structural relations (direct and indirect effects) between the variables were presented in the proposed theoretical model (Fig. 1) and were conducted using the software AMOS (v.22, SPSS Inc., Chicago, IL, USA). This theoretical model was designed to test if early memories of warmth and safeness with peers and BMI (exogenous variables) would associate with disordered eating (endogenous variable) when mediated by self-compassion, social safeness and body appreciation (endogenous mediator variables).

Fig. 1
figure 1

Final path model. Standardized path coefficients among variables are presented. *p < 0.05; ***p < 0.001

Full size image

The estimation method used in these analyses was the maximum likelihood, with 95% confidence interval, to test for the significance of the regression coefficients, and to compute fit statistics. Significance of direct, indirect and total effects was measured by Chi-square tests. The Bootstrap resampling method was used to test the significance of the mediational paths, using 5000 Bootstrap samples and 95% confidence intervals [33]. Several goodness-of-fit measures were used to assess the adequacy of the overall model: Chi-Square (χ2), Normed Chi-Square (χ2/df), Tucker Lewis Index (TLI), Comparative Fit Index (CFI), and Root-Mean Square Error of Approximation (RMSEA) with 95% confidence interval.

Results

Descriptive statistics and correlations

Means and standard deviations are presented in Table 1. Correlation analyses’ results (Table 1) showed negative correlations of BMI with early positive emotional memories with peers (albeit weak) and with body appreciation, and a positive moderate correlation with eating psychopathology. Early positive emotional memories with peers correlated positively with self-compassion, social safeness (with strong magnitude) and body appreciation, and negatively and weakly with eating psychopathology. Self-compassion, social safeness and body appreciation presented positive associations between each other, and negative correlations with eating psychopathology.

Table 1 Cronbach’s alphas (α), means (M), standard deviations (SD), and intercorrelation scores on self-report measures (N = 387)
Full size table

Path analysis

The proposed theoretical model was tested through a fully saturated initial model which consisted of 27 parameters. This model accounted for 15% of self-compassion, 31% of social safeness, 37% of body appreciation, and 51% of eating psychopathology’s variances. Nonetheless some of the paths were not significant, and thus were sequentially eliminated: (a) the direct effect of early positive memories with peers on eating psychopathology (bEMWSS−Peers =0.00; SEb = 0.00; Z = 0.18; p = 0.859); (b) the direct effect of social safeness on eating psychopathology (bSSPS = 0.00; SEb = 0.01; Z = − 0.14; p = 0.888); (c) the direct effect or early positive memories with peers on body appreciation (bEMWSS−Peers = 0.00; SEb = 0.00; Z = 0.52; p = 0.603); (d) the direct effect BMI on self-compassion (bBMI = − 0.01; SEb = 0.01; Z = − 0.60; p = 0.547), and (e) on social safeness (bBMI = − 0.05; SEb = 0.10; Z = − 0.48; p = 0.631); and, finally, (f) the direct effect of self-compassion on eating psychopathology (bSCS_SC = 0.09; SEb = 0.07; Z = 1.27; p = 0.205). Then, the model was readjusted, and effects’ sizes were recalculated.

The readjusted model (Fig. 1) explained 17% of self-compassion and 29% and social safeness’ variances and accounted for 36% of the variance of body appreciation, and for 51% of eating psychopathology’s variance. All path coefficients showed to be statistically significant (p < 0.050), and model fit indices revealed an excellent fit to the empirical data [χ2(6) = 2.52, p = 0.867, CMIN/df = 0.41; TLI = 1.01; CFI = 1.00; RMSEA = 0.00, p = 0.984, 95% CI = 0.00 to 0.03]. Early memories of warmth and safeness with peers (EMWSS_Peers) showed positive direct effects of 0.41 on self-compassion (SCS_SC) (bEMWSS−Peers = 0.02; SEb = 0.00; Z = 8.30; p < 0.001), and of 0.54 on social safeness (SSPS) (bEMWSS−Peers = 0.40; SEb = 0.03; Z = 13.15; p < 0.001). The effect of early memories of warmth and safeness with peers on body appreciation (BAS) was significant only at an indirect level, i.e., when mediated by self-compassion and social safeness (β = 2.93; 95% CI = 0.23–0.37). The effect of early memories of warmth and safeness with peers on disordered eating (EDE-Q) was also indirect, carried by the effects of self-compassion, social safeness, and body appreciation (β = − 0.18; 95% CI = − 0.23 to − 0.13). Regarding self-compassion (SCS_SC), results showed a positive direct effect of 0.31 on body appreciation (bSCS_SC = 0.38; SEb = 0.06; Z = 6.57; p < 0.001), and an indirect effect on disordered eating, mediated by body appreciation (β = − 0.18; 95% CI = − 0.25 to − 0.12). Social safeness (SSPS) had a positive direct effect of 0.31 on body appreciation (bSSPS = 0.03; SEb = 0.01; Z = 6.90; p < 0.001), and an indirect effect on disordered eating, mediated by body appreciation (β = − 0.19; 95% CI = − 0.27 to − 0.12). Body appreciation (BAS) was showed to have a direct negative effect on disordered eating (β = − 0.60; bBAS = − 0.81; SEb = 0.05; Z = − 16.16; p < 0.001). Finally, BMI revealed a negative direct impact on body appreciation (β = − 0.26; bBMI = − 0.06; SEb = 0.01; Z = − 6.30; p < 0.001), and a total effect of 0.40 on disordered eating which was revealed as being both direct (β = 0.24; bBMI = 0.08; SEb = 0.01; Z = 6.36; p < 0.001), and indirect, mediated by body appreciation (β = 0.16; 95% CI = 0.11–0.21).

In summary, the accessibility to early memories of warmth and safeness with peers associates with an attitude of acceptance and care toward one’s own body, via a higher ability to be compassionate to oneself in times of suffering, and a safe and enjoyable experience of the social context (social safeness). Moreover, body appreciation associates with a lower expression of disordered eating attitudes and behaviours. Regarding BMI, this variable was showed to be exclusively associated with body appreciation (inversely), and with disordered eating (directly), and not with self-compassion or social safeness.

Discussion

The rationale for the present work was based on widely extant research data on constructs which have been found to be important in the understanding of body image, namely peer relationships, self-compassion and social safeness. The present study aimed at analysing the suitability of a theoretical model which hypothesized that the recall of early memories of warmth and safeness with peers would associate with higher levels of self-compassion and feelings of safeness and connectedness, and thus associate with higher body appreciation and lower disordered eating attitudes and behaviours.

Early memories of warmth and safeness, self-compassion and social safeness are all reported in literature as being inversely associated with body dissatisfaction and disordered eating, while BMI has been shown to be positively associated with these outcomes [6, 8]. Thus, it would be expected that these same variables would present opposite direction associations with body appreciation, given that this facet of positive body image implies an appreciative, protective and overall healthy relationship with body image and function [2]. Correlation analysis’ results were as expected: body appreciation presented positive correlations with the recal of early memories of warmth and safeness with peers, self-compassion and social safeness, and negative associations with BMI and disordered eating.

To better clarify these relationships, a theoretical model was tested via path analyses. These analyses suggested that, in women, the recall of early memories of warmth and safeness with peers is not directly associated with higher body appreciation, in fact, its significant and positive effect is carried by self-compassion and social safeness. Specifically, self-compassion and social safeness appeared as significant mediators of the relationship between recalling early memories of warmth and safeness with peers and body appreciation. Firstly, the finding that early memories of warmth and safeness with peers significantly predicts both self-compassion and social safeness may be understood under the light of the Tripartite Model of Affect Regulation [19, 20], which argues that the safeness-soothing system grows in parallel with the attachment system, i.e., positive social interactions stimulate this system to emit signals of connectedness, trust and safeness [19, 20]. By stimulating this important low-arousal emotion regulation system, early affectionate and trusting relationships with peers may contribute to the creation of an optimal emotional environment for the development of self-compassionate abilities to deal with failure and suffering, and to the experience of the social world as pleasurable and trustworthy. Our results showed a positive relationship of self-compassion and social safeness with body appreciation. These results confirm and extend previous literature. Firstly, the link between self-compassion and body appreciation has been previously reported in literature [16, 17]. Regarding the association of social safeness and body appreciation, since body dissatisfaction may be considered a symptom of social insecurity in women [6], it was predictable that, in females, a safe social environment would contribute to an appreciative, protective and balanced experience of one’s own body characteristics. The overall model accounted for the explanation of the variance of disordered body and eating attitudes and behaviours (such as diet and over-concern about body image). Specifically, a higher recall of early memories of warmth and safeness with peers associated with higher self-compassion and social safeness, and higher body appreciation, which, in turn, associates with a lower display of disordered eating. Finally, BMI only appeared to have significant effects on body image-related variables, i.e., body appreciation and disordered eating. The relationship between BMI and disordered eating (as measured by EDE-Q) was revealed as being almost equally strong at a direct and indirect level, that is when mediated by body appreciation. This result may suggest that, beyond the impact of an objective measure of body weight vs height, it is the valence of the affective relationship established with one’s body characteristics, despite some level of dissatisfaction, that can make women vulnerable to engage in unhealthy body image-concealing behaviours.

The results of the present study, although promising, should be understood while considering some limitations. Prospective studies could be conducted to validate the nature and direction of the present findings, since cross-sectional designs compromise causal inferences. Regarding the sample used in this study (female emerging adult sample), although deliberate, due to the fact that female body image concerns significantly differ from men’s [34] and is continuously affected by messages from significant others who compare them to media-prescribed appearance standards [1], upcoming studies should explore the generalization of the tested model to other samples, such as male samples and other age groups. Finally, the use of self-report measures may compromise the generalization of the data, and so future studies could benefit from the use of other research methods, namely structured interviews.

In summary, the present work seems to underline the importance of childhood and adolescent experiences to the development of self-to-self and self-to-others secure relationships, with consequences on the way women behave and relate with their unique body characteristics. Specifically, early relationships with peers characterized by a sense of warmth and safeness seem to enhance self-compassionate abilities and the capability of feeling safe around and connected to others. In consequence, both intrapersonal and interpersonal factors (self-compassion and social safeness), associate with appreciative and protective attitudes toward body image, which, in turn, associates with a lower tendency to display disordered eating behaviours and attitudes. These results seem to hold important consequences to the development of intervention programs among females, in which the cultivation of self-compassion and the development of affiliative skills may contribute to a more kind and understanding evaluation of their body characteristics and of a health-focused relationship with one’s body. Moreover, when clinically approaching body image and eating-related disturbances, the once again underlined association between self-compassion and body appreciation seems to support the pertinence of applying strategies of Compassion Focused Therapy.