Eating disorders (EDs) are associated with functional impairment, chronicity, psychological comorbidities, medical complications, and increased rates of suicidality and mortality (Stice et al., 2017; van Hoeken & Hoek, 2020). Australian cohort studies suggest that the point prevalence of EDs may be as high as 15% in young females (e.g., Allen et al., 2013). The consequences and prevalence of ED symptomatology, in combination with access to effective ED treatments remaining limited (Kass et al., 2013; Le et al., 2017), emphasise the need to improve our understanding of the underpinning mechanisms; in particular, those that are modifiable.
Cognitive theories of EDs implicate the relevance of cognitive biases in the processing of body shape and weight-related information in the potential development and maintenance of disordered eating behaviours (e.g., dietary restraint and binge eating), as well as body dissatisfaction (Williamson et al., 1999, 2004). Research has primarily focused on biases in attention, with some research on memory and interpretation biases, and is mostly based on cross-sectional research paradigms. Recent reviews provide support for attentional, memory, and interpretation biases for body image-related stimuli in people with elevated body dissatisfaction (Rodgers & DuBois, 2016) and body-specific attentional biases in people with eating disorders (Stott et al., 2021). There is also some experimental evidence for a causal role of body-specific attentional biases in body dissatisfaction (Smith & Rieger, 2006, 2009).
Theoretical notions have also implicated biases involving judging body stimuli (this can include one’s own body or others) in terms of one particular dimension in the potential development and maintenance of body image and eating disturbances (Williamson et al., 1999, 2004). There is currently no research, however, that has empirically tested for a relationship between biased judgement processing of bodies and body image and eating disturbances. Though previous systematic reviews of cognitive biases (e.g., Rodgers & DuBois, 2016) have conflated the terms ‘judgment bias’ and ‘interpretation bias’ to be synonymous, they refer to two distinct constructs. The methodologies cited in previous reviews indicate that the construct that has been previously assessed is that of interpretation bias – i.e., the tendency to interpret ambiguous information in specific ways (Rodgers & DuBois, 2016). In order to adequately assess judgement biases, the paradigm employed must have the capacity to capture the tendency to judge bodies in terms of one particular dimension (e.g., shape and weight) versus other irrelevant dimensions (e.g., body posture). The biased judgment-dimension task (Dondzilo et al., 2021b) is a recently developed task that both permits such an assessment and has the capacity to reveal associations between judgement processing biases and disorder-specific symptomatology.
Dondzilo et al. (2021b) employed this novel paradigm to examine the role of biased judgement processing of faces in body dysmorphic symptomatology. Female participants were required to make either attractiveness-based judgements or attractiveness-irrelevant judgements (in this study the chosen attractiveness-irrelevant dimension was eye gaze direction) of photographs of faces that they had previously selected as being attractive or unattractive. Such an approach permitted a measure of objective accuracy (i.e., accuracy of judging photos as being attractive or unattractive and eye gaze direction facing right or left). Accurate responses were used to compute a measure of readiness to judge facial stimuli in terms of their attractiveness, rather than in terms of an attractiveness-irrelevant dimension. Results were novel in revealing an association between biased judgement processing of faces and body dysmorphic symptomatology (Dondzilo et al., 2021b). These findings further emphasise the importance of addressing this gap in the ED literature, through investigating the relationship between biases in judging bodies and ED symptomatology. Specifically, it is plausible to assume that individuals with higher levels of ED pathology have a greater tendency to disproportionately judge bodies (either one’s own or others) in terms of their shape and weight (i.e., whether they are fat or thin) relative to other dimensions (e.g., clothing, body posture/position). Further, the relationship between biased judgement processing of bodies and other important ED-related constructs, such as ED-specific rumination, has not yet been assessed.
ED-specific rumination, conceptualised as preoccupation with eating and body weight control (Fairburn et al., 2003; Park et al., 2011, is a salient process in ED symptomatology. In support of this notion, a recent meta-analysis revealed a robust relationship between ED-specific rumination and ED symptomatology (Smith et al., 2018), with some evidence that ED-specific rumination may predict subsequent body dissatisfaction (Smith et al., 2021b) and binge eating episodes (Smith et al., 2021a). Research further suggests that ED-specific rumination may serve to underpin the relationship between cognitive biases and ED symptomatology. Specifically, research provides converging support for an indirect relationship between selective attentional processing and ED symptoms via ED-specific rumination (Berrisford-Thompson et al., 2021; Dondzilo et al., 2017, 2021a). These findings implicate that selective attentional processing of idealised bodies may contribute to rumination on eating, shape, and weight concerns, which in turn serves to contribute to body dissatisfaction and dietary restraint. While no research has yet investigated the relationship between ED-specific rumination and other cognitive biases, such as judgement biases, it is plausible to assume that a relationship would exist. Specifically, it is hypothesised that when confronted with bodies, individuals who have a greater dispositional tendency to ruminate about shape and weight have a speeded tendency to judge bodies in terms of their shape and weight rather than in terms of other dimensions.
ED-specific rumination consists of two distinct subtypes, namely reflection and brooding. ED-specific reflection refers to contemplative, intentional pondering, with a focus on actively gaining insight into ED symptoms, whereas ED-specific brooding refers to passive comparison of one’s current situation, which is predominantly focused on the control of eating, shape, and weight, with some ideal standard (Cowdrey & Park, 2012). Critically, research suggests that ED-specific reflection and brooding function differently with respect to their associations with ED outcomes. Specifically, there is some evidence that rumination subtypes predict different ED outcomes (Smith et al., 2021b), and that reflection is more strongly related to ED symptomatology in people with anorexia nervosa (Cowdrey & Park, 2012). This important distinction between rumination subtypes raises the intriguing possibility that ED-specific reflection and ED-specific brooding are associated with different cognitive mechanisms. Although prior research examining the association between attentional biases and ED-specific rumination did not differentiate between rumination subtypes (Berrisford-Thompson et al., 2021; Dondzilo et al., 2017, 2021a), research in the depression domain has shown the depressive brooding, but not reflection, was associated with an attentional bias for negative information (Duque et al., 2014; Joormann et al., 2006; Owens & Gibb, 2016). Therefore, it is entirely plausible that biased judgement-processing of bodies is associated with only one of the rumination subtypes.
In light of the above, the present study sought to investigate the relationships between biased judgement processing of bodies, ED-specific rumination subtypes, and ED symptomatology. Female undergraduate students were recruited to complete the novel biased judgement dimension task developed by Dondzilo et al. (2021b) in order to assess variation in the tendency to judge bodies on different dimensions. In order to ensure the utmost experimental rigour in the assessment of biases in judging bodies in terms of shape and weight, it was necessary to employ an equal number of stimuli in the experimental task that participants would judge as being either thin or fat. Thus, it was necessary to first expose participants to an image selection task to choose images of bodies that they subjectively perceived as being either thin or fat. The use of image selection tasks for the purpose of generating personalised stimulus sets for use in cognitive tasks where participants are making subjective judgments (e.g., attractiveness-based judgements), is in line with prior research (Dietel et al., 2023; Dondzilo et al., 2021b, 2022). It was hypothesised that a greater readiness to judge bodies in terms of their shape and weight would be associated with greater ED symptomatology.1 Given evidence for an indirect relationship between attentional bias and ED symptomatology via ED-specific rumination (Berrisford-Thompson et al., 2021; Dondzilo et al., 2017, 2021a), the opportunity was taken to examine the potential indirect relationship between judgement processing bias and ED symptomatology via ED-specific rumination subtypes. Specifically, two exploratory indirect effect models were considered: i.e., a model which considered ED-specific reflection as the intermediary variable (Fig. 1a) and a model which considered ED-specific brooding as the intermediary variable (Fig. 1b).
Fig. 1
Conceptual diagrams for ED-specific reflection as the intermediary variable in A and for ED-specific brooding as the intermediary variable in B. Path a represents the relationship between biased judgement processing of bodies and the proposed intermediary variable. Path b represents the relationship between the proposed mediator and ED symptomatology when controlling for the proposed intermediary variable. Path c represents the relationship between biased judgement processing of bodies and ED symptomatology (total effect). Path c’ represents the relationship between biased judgement processing of bodies and ED symptomatology when controlling for the proposed intermediary variable (direct effect). The indirect effect is the product of a × b
×
![]()
Method
Participants
Three-hundred-and-forty-four participants (Mage = 19.05, SDage = 1.23) were recruited online through the Research Experience Program sign-up system for first-year psychology students at the University of Melbourne. Only females aged between 17 and 24 years who were fluent in English were eligible to participate. The mean BMI of the sample was 21.71 (SD = 4.74), and 8.14% reported having previously received a formal eating disorder diagnosis. Participants identified as Asian (62.20%), Caucasian (28.50%), and other (9.30%).
Procedure
Following approval from the University of Melbourne Human Research Ethics Committee, the study was advertised via the online Research Experience Program system. Recruitment commenced in May 2020, and data was collected from May 2020 to May 2021.
After signing up for the study, participants were provided with a web link directing them to the online Qualtrics survey platform. Upon providing informed consent, they were asked to complete a series of measures assessing sociodemographics, ED symptomatology and ED-specific rumination. Following completion of the Qualtrics measures, participants were redirected to complete the biased judgment-dimension task via Inquisit Web. Prior to completing this task, participants completed an image selection task, where they selected body images subjectively perceived as thin and as fat. These individualised images were employed in the biased judgment-dimension task. Participants were then debriefed and compensated with one course credit.
Measures
Sociodemographics
Participants indicated their age, current self-reported height and weight (for calculation of body mass index; [BMI]), ethnicity, and whether they had ever been formally diagnosed with an ED.
ED Symptomatology (EDE-Q 6.0; Fairburn & Beglin, 2008)
The Eating Disorder Examination Questionnaire 6.0 (EDE-Q 6.0) assesses the self-reported presence and severity of ED symptomatology within the previous 28 days. Four subscale scores related to the cognitive features of EDs (i.e., restraint, eating concern, shape concern, and weight concern) are derived from 22 items. Participants responded to each item on a 7-point Likert scale ranging from 0 (no days/not at all) to 6 (every day/markedly). A global score is computed as the average of the four subscale scores, with higher scores indicating greater levels of ED symptomatology. Previous research has demonstrated the sound psychometric properties of the EDE-Q 6.0 in both community (e.g., Mond et al., 2004), and clinical samples (e.g., Calugi et al., 2016).
ED-Specific Rumination (RRS-ED; Cowdrey & Park, 2011)
The Ruminative Response Scale for Eating Disorders (RRS-ED) is a nine-item self-report measure that evaluates rumination pertaining to eating, shape, and weight concerns (i.e., ED-specific rumination; Cowdrey & Park, 2011). The measure consists of two distinct subscales, with three items relating to ED-specific reflection (e.g., “Write down what you think about your eating, weight and/or shape and analyse it”), and six to ED-specific brooding (e.g., “Think ‘why do I have problems with controlling my eating, weight and/or shape when other people don’t?’”). Only these subscale scores were relevant for the purposes of the present study. Participants responded to each item on a 4-point Likert scale ranging from 1 (almost never) to 4 (almost always), with higher scores indicating greater levels of ED-specific rumination. In previous studies, the ED-specific reflection and ED-specific brooding subscales have demonstrated adequate to good internal consistency, test-retest reliability, and convergent validity (Cowdrey & Park, 2011; Smith et al., 2018).
Individualised Stimuli Development Protocol
Stimuli Utilised in the Image Selection Task
The biased judgment-dimension task required each participant to judge bodies in terms of their shape and weight (i.e., thin versus fat) and in terms of a shape- and weight-irrelevant dimension. Given the shape/weight dimension was two-dimensional (i.e., thin versus fat), it was necessary to choose a two-dimensional shape- and weight-irrelevant dimension. Body profile direction was chosen (i.e., profile right versus profile left), which meant selecting photographs of body stimuli facing to the right, and replicating and horizontally flipping this subset, to in turn yield an equivalent subset of bodies facing to the left. This ensured the highest level of experimental control over stimuli, as stimuli only differed in terms of body shape or profile direction. Additionally, focusing on a directional aspect of appearance is consistent with previous research; researchers investigating biased judgment-processing of attractiveness used eye gaze direction as the attractiveness-irrelevant dimension (Dondzilo et al., 2021).
A subset of 16 images were selected from a standardised photographic stimulus set developed by Horndasch et al. (2015). These images were black-and-white photographs of women’s bodies clothed in underwear, with each photograph taken under equivalent conditions. To ensure a wide variety of body shapes and sizes to choose from in the image selection task, an equal number of images representing underweight bodies (ranging between 15 and 17 kg/m2), healthy weight (ranging between 20.7 and 21 kg/m2), overweight (ranging between 25.2 and 29.6 kg/m2), and obese (ranging between 37 and 37.6 kg/m2) were selected (i.e., four per weight category). Critically, each of the bodies depicted in this subset was facing to the right, which permitted the creation of an equivalent profile left subset. The final set of 32 images were utilised in the image selection task.
Image Selection Task
The image selection task was employed to generate an individualised pool of images for each participant that varied widely in perceived body shape (i.e., thinness versus fatness). The selected images were utilised in the biased judgment-dimension task.
Each trial commenced with the presentation of four images depicting bodies with equal representation of the four weight categories (i.e., underweight, healthy weight, overweight, and obese) in order to ensure a wide variety of body shapes and sizes. The specific images selected from each weight category was randomised for each trial and across participants. Above the bodies was an instruction cue indicating, with equal probability, whether the selection process should be based on perceived body shape (i.e., “Please select a good example of a THIN/ FAT body shape by directly clicking on that body”), or profile direction (i.e., “Please select a body which is facing to the RIGHT/ LEFT of the screen [i.e., RIGHT/ LEFT PROFILE] by directly clicking on that body”). Thirty-two trials were presented in total across the image selection task, with an equal number of trials instructing participants to select a thin body, a fat body, a right profile, and a left profile. The purpose of including an equal number of body shape and body profile direction selection trials, was to ensure that neither a shape nor shape-irrelevant processing style was induced prior to the biased judgment-dimension task. See an example of each trial type (i.e., perceived body shape and profile direction) in Fig. 2.
This selection procedure yielded a total of 16 experimental stimuli for use in the biased judgment-dimension task: eight images of subjectively perceived thin bodies, and eight images subjectively regarded as fat bodies. In each case, four of the eight images depicted bodies facing to the right, while the remaining four images depicted bodies facing to the left.
Fig. 2
Perceived Body Shape (A) and Profile Direction Trials (B) of the Image Selection Task
×
![]()
Biased Judgment-Dimension Task
The biased judgment-dimension task (Dondzilo et al., 2021) was employed to assess readiness to judge bodies in terms of their shape and weight relative to profile direction. Key task specifications (i.e., number of trials and stimulus exposure duration) followed prior work (Dietel et al., 2023; Dondzilo et al., 2021b) given that the use of these task parameters in this previous research served to reveal associations between biased judgment-processing and disorder-relevant symptomatology. The task involved eight practice trials, followed by 256 experimental trials. Each trial commenced with the presentation of a fixation cross in the centre of the screen for 1,000 ms. This was followed by a 1,000 ms presentation of a body silhouette in the eight practice trials (see Fig. 3A), and one of the 16 images from the photo selection task in the experimental trials (see Fig. 3B).2 One of four potential questions about the body was then presented. Two of these questions concerned whether the body was previously selected as being a good example of a thin body shape (i.e., “Thin body?”), or a good example of a fat body shape (i.e., “Fat body?”). The remaining two concerned whether the body was facing to the right of the screen (i.e., “Right profile?”), or to the left of the screen (i.e., “Left profile?”). Participants were instructed to respond as quickly and accurately as possible, by pressing the up or down arrow on the keyboard to indicate yes or no, respectively. Response latencies (ms) were recorded, yielding an index that revealed the tendency to judge bodies in terms of their shape and weight, as opposed to in terms of the shape- and weight-irrelevant dimension of profile direction. Each new trial commenced following detection of this response on the previous trial, and incorrect responses for the experimental trials triggered error feedback and a 5 s delay.
The 256 experimental trials were delivered across four blocks, with each block of 64 trials separated by a self-timed rest period. Within each block, each unique body image was displayed four times. Trial types differed on the following dimensions: body shape (thin fat), profile direction (right/left), and the question presented (body shape/ profile direction). Participants were presented with an equal number of each trial type in each block of 64 trials. Order of presentation of both practice and test trials was randomised for each participant.
Fig. 3
Example Trials from Practice Task (A) and the Biased Judgment-Dimension Task (B)
×
![]()
Computation of Judgment Bias Index Scores
Median response latencies (ms) to make correct body shape- and profile direction-related judgments were used to calculate a judgment bias index score for each participant. Prior to computing this bias score, pre-defined criteria in line with prior practice (Dietel et al., 2023; Dondzilo et al., 2021b) was used to handle outlier response latencies. Specifically, response latencies below 200 ms and above 3000 ms were excluded, resulting in the exclusion of 2% of trials. The judgement bias index score was then computed by subtracting the median response latency for shape- and weight-related judgments (ms) from the median response latency for profile-related judgments (ms). Thus, higher scores reflected a greater tendency to judge bodies in terms of their shape and weight rather than in terms of the shape and weight-irrelevant dimension of profile direction.
Statistical Methods
Pearson’s correlations were first conducted to evaluate the associations between judgement processing biases, ED-specific rumination subtypes, and ED symptomatology. Next, the Hayes (2018) PROCESS macro for SPSS, version 3.5.3 was used to evaluate the hypothesised indirect effects (Fig. 1). This macro uses bootstrapping—a nonparametric method based on repeated random resampling of the data with replacement (Hayes, 2018)—to construct an empirical approximation of the sampling distribution and estimate path coefficients and a confidence interval (CI) for the indirect effect. Bootstrapping makes no assumptions about the sampling distributions of the paths a, b, or the indirect effect, ab (Hayes, 2018; Preacher & Hayes, 2008). Further, it is preferred over other methods—such as the Sobel Test—due to higher power, and as such, lower probability of Type II errors (MacKinnon et al., 2002). Critically, process analyses permit testing an indirect effect of X on Y through M in the absence of an association (i.e., total effect) between X and Y (Hayes, 2009; Preacher & Hayes, 2004, 2008). In such cases, it is recommended that the term ‘mediator’ be avoided as there is no relationship to mediate. A failure to test for indirect effects in the absence of a total effect (i.e., completely indirect effects) might lead researchers to miss potentially interesting and important information (Hayes, 2009). Indeed, there is evidence for a completely indirect effect of attentional biases on body dissatisfaction via ED-specific rumination (Dondzilo et al., 2021a). Thus, the absence of an association between judgement processing biases and ED symptomatology did not preclude testing for a completely indirect association via ED-specific rumination subtypes. Following Hayes (2018), the present study computed 95-percentile CIs for indirect effects, using the recommended 5,000 bootstrap samples. Indirect effects were considered statistically significant if the CI did not straddle a zero in between. Given the hypothesised pathway from biased judgment-processing to ED symptomatology via ED-specific rumination (and its two subtypes), PROCESS model 4 was used.
Results
In line with Dondzilo, Dietel, et al.’s (2021b) protocol, participants who demonstrated less than 75% accuracy on the biased-judgment dimension task were excluded, resulting in the exclusion of five participants. Accuracy was very high across the remaining 339 participants, averaging 96% accuracy. Descriptive statistics for questionnaires and the experimental task are presented in Table 1. Observed means and standard deviations for measures of ED symptomatology and ED-specific rumination were in line with previous studies employing undergraduate female samples (Dondzilo et al., 2016, 2017).
Table 1
Descriptive Statistics for Questionnaires and Experimental Task
Variable | M (SD) | Range |
|---|---|---|
EDE-Q 6.0 global score | 2.23 (1.36) | 0.00–5.65 |
RRS-ED reflection subscale score | 5.02 (2.02) | 3–12 |
RRS-ED brooding subscale score | 12.52 (4.85) | 6–24 |
Judgment bias index score | 170.02 (132.55) | -128.50–698.00 |
Relationships Between ED Symptomatology, ED-Specific Rumination, and Biased Judgment-Processing of Bodies
A first order Pearson’s correlation revealed that judgment bias index scores (reflecting biased-judgment processing of body shape and weight relative to the body shape- and weight-irrelevant dimension of direction) were not associated with ED symptomatology scores (r = .07, p = .232).3
The observed pattern of associations, however, provided support for a potential complete indirect effect via ED-specific reflection (see Fig. 4a). That is, results revealed a significant association between judgment bias index scores and ED-specific reflection scores (see path a; Fig. 4a). Results also revealed a significant association between ED-specific reflection scores and ED symptomatology scores (see path b; Fig. 4a). A percentile bootstrap CI for the hypothesised indirect effect was entirely above zero (95% CI: 0.03 to 0.16), indicating a statistically significant indirect effect. Both the direct and total effects were not statistically significant. Thus, a completely indirect relationship between judgment bias and ED symptomatology was observed, such that a greater tendency to judge bodies in terms of their shape and weight was associated with greater levels of ED-specific reflection, which in turn was associated with higher levels of ED symptomatology.4
The observed pattern of associations, however, did not provide support for a potential complete indirect via ED-specific brooding (see Fig. 4b). That is, results did not reveal an association between biased judgement-processing and ED-specific brooding (see path a; Fig. 4b). Nonetheless, the significance of the indirect effect via ED-specific brooding was evaluated for sensitivity purposes. The computed CI for the indirect effect contained zero within its range (95% CI: -0.01 to 0.16), indicating no significant indirect effect of biased judgement processing on ED symptomatology via ED-specific brooding.5
Fig. 4
Results from indirect effect analyses. All numbers are standardised OLS regression coefficients.*p < .01; **p < .001. ¥-The 95% percentile confidence intervals indicate a significant indirect effect
×
![]()
Discussion
The overall aim of the present study was to investigate the relationship between biased judgment processing of bodies, ED-specific rumination subtypes and ED symptomatology using a novel measure of biased judgement processing developed by Dondzilo et al. (2021b). Contrary to the hypothesis under test, results revealed no direct association between biased judgement processing of bodies and ED symptomatology. Instead, results provided support for a completely indirect association between biased judgment processing of bodies and ED symptomatology via ED-specific reflection (but not via ED-specific brooding).
This is the first time, to the best of our knowledge, that research has evaluated the relationship between biases in judging bodies and ED symptomatology. Given no evidence for a direct association between biased judgement processing and ED symptomatology, and only evidence for an indirect association through ED-specific reflection, the obtained results suggest that biased judgement processing of bodies might only associate with ED symptomatology via an indirect pathway, rather than a direct one. It is possible, however, that a direct relationship is only observed in people with clinically diagnosed eating disorders, warranting replication of the current study in clinical samples. Alternatively, it could be that a direct relationship only exists when judgments concern one’s own body, as opposed to other’s bodies (as in the present study, where the stimuli employed in the biased judgment-dimension task were other women’s bodies). This suggestion is supported by the attentional bias literature, which has demonstrated different patterns of selective attention for self vs. other bodies in elevated ED symptomatology (Rodgers & DuBois, 2016). Thus, it will be important to evaluate the role of biased judgement processing of one’s own body in future replications of the current study.
The observed indirect effect builds on literature examining the indirect association between biased attentional processing and ED symptomatology through ED-specific rumination (Berrisford-Thompson et al., 2021; Dondzilo et al., 2017, 2021a) by showing that the relationship exists for other types of cognitive biases. Further, the indirect pathway was examined separately for the two ED-specific ruminative subtypes (i.e., ED-specific reflection and ED-specific brooding), whereas prior studies have only considered an indirect pathway via a global measure of ED-specific rumination (Berrisford-Thompson et al., 2021; Dondzilo et al., 2017, 2021a). The finding of an indirect effect only through ED-specific reflection highlights the importance of examining these two ruminative processes separately. Biased judgment processing of bodies arguably represents an analytical type of processing (i.e., it requires an individual to analyse bodies in order to make judgments about a particular characteristic). In turn, this bias may be more closely related to a ruminative process characterised by intentional, analytical processing of information (i.e., ED-specific reflection; Dondzilo et al., 2016), versus a ruminative process characterised by passive processing of information (i.e., ED-specific brooding). It is plausible to presume that ED-specific brooding is thus underpinned by more passive cognitive processes, such as non-volitional attentional processes. This suggestion is supported by research within the field of depression, where studies have demonstrated that depressive brooding, but not reflection, was associated with an attentional bias for negative information, in both non-clinical (Duque et al., 2014; Owens & Gibb, 2016), and clinical samples (Joormann et al., 2006). As such, it may be possible that two individuals who exhibit ED-specific reflection and ED-specific brooding respectively, are both at risk for ED symptomatology, but are characterised by different types of underlying cognitive biases (i.e., biased judgment processing versus attentional bias).
Of course, it should be acknowledged that the effect size (0.09) obtained for the indirect effect via ED-specific reflection is small, and thus suggests that other variables, such as for instance appearance comparisons, may exist. In support of this notion, recent research revealed that attentional bias towards thin-ideal bodies exerted an indirect influence on body dissatisfaction through appearance comparisons and ED-specific rumination (Dondzilo et al., 2021). This invites the possibility of the following serial indirect pathway: an increased propensity to judge bodies in terms of their shape and weight might lead to a heightened tendency to engage in appearance comparisons, in turn increasing ED-specific reflection, leading to elevated ED symptomatology. Future researchers are encouraged to evaluate this proposed serial indirect pathway by manipulating biased judgement processing of bodies and evaluating consequent changes in appearance comparisons, ED-specific reflection, and ED symptomatology.
Given the novelty of the present study, replication studies are warranted in order to establish the observed association more robustly. There would also be value in conducting longitudinal studies to determine whether biases in judgement processing play a role in the onset and/or worsening of ED symptomatology. In turn, this would inform whether biases in judgement processing serve as an appropriate target for therapeutic interventions designed to either prevent the onset of, and/or reduce, ED symptomatology. In support of the idea that biases in cognitive processing may serve as appropriate therapeutic targets, a recent systematic review provided evidence for the effectiveness of procedures which serve to modify cognitive biases modification in reducing ED symptomatology within community samples (Matheson et al., 2019). Critically, preliminary evidence supports the capability of a novel judgement processing manipulation procedure in altering patterns of biased judgement processing, and in turn, disorder-relevant symptomatology (Dondzilo et al., 2022).
There are, of course, a few limitations that deserve acknowledgement. First, the cross-sectional nature of the present study does not allow for causality to be determined for the observed associations. It is worth noting, however, preliminary evidence suggesting that biased judgement processing of faces plays a causal role in body dysmorphic symptoms (Dondzilo et al., 2022). Thus, there would be value in future studies evaluating whether biased judgement processing of bodies plays a causal role in indirectly driving ED symptomatology through ED-specific reflection by manipulating judgement processing of bodies and evaluating consequent changes in rumination and ED symptomatology. Second, the current findings are restricted to young undergraduate females and so it will be important to replicate the current study in other relevant populations, such as men and people with clinical EDs. Finally, a few methodological considerations related to the biased judgment-dimension task should be acknowledged. First, given that the employed body stimuli were black-and-white and did not include heads (see Fig. 2 for examples), cultural differentiation of the stimuli was not readily possible. It would be valuable to replicate the current study using stimuli that both vary in, and permit differentiation of, cultural characteristics, as such characteristics may play a role in how people judge bodies. Second, it is possible that the response format (i.e., up and down arrow keys to signify yes or no responses, respectively) may have caused some interference. However, given that the use of this response format both in the current study and in prior implementations of the biased judgment dimension task served to consistently reveal associations between judgement processing biases and disorder-specific symptomatology (Dietel et al., 2023; Dondzilo et al., 2021b, Dondzilo et al., 2022), such potential interference was evidently negligible. Furthermore, the current study examined readiness to code body stimuli on alternative dimensions using a single stimulus exposure duration of 1,000 ms. This duration did not permit distinction between automatic and controlled information processing (Dondzilo et al., 2021b). Future studies may employ variants of the biased judgment-dimension task that allow for the differentiation between these two types of processing; the key requirement would be to employ a wide range of stimulus exposure durations (e.g., ranging from 200 ms—2, 000 ms; Dondzilo et al., 2021b). Relatedly, it may also be possible that the hypothesised association between biased judgment-processing of bodies and ED symptomatology is driven by automatic information processing, and as such, would only be evident at shorter stimulus exposure durations (i.e., < 500 ms; Dondzilo et al., 2021b).
For the moment, the present findings serve to advance understanding of the types of cognitive processes involved in ED symptomatology. Findings were novel in revealing that a heightened tendency to judge bodies in terms of their shape and weight associated with greater ED-specific ruminative reflection, which in turn associated with greater ED symptomatology. These findings suggest that biases involving judging bodies in terms of their shape and weight are indirectly related to ED-related symptoms via ED-specific ruminative reflection.
Acknowledgements
We would like to thank Minhong Shi for assisting in data collection.
Declarations
Conflict of Interest
The authors declare no conflict of interest.
Informed Consent
Informed consent was obtained from all individual participants included in the study.
Research Involving Human Participants and/or Animals
No animal studies were carried out by the authors for this article.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.