Religion and culture can both play intricate and provoking roles in our everyday lives. Most individuals grow up surrounded by culture and religion, even if they do not practice the beliefs or practices. So, when asked “Is a religion like a culture, that a cult is like a culture, or that a culture is like a cult or a religion?”, it can be quite complex. Both the concept of religion and culture can play influential roles, leading to questions regarding the role they have on each other. While cults are often viewed as obscure communities, they often embody many of the same type of belief system and core operations. This question of the influential role religion and culture have on each other is important because many individuals either seek out culture and religion as a way to develop their identity along with others who experience almost the opposite, finding identifying both with a culture and religion to be overwhelming and contradicting at times. Along with this, in many regions, culture, religion, or both are engrained historically in the area, leading to the generational passing of the beliefs, practices, and traditions. This adds even more complication to understanding the relationship between both and how the growth of each operates.

The most intuitive response is that religion and culture are both mutually beneficial and influence each other. This is a very region-based response though. Not everywhere will have strong cultural and religious entities or presence that exist, however in regions where both are present, they seem to exchange information. For example, the culture of Israel is very engrained into the beliefs, practices, and traditions of Judaism, and vice versa, the religion Judaism is heavily embedded into Israeli culture. This is because the idea, morals, and practices were exchanged through social transmission and writing which allowed the culture and religion of the area to grow closed together and share more beliefs over time. However, in areas where religion is not strongly practiced or there is not a strong cultural appearance, there may not be as clear of a defining guide for where the influence comes and goes.

It would be hard to justify religion being above or encompassing culture, in the same manner that it would be hard to justify culture as being above or encompassing all religion. The reality is that many individuals lives full of culture yet have no influence nor partake in any religious activity. While external influences may still exist to a point, they do not seem scientifically a large reason for religiousness. The same is for many religious individuals who are raised with little culture, or at least culture that isn’t created and defined by just the religion. This brings up two final key points. Many religious groups, especially those that are defined as “cults”, allow or embrace little culture into their system. Culture is often seemed to plague or create sin within religions, leading to my second key point. Many religions, with an emphasis on cults, create their own “culture” including things such as traditions, cuisine, language, practices, rules, social norms, and monetary systems. This leads to further questioning into how valid these religion-based cultures are, and how they truly compare to or influence other religiously-detached cultures.

Dulin’s Research in Emotion and Religious Culture

Dulin (2011) takes an observant and analytical approach towards seeing the interactions between religion and culture. He focuses on how religious concepts are socially transmitted with an emphasis on emotion. He insinuates that there needs to be a cross-culturally look at the effects universal emotional processes can have on religion and representation. Throughout the article, he emphasizes how important representation is because emotion shapes cultural representations leading to the different interpretations of religions (Dulin, 2011)

Dulin (2011) hypothesized casual relationships between religion and culture, where religion is socially transmitted through causal chains until they reach the final stage. He discusses how many religious imagery, concepts, and framework are sent through various stages starting at the cognitive causal chain (CC: causal chain) leading to the social cognitive causal chain, and ending at the cultural cognitive causal chain. This is where he really aims to answer the question, is religion like a culture or is culture like a religion?

Dulin (2011) answers that question by explaining that a causal relationship exists, however, that path does not automatically lead to cultural significance. He states that religious representations go through the cognitive CC first, by spreading the idea to another’s person cognitive CC. This always a social cognitive CC to form where individuals will need to first process and externalize their religious beliefs and representations as a way to socially transmit them to someone else. If this social transmission has the stability to become long-lasting then it is finally able to become a cultural cognitive CC (Dulin, 2011).

Through his explanation of the process, he argues that it is a causal process where both religion and culture influence each other, yet also sets up his theoretical process for religion to be like culture. In a sense, religion starts at the bottom and has to develop towards enough stability to be seen as cultural. This places culture at a superior level, where social transmission and externalization of reactions are necessary to motivate it to cultural status. This is partially because Dulin (2011) views culture as long-standing/stable practices, beliefs, and representations. This can be problematic because it disregards cultures in their origin or beginning phases of development while also does not acknowledge that not every culture is or has been stable in history, such as the Mestizo/Indigenous culture in Central America, where both ethnically almost extinct and unstable culturally.

While Dulin (2011) makes great points about how religion can be interpreted differently depending on the cultural background of the individual trying to dissect it, he only utilizes the concept of representations to back this statement up. He focuses on witchcraft and benevolent agents which do not encompass or directly relate to most common religions today. He focused on religions where the imagery was very strong, fear-inducing, and almost negatively exaggerated which contrasted the type of imagery in most common religious readings and tales such as the Bible and Noah’s Ark (Dulin, 2011).

Overall, Dulin (2011) answers the question, “Is a culture like a religion or is a religion like a culture?” by indirectly providing a framework and evidence to show that religion is like a culture. He proves this point using the CCC (cognitive causal chain) framework to prove that religious representations need to undergo processes like social transmission to increase the chance of stability to allow it to develop into a culture. So, while he believes that both religion and culture exist in a causal chain, spreading information between each other, they do not exist on the same level. Culture sits above religion as a stable presence and relies heavily on the human mind (through cognitive causal chains) to determine what representations are worthy of stability. Dulin’s view of culture shaping psychology, vice versa, completely leaves out religion as being an entity of its own outside of the cultural world he has defined. He states universal psychological process influence culture, which automatically influences religion, as it is apart or like culture (Dulin, 2011). This is ignorant of the idea that religion or culture exists in regions of the world where the other concept is not highly present. Such as cults where religion is heavily present yet there is an often a lack of culture due to the risk of corruption or sin.

Saroglou’s Five-Factor Model

The second article takes a critical look at the religious experience, and religiosity as being influenced from culturally adapted personality traits. The author, Vassilis Saroglou, discusses how individual differences in areas such as personality to influence varying levels of religiousness. They opt to use the Five-Factor Model for personality, often referred to as the acronym OCEAN. He presents the argument that two of the factors experienced universally, agreeableness and conscientiousness, have been culturally adapted and now influence religiousness in individuals (Saroglou, 2010).

The first key thing Saroglou (2010) does differently compared to Dulin (2011) is that they focus on how individual differences in personality can influence religiosity instead of the more collective based representations that Dulin used. This adds new depth and really provides new insight to a path outside of what Dulin described where there was lack of consideration for groups where both culture and religion did not exist together such as cults. The framework and evidence Saroglou provide is based on individualism such as personality much more than collective actions, such as the Semai tribe in Dulin (2011). The emphasis on how our personality is culturally adapted provides a new view on whether or not we are raised with culture or religion first. Saroglou throughout his article continues to stress that there is a correlation between the two, where information is being shared, however he does not ever insinuate that religion is like culture or that culture is like religion. More so, he provides the unique perspective that personality, influenced by culture, acts almost like a mediator towards religiousness (Saroglou, 2010).

Saroglou (2010) meta-analyzed seventy-one samples based on his subject of how individual differences affect religiousness. He looked specifically for how the presence of agreeableness and conscientiousness affected the level of religiousness an individual affiliated with. He specifically chose those two factors because he predicted agreeableness would positively correlate with religiousness due to the stress on social harmony and relationships, while conscientiousness would positively correlate with religiousness due to the need for orderliness that can be found in many practices (Saroglou, 2010).

His hypothesis about culturally adapted agreeableness and conscientiousness influencing religion was supported throughout the article, first by the meta-analysis. The results show that both of those factors did positively correlated with religiousness, which was further broken down into religiosity, fundamentalism, and spirituality. His review of related literature provided great supplemental support for his theory about how religiousness can be directly influenced by individual differences. However, he does acknowledge that many of the studies discussed are based on samples with a Christian background, which is not good for diversity of results. He also acknowledges that in the meta-analysis that age ranges are not provided for many of the studies, making it hard to specifically classify data or analyze it to the year of age. His conclusion ties together the three sections of his paper very well and reinforces his answer to the key question (Saroglou, 2010).

To Saroglou (2010) there is association between culture and religiousness where they mutually influence each other. They are both similar each other, and not mutually exclusive, however they are not the same. Culture often shows itself through influences such as personality which can influence other factors such as religion. While religion can influence culture through someone’s personality changing and the sharing of new information, it seems less researched and discussed. More so, specific culturally adapted personality traits (agreeableness and conscientiousness) have been able to play important roles influencing religion such as those traits being present in many religious figures, heroes, or saints. These traits have been able to culturally influence how religious figures are created and perceived, while not dominating or controlling religion. This leaves the relationship between both culture and religion to be mutually beneficial and associated but one is truly not like the other (Saroglou, 2010).

The Intersection of Culture and Religion

There is no accurate method in which we can clearly answer whether religion is like a culture or culture is like a religion. Throughout these articles, it has been shown the mutual relationships that exist between the two concepts. Religion and culture transmit and share information amongst each other, allowing for the growth of new religions, new cultures, and new connections between them. As religion and culture continue to share information, it may become questionable how distinctive each concept will truly be, or if they combine into something much bigger than discussed in these articles. There is no way to accurately say that one is more like the other, both have historically existed and built upon each other through social transmission and individual differences.

To truly answer if culture is like a religion or if religion is like a culture, bountiful information would be required. First, we would have to define and understand how others defined the concepts religion and culture; not everyone has the same answer, and region/age/race could change the interpretation a lot. Along with this, we would have to determine if children develop and internalize religion or culture first to understand which influences the other first. This could become more complex if there is no defined pattern because there will be added emphasis on areas like genetic information, emotional security, and personality as influences over which is developed first. Overall, there is no way to clearly draw lines between the path of influence and social transmission between religion and culture. There can only be inferences made, and those have shown a mutually beneficial relationship where religion and culture often spread together and influence each other. This means there is no superiority or imbalance when it comes to the roles culture and religion have, because it is so individualistic (based on personal needs and differences) yet so collective as the same time (often embodying and rewarding social activities, communication or relationships).

Reference List

Dulin, J. (2011). How emotion shapes religious cultures: A synthesis of cognitive theories of religion and emotion theory. Culture & Psychology, 17(2), 223-240.                     https://doi.org/10.1177/1354067X11398310

Saroglou, V. (2010). Religiousness as a Cultural Adaptation of Basic Traits: A Five-Factor           Model             Perspective. Personality and Social Psychology Review, 14(1), 108–125.             https://doi.org/10.1177/1088868309352322

Abstract

Maternal emotion dysregulation is a debilitating issue that can greatly impact relationships with their children; children can develop problematic behaviors as a way to cope with their mother’s instability of emotions; a child’s gender may also impact their susceptibility to being affected by maternal emotion dysregulation. The Difficulties in Emotion Regulation and the Child Behavior Checklist were utilized to help measure the variables of child behavior and maternal emotion dysregulation; demographic information was used to study gender. It was hypothesized there would be a 1st main effect that children with mothers who are dealing with a clinical level of emotion dysregulation will have higher problem item scores on the Child Behavior Checklist (CBCL) than children who do not have mothers dealing with a clinical level of emotion dysregulation, a 2nd main effect that males will have higher problem item scores on the Child Behavior Checklist than females, and an interaction effect of a child’s mother dealing with a clinical level of emotional dysregulation on CBCL problem item scores will be stronger for females than males. A total of 68 mother-child dyads (Age: M= 48, SD= 7.6 months, Gender of Children: Female (46%) and Male (54%)) participated in a random clinical trial, where mothers completed the DERS and CBCL. The results showed that there was a main effect that children with mothers who experience a clinical level of emotion dysregulation have higher problem item scores on the Child Behavior Checklist (CBCL) than children who do not have mothers experiencing a clinical level of emotion dysregulation. However, there was no evidence of a 2nd main effect or interaction effect. The implications of this study are to help find more information regarding how maternal emotion dysregulation affects children and what types of influences can alter child behavior or cause the development of problem behaviors; along with providing data to improve therapy and treatment methods for mothers and children.

Maternal Emotion Dysregulation Impacts on Child Behavior: Through the Lens of Child Gender

Introduction

Parent-child interactions are a key part of how children learn to socialize and communicate with others. These interactions can help guide children in developing good or bad behaviors. Emotion regulation is essential to helping individuals form healthy relationships and perform everyday tasks in a stable emotional state. Emotion regulation is defined as, “an ability to recognize, control, and express the emotions of oneself and others and an in/external process of monitoring, assessing, and modifying emotional reactions according to individual goals (Hoeksma et al. 2004; Thompson 1994, cited in Suh & Kang, 2020). While emotion regulation has always been viewed as a valuable skill for children to develop, it has rarely been discussed as an important skill for parents to have. Research has been accumulating around the effects maternal emotion regulation could have on child development. Evidence has shown that emotion regulation skills in mothers are important to help them handle their children’s negative emotions while also showing behavior that is acceptable for their child to adapt to. Children look up to their parents and often develop experiencing their parent’s emotions and interactions first-hand. This means that children may develop and replicate their parent’s actions from experience, leading to possible behavioral issues. Maternal emotion dysregulation can set poor boundaries and examples for how children should behave, and some research has found this to be shown more prominently in situations with peers such as in classroom settings.

How a child is able to express and regulate their emotions through their behaviors, is essential to development. Children rely on good behavioral skills to form social skills, educational skills, health habits, and communication. Without these skills, outcomes such as the development of Oppositional Defiant Disorder, or Childhood-onset depression may be more likely. Without proper support, and the ability to learn healthy behavior from an adult, children may be compromised to many other long-term issues such as anger issues, or negative mental health.

Literature Review

The literature reviewed for this paper has researched the effects that maternal emotional regulation can have on the development of toddlers and adolescents. While each journal article touched on its own distinct measures, there was a commonality in the predicted effects maternal emotional regulation could have on child behavior and emotion regulation. Much of the past literature has identified maternal emotional regulation to be impactful for childhood development. This is especially true for behavior because many children learn from and are directly interacting with their parents’ behaviors and emotions. They may grow to have resemblances in their own emotion regulation patterns compared to their parents.

A research study around the program Tuning into Kids stressed the importance of parents having good emotional regulation skills, especially in the area of emotion socialization. The researchers recruited families at school districts located in lower to middle-class areas of Melbourne, Australia. The participants were parent-child dyads where the children were between the ages of three to four years old, and the parents were primarily mothers (207 mothers, 9 fathers). They randomly placed these dyads into either the regular family services and help they were already receiving, or the Tuning into Kids program, which focused on increasing the emotional connection between parents and children. The structured program involved emotional coaching that would help alter parenting beliefs and behaviors. Two keys measures used to help facilitate this study were the Difficulties in Emotion Regulation Scale (DERS; Gratz & Roemer, 2004) and the Eyberg Child Behavior Inventory (ECBI; Eyberg & Pincus, 1999, cited in Havighurst et al., 2010). The results found that parents who went through the program displayed better emotion socialization skills, which in return improved emotion regulation skills. There was a significant improvement in emotional awareness and regulation compared to the control group; along with reported improved empathy, by the end of the study. This study supports the key idea that emotion regulation can have a large impact on parent-child interactions which can influence a child’s behavior. They display this by showing how increased emotion socialization can lead to better handling of emotions.

Emotional socialization becomes a key theme in a lot of the literature surrounding maternal emotion dysregulation because it is a foundation for healthy communication between children and mothers. Another piece of literature reviewed also focused on the effect’s emotion socialization can have on child behavior. Researchers in Turkey conducted a study around the effects emotional socialization could have on child problem behaviors. They recruited participants via local advertising and had the participating mothers fill out a set of questionnaires. Two key measures were the Child Behavior Checklist (CBCL; Achenbach & Ruffle, 2000), and the Parental Reflective Functioning Questionnaire-1 (PRFQ-1; Luyten et al., 2009, cited in Arikan & Kumru, 2020) which were respectively used to measure child problem behaviors, and maternal mentalization. The study found that poor maternal emotion socialization, as an indicator of emotion regulation, did have a negative effect on child problem behaviors, in terms of children internalizing and externalizing problem behaviors. Researchers struggled with identifying which symptoms of emotion regulation, such as anxiety or anger, directly contributed to certain child problem behaviors. They found more generally that poor emotion regulation could root a variety of possible child problem behaviors, without an exact pattern to exist (Arikan & Kumru, 2020).

The last piece of literature did specifically look at gender in children as a possible influencing factor to how mothers regulate their emotions, deal with challenging child behavior, and handle their children’s emotions. Mother-child dyads were recruited and the participating mother completed a set of questionnaires regarding child behavior and maternal emotion regulation. The study found that emotion regulation was consistently higher for females compared to male children and that the mother’s reaction does seem to have an influence on child aggression. They found that mothers of sons displayed more punitive reactions compared to the mothers of daughters; this could explain why the females had better emotion regulation skills (Suh and Kang, 2020).

There were key similarities between the literature reviewed. Most focused-on mothers are the primary parent or caregiver. Along with this, measures with similar constructs were used across the studies reviewed. There was a strong focus on key mediators and symptoms of emotion regulation such as emotional socialization and aggression. This is in part because they have key roles in how individuals are able to cope with and regulate their emotions. The age range was also consistent throughout the studies because it is important to stress that while the studies found similar results, this does not make their research correct. The studies all discussed the possibilities of third variables such as environmental or peer factors that could also affect problem child behaviors, along with the possible inaccuracy of self-report forms filled out by mothers

The studies each focused on their own mediators, and ideas in terms of why emotion regulation plays a role in child behavior. Some find that children will model after this behavior, others find children will develop feeling fearful of it, leading to other behavioral issues. Lastly, many are unsure of how exactly children interpret daily parent-child interactions, especially as toddlers. This leads to a lot of possibilities surrounding why maternal emotion regulation seems to influence problem child behaviors. Each study looked at different defining qualities from socio-economic status to gender; these variables were thought to influence why parents have or change their interaction habits compared to other parents.

Purpose of the Current Study

Our study is going to be able to expand from previous studies in the area by looking at different interactions. Gender has not commonly been researched as an interacting factor in maternal emotional regulation and possible outcomes in child behavior. Much of the prior literature looks at gender in terms of the parents, where the difference between mother and fathers may be key to understanding other issues such as child abuse potentials. However, these articles do not directly look at how that abuse and gender difference can shape long-term child behavior. Along with this, many past studies only utilize aggression to help measure variables like emotional dysregulation, or problem behaviors in children. Our study expands from prior ones while still keeping the core research point of the effects of maternal emotional regulation abilities.

Much of the prior literature has focused either solely on child behavioral issues or parent-child interactions. This has left gaps in research surrounding the role gender plays in child behavioral problems. Along with this, there is not a lot of background literature regarding the effects and differences a mother’s emotion regulation has on behavioral problems for female children compared to male children. Gender has often to be viewed through the parental lens, where researchers are comparing mothers against fathers. While this is because of the possible gender differences in emotional regulation that could affect parenting skills between mothers and fathers, it has not commonly been viewed from the perspective of possible gender differences in emotional regulation for male and female children that could affect behavior. This is especially intriguing because the development of emotion regulation in children could not only be affected by parental emotional regulation (and the differences there within gender) but also their gender. Along with this, much of the prior literature focuses solely on aggression, and depression to fulfill measuring emotion regulation in mothers, while also commonly using aggression to measure child behavior.

Our study is able to expand on previous information by utilizing the new areas of research as described above. By utilizing different testing measures, there will be more information regarding child behavior on a spectrum outside of just aggression. By including gender, there will be the addition of new information regarding interactions it can have on emotion regulation. It will also provide new information regarding the effect’s child gender can have on how they interact and interpret maternal emotional regulation and negative emotions.

Variables and Hypothesis

A questionnaire along with a task was used to help create the categorical and continuous variables for gender, presence of a clinical amount of emotional dysregulation in the mother, and problem item scores on the Child Behavior Checklist. The presence of a clinical level of emotional dysregulation in mothers was measured using the Difficulties in Emotion Regulation (DERS; Gratz & Roemer, 2004). A higher score in total indicates a higher level of emotional dysregulation; scores at or above 96 indicated a clinical level of emotional dysregulation. The demographic information collected from the study was used to help assess the gender variable. Lastly, to help measure child behavior problems, problem item scores from the Child Behavior Checklist (CBCL; Achenbach & Ruffle, 2000) will be utilized. Higher scores on the CBCL indicate the child may struggle with more emotional and behavioral problems.

I hypothesize there will be the main effect that children with mothers who are dealing with a clinical level of emotion dysregulation will have higher problem item scores on the Child Behavior Checklist (CBCL) than children who do not have mothers dealing with a clinical level of emotion dysregulation. Secondly, I hypothesize a main effect that males will have higher problem item scores on the Child Behavior Checklist than females. Lastly, I hypothesize an interaction effect of a child’s mother dealing with a clinical level of emotional dysregulation on CBCL problem item scores will be stronger for females than males.

Methods

Participants

A total of 68 mothers (M= 48, SD= 7.6 months), and their 3 to 4-year-old children, were recruited to participate in an in-person study in a university campus office. The child participants’ gender breakdown was 46% females and 54% males. The racial background of the child participants was: 63.2% European American, 5.9% Latino or Hispanic, 1.5% African American, 29.4% multiple racial and ethnic backgrounds, including individuals who identified their children as Asian American, American Indian, Native Hawaiian, or Pacific Islander. Demographic information regarding the mother’s education level and marital status was also recorded. For completed education level the breakdown was, 6% mothers with some high school attainment, 4.5% completed high school, 35.8% with some college, 14.9% technical school or professional school, 26.9% college graduates, and 12% with postgraduate education. The distribution for marital status included 62% percent were married or had longtime partners, 25% were never married, and 13% were separated, divorced, or widowed and were single heads-of-household. The income of families was distributed into quartiles: (1) less than $17,000, (2) $17,000-$29,000, (3) $29,001-$50,000, and (4) $50,000 or more.

Mothers with elevated Borderline Personality Disorder symptoms were specifically recruited for this study, because of the positive connection between BPD and emotion dysregulation.  Along with this, mothers with low BPD symptoms and low SES were recruited to offset the high prevalence of mothers with low SES who have high BPD symptoms. Mothers were screened for BPD using the McLean Screening Instrument (Zanarini et al., 2003; range = 0-9) and 27% (n = 18) reported 0-1 symptoms, 24% (n = 16) reported 2-4 symptoms, and 49% (n = 33) reported 4 or more symptoms.

Materials

Self-report questionnaires were administered to the participating mothers; these surveys included the DERS and the CBCL, as previously mentioned. Children did complete their own separate task as a part of a larger study, however, that information was not used for the purpose of this paper. Gender was measured using demographic information collected by the researchers. The mother either reported their child as M (male) or F (female). They also collected child race/ethnic information, age information, education level of the mother, the marital status of the mother, and socio-economic status of the family.

The presence of a clinical level of emotion dysregulation in the mothers was measured using the Difficulties in Emotion Regulation (DERS; Gratz & Roemer, 2004), a 36-item measure that uses statements to assess difficulty in regulating emotion. Example items on the DERS include: “I have difficulty making sense out of my feelings”, “When I’m upset, my emotions feel overwhelming”, and “When I’m upset, I have difficulty controlling my behaviors”. Respondents use a 5-point Likert Scale, from 1 (almost never [0-10%]) to 5 (almost always [91-100%]), to gauge how much they experience or relate to the statements. Total scoring works by summing all of the scores for each item together. Questions 1, 2, 6, 7, 8, 10, 17, 20, 22, 24 and 34 are reverse-scored. Subscale scoring consists of these sections: nonacceptance of emotional responses (11, 12, 21, 23, 25, 292), difficulty engaging in goal-directed behavior (13, 18, 20R, 26, 333), impulse control difficulties (3, 14, 19, 24R, 27, 324), lack of emotional awareness (2R, 6R, 8R, 10R, 17R, 34R 5), limited access to emotion regulation strategies (15, 16, 22R, 28, 30, 31, 35, 366), and lack of emotional clarity (1R, 4, 5, 7R, 9). Higher scores indicate higher levels of emotion dysregulation, and a score of 70 is considered a nonclinical community average, while a score of 96 or above indicates a clinical level of emotion dysregulation (Gratz & Roemer, 2004). Reliability for the DERS was strong, Cronbach’s α= .94.

To measure child behavior, the continuous variable, item scores on the CCBL, is being utilized through the Child Behavior Checklist (CCBL; Achenbach & Ruffle, 2000). The CCBL is a 64-item measure that has caregivers mark down how much they identify or feel the items describe their children with the listed behaviors, characteristics, and moods. Example items on the CCBL are: “Easily frustrated” and “Punishment does not change his/her behavior”. The 3-point scale ranges have the choices of, 0 (not true), 1 (somewhat or sometimes true), or 2 (very true or often true). Each mother’s responses to CBCL items were summed to calculate the child’s total score for problem behaviors. There are also subscales apart of the CBCL that can be utilized and they include: Somatic Complaints, Anxious/Depressed, Aggressive Behavior, Delinquent Behavior, Withdrawn, Social Problems, Attention Problems, and Thought Problems. Higher scores reflect higher levels of problem behaviors. The reliability of the CBCL questionnaire was strong, Cronbach’s α = .81.

Procedure

This study was a randomized clinical trial. Researchers recruited mother-child dyads for participants by Craigslist’s posts, fliers at local mental health clinics, and via a developmental database that is maintained by the psychology department. Once recruited, the participating families were assessed in a university campus office. Both the mother’s informed consent and the child’s assent were secured before participation, with IRB approval. There was a 2.5-hour session, where children completed assessments in one room while mothers completed questionnaires in an adjacent room. The questionnaires consisted of the 36-item DERS and the 64-item CCBL. The children’s assessment was Locked Box Task, which was a task aimed at trying to create frustration in children to measure behaviors such as distress, problem-solving, play activity, and distraction. Next, mothers and children reunited for parent-child interactions. Lastly, mothers received training in home cortisol data collection. They were debriefed about the research they were participating in. Participating families received $50 for the laboratory visit.

Results

A 2×2 factorial ANOVA determined that there was a main effect for the presence of maternal emotion dysregulation, however, there was no main effect for gender.  There was a difference in mean score for children who have mothers experiencing a clinical level of emotion dysregulation (M = 53.8, SD = 23.1) and for children who do not have mothers experiencing a clinical level of emotion dysregulation (M = 34.4, SD = 20.2), F (1, 61) = 12.2652, p > .001.  There was no significant difference in mean score for female children (M = 42.6, SD = 24.2) and for male children (M = 41.8, SD = 22.7), F (1, 61) = 0.031, p > .86. There was not a significant interaction between gender and having a mother who experiences a clinical level of emotion dysregulation, F (1, 61) = 0.104, p < .74. CBCL scores were not higher for female children with mothers who experience a clinical level of emotion dysregulation, compared to male children with mothers who experience a clinical level of emotion dysregulation (See Table 1 for cell means; see Figure 1 and Figure 2 for estimated marginal means). The range for the DERS scores was 40 to 125 (M = 68.4, SD = 21.7). The range for the CBCL scores was 5 to 124 (M = 42.2, SD = 23.3).

Discussion

Summary

It was hypothesized that there would be a main effect in which children with mothers who were dealing with a clinical level of emotion dysregulation would have a higher problem item score on the CBCL. Along with this, it was hypothesized a main effect would be present where males would have higher problem item scores on the Child Behavior Checklist than females. An interaction effect was hypothesized that female children with mothers who are dealing with a clinical level of emotion dysregulation would score higher on the CBCL compared to male children with mothers who are dealing with a clinical level of emotion dysregulation.

The results of our analysis provided insightful information surrounding the effects present between our variables. There was a statistically significant effect between children with mothers who were dealing with a clinical level of emotion dysregulation and higher problem item scores on the CBCL. However, there was no significant effect between children with mothers who were dealing with a clinical level of emotion dysregulation and problem item score on the CBCL. Along with this, our interaction effect, female children with mothers who are dealing with a clinical level of emotion dysregulation would score higher on the CBCL compared to male children with mothers who are dealing with a clinical level of emotion dysregulation, was non-significant. This means while there was visible evidence on a trend for the first hypothesis, no concrete evidence backing trends or interactions could be seen for the other two.

Our analysis provided partial support for this study’s hypotheses. The core idea of maternal emotion dysregulation affecting CBCL scores was supported, however, the intersection of gender was not. There was no evidence to prove the hypothesized effects that gender would have for the main effect on CBCL scores nor the interaction effect between a child’s mother dealing with a clinical level of emotional dysregulation and CBCL scores. There were no associations between gender, and child problem behaviors through these findings. Over our results supported our first main effect hypothesis, however, our second main effect hypothesis and interaction effect hypothesis did not hold to be true with our results.

Our results both agreed and disagreed with the previous findings in the literature review performed. Our first hypothesis, regarding the main effect that children with mothers who are dealing with a clinical level of emotion dysregulation will have higher problem item scores on the Child Behavior Checklist (CBCL) than children who do not have mothers dealing with a clinical level of emotion dysregulation, did compare to previous findings. The previous literature that utilized the CBCL, all found the same trend here at the surface level. However, when broken down into sections of emotion regulation, there was less consensus. While our study did not break down emotion dysregulation into subscales, many studies chose to separate qualities such as anxiety, aggression, distraction, and fear from each other. With the other studies, there was more focus on the aspect of emotion socialization within emotion regulation compared to ours. So, these are factors that may have affected the comparison of findings.

In terms of gender, our results contrast with the previous literature findings. The main literature piece on gender found that girls tended to regulate their emotions better and that mothers seemed to be more punitive towards their sons, leaving a significant difference in child labor. However, this study also took into account teacher opinions to try and avoid some bias from maternal figures. This bias may be important to why we did not receive statistically-significant outcomes for our secondary main effect and interaction effect. However, other studies that only performed secondary analyses on gender, did not find significant differences, which would align more with the findings received.

Strengths

This study possessed strengths surrounding the methodology and constructs used. Both measures have been frequently used tests in this area of child behavior and emotion regulation research, along with holding validity to accurately measure the variables. Along with this, the study was a randomized clinical trial, providing a more varied sample. By recruiting subgroups like low BPD, low-income individuals, there were advance steps taken to counteract predicted majorities. This also added some protection against researcher bias or participant bias, as the methodology was explicitly handled in steps and there was a low chance of manipulating any of the variables being researched. The way the hypotheses and variables are operationalized allowed for clear, direct research relating to valid constructs. It also allowed for a two by two factorial ANOVA to be completed, taking this research past the level of correlational. The study overall had rigid guidelines and objectives that allowed for transparency and generalizability.

Limitations and Possible Mechanisms

One significant limitation of this study is the lack of generalizability to a variety of demographics. The study contained very specific details that unfortunately limit how applicable the results are. One example of this is the children’s ages only ranging between three to four years old. While an important period developmentally, it may not be the key age range to target when trying to analyze differences due to gender. Children between the ages of three to four years old may not have a solid concept of gender identity, meaning that they may not be influenced even by a parent who was influenced by child gender. Internalized or externalized feelings about gender may not be present at this age because they do not understand their parents are treating them differently due to their gender. Along with this, three to four years old is long before puberty where gender differences become prominent, meaning that hormones and other variables may not impact behavior yet in our participants.

Along with this, the focus on only mothers limits the generalizability to general caregivers. By focusing on mothers, there is a stress placed on the mother-child relationship which is not often equally weighed to a caregiver-child relationship or a father-child relationship. It will be hard to generalize whether or not a father’s emotion dysregulation will have a similar impact compared to mothers. There is a lack of understanding of the father-daughter dyad or the father-son dyad. This may have partially contributed to why our interaction effect and our second main effect hypothesis were determined to be non-statically significant. The lack of ability to generalize to a larger audience, and collect a broader amount of information negatively impacts the ability to get the most accurate results. If experiment variables like age range were modified, it may have contributed to a significant result for gender effects.

A significant methodological limitation is not being able to directly assess child behaviors, and being limited in the questions asked. Child behaviors are assessed through the responses of participating mothers, this limits the child’s role in explaining and logically working through their behavior. By asking mothers to assess the child’s behavior, it is more so providing information on how mothers view certain behaviors and classify them. Not every mother may view aggression or irrational behaviors in the same manner. There is a strong possibility of bias in the reporting mothers, for them to exaggerate their good qualities while understating their downsides, and vice-versa for their child. Along with this, questions were limited because children were only the ages of three to four years old. Questions related to anxiety, behavioral disorders, or more complex-relationships would be non-applicable in this study thus, there is a lack of information there.

Our first hypothesis proved to be significant. However, there are possibly associated third-variables that could better explain the relationship analyzed. A very prominently discussed possible confound or moderator is the environment. There are many environmental influences in everyday life that can impact child behavior. This could include education environments such as child-teacher interactions, peer acceptance/social rejection, and the quality of the education in general. Other possible explanations could also be socio-economic status, where it is linked to several factors including food insecurity, healthcare, and homelessness. When children are forced into these situations without a way to pull themselves out, such as trying to get a job, an intense amount of frustration and even trauma, may build.

Those third variables also could be reshaped and analyzed similarly to gender, where they act as nominal independent variables, to view key differences between the groups, such as students with A-grades versus students with D-grades. The possibility of mediation could also be what explains why we saw a significant association between mothers’ scores for the CBCL and mothers’ scores for the DERS. One possible reason why the presence of a clinical level of maternal emotion dysregulation has a main effect with a child behavior problem items is because when mothers have a clinical level of emotion dysregulation, there is less validation from them regarding the child’s negative emotions. This leads to poor reactions from the mother that upset the child, increase frustration, and may leave many children feeling like they have to bottle-up negative emotions. This in return, leads to poor behaviors developed as children become angry for never feeling valid. Therefore, one possible mediator that could explain the relationship between a clinical level of emotion dysregulation in mothers and a child’s problem behaviors could be the lack of validation from mothers about a child’s negative emotions.

The lack of significant results, specifically for our gender-involved hypotheses, limits their applicability because it was not shown that gender effects problem behaviors. A large amount of theory backing up the impacts of maternal emotion dysregulation is because researchers make automatic assumptions that all children attach better to their mothers and that their mothers are inherently the emotional ones in a set of parents. Along with this, the most theory surrounding mothers’ emotions shaping child outcomes, only usually look at mothers in heterosexual relationships, who naturally produced the children, and are not immigrants or immersed in the culture in where they are living at. This type of theory limits the breadth of information that could be collected about minority groups, that could later be applied wide-scale. Issues that minority groups face, can be solved with benefitting non-minority groups still. Our results also limit the ability for us to back-up inherent gender differences that are theorized to be evident since infancy. While it was predicted that gender would cause a difference, it could easily be social rejection which tends to separate itself by gender once adolescence begins. Social rejection is theorized to be a defining characteristic as to why some people become bullies or develop extreme behavioral reactions; it has been applied to mass shooting perpetrators commonly as a way to explain why social rejection on youth can harvest bad behavioral issues long-term.

Implications

While not every result of the analysis was significant, the findings can still have implications for both research, and daily life. New contributions have been made in the field of child behavior development, especially in preschoolers. Our results were able to show that gender differences, focusing on those that affect behavior and emotions, may not develop as quickly as assumed. Children may not have as strong of genetic differences that predispose them to react in certain manners, reinforcing that either long-term maternal emotion dysregulation may have a stronger effect, along with puberty reinforcing gender differences. While our results contradict our literature review on the effects of gender and child behavior, many of these studies utilized teachers or peers to rate child behavior instead of a parent figure who could be biased. Technically, our results do contradict previous findings that males score worse on the behavior measures and that girls seem to be more emotionally impacted by their mothers. However, our base-level idea did hold. It was hypothesized there would be a main effect that children with mothers who are dealing with a clinical level of emotion dysregulation will have higher problem item scores on the Child Behavior Checklist (CBCL) than children who do not have mothers dealing with a clinical level of emotion dysregulation. This supported all of the previous findings, which indicated they also found higher behavior problem scores when the mother was experiencing higher emotion dysregulation. This has implications for future research because there is a new perspective on why gender may influence some groups compared to others. Parts of identity such as geographic location, culture, and family values may also influence the outcome gender has on emotion differences displayed in behaviors.

In daily life, our findings still have just as impactful implications. Behaviors and emotions are something that exists inside of everyone, whether children, adults, males, or females. Everyone is susceptible to bad behavior, and negative emotions. Understanding how to better regulate our emotions, and help children do too, is an important goal. It could help break down a lot of the child-parent interaction barriers that are experienced when there’s an emotional disconnect. By helping children develop more good behaviors than bad, there are long-term impacts such as a better chance at finishing high-school, less chance of delinquency, and fewer risk levels for abusing federally classified “hard substances”. This type of research could be applied at early learning centers, public education systems, juvenile detention centers, or even everyday situations where good behavior skills are necessary to excel.

Future Directions

The varied results of our study provide many directions to move towards this specific research question. Our first hypothesis, regarding the main effect in which children with mothers who were dealing with a clinical level of emotion dysregulation would have higher problem item scores on the CBCL, was accurate to the results. This opens up discussion for why this may be better explained outside of the parameters of emotion regulation. Emotion regulation encompasses so many different facets of daily behavior, moods, and personality, where it is so broad that treatments may need to be more specific. Other variables to look at when trying to explain changes in child behavior would be environmental factors or internal factors. For example, it was never specifically answered in emotional dysregulation plays the largest role in shaping a child’s behavior. By researching environment living situations for children, more inferences can be made about socioeconomic status, and type of area (rural vs. urban). Internal factors such as feeling invalidated may also greatly shape how children view themselves. This self-view or self-confidence could shape behavior depending on if the child had negative views of themselves, along with peer reactions.

While it would be incredibly difficult to have a true experimental reaction for a research question like this, especially due to the ethical implications of manipulating certain variables; this is even more prominent for minors. However, more study designs could be shaped to identify other demographic associations similar to gender for this association. Studies could be shaped to assess possible mediators such as validation, emotion-socialization, and social-rejection.

While our results surrounding the effects gender would have on child behavior, the research does not have to stop here. Older age groups such as high-school or college students could be utilized as a way to see if there is a difference surrounding the effects gender has, once puberty has occurred for the participants. This would be one key way to alter the methodology instead of changing the research focus. However, the other is possible by looking at other factors outside of gender that explain why we see develop differences between men and women’s emotion regulation in adulthood. One possible mechanism is different social stereotypes and expectations placed on males and females in puberty. Males and females are expected to develop different interests, have different conversations, and different ways of socializing with same-sex peers. Males are often expected to take on rugged, athletic, and sometimes violent interests as a way of showcasing their masculinity. This is not as prominent until children are old enough to be in these activities, which is why the difference may not show up in 3 and 4-year olds. This also means the differences aren’t rooted in gender but more so, gender-based social expectations.

References

Arikan, G., Kumru, A. Patterns of Associations Between Maternal Symptoms and Child Problem             Behaviors: The Mediating Role of Mentalization, Negative Intentionality, and       Unsupportive Emotion Socialization. Child Psychiatry Hum Dev (2020). https://doi.org/10.1007/s10578-020-01046-w

Achenbach, T., & Ruffle, T.M. (2000). The Child Behavior Checklist and related forms for assessing behavioral/emotional problems and competencies. Pediatrics in review, 21 8, 265-71.

Binion, G., & Zalewski, M. (2018). Maternal emotion dysregulation and the functional organization of preschoolers’ emotional expressions and regulatory behaviors. Emotion, 18(3), 386–399. https://doi-org.libproxy.uoregon.edu/10.1037/emo0000319

Gratz, K. L. & Roemer, L. (2004).  Multidimensional assessment of emotion regulation and dysregulation: Development, factor structure, and initial validation of the Difficulties in   Emotion Regulation Scale. Journal of Psychopathology and Behavioral Assessment, 26, 41-54.

Havighurst, S.S., Wilson, K.R., Harley, A.E., Prior, M.R. and Kehoe, C. (2010), Tuning in to Kids: improving emotion socialization practices in parents of preschool children – findings from a community trial. Journal of Child Psychology and Psychiatry, 51: 1342 1350. DOI:10.1111/j.1469-7610.2010.02303.x

Suh, B.L., Kang, M.J. Maternal Reactions to Preschoolers’ Negative Emotions and Aggression:   Gender Difference in Mediation of Emotion Regulation. J Child Fam Stud 29, 144–154     (2020). https://doi-org.libproxy.uoregon.edu/10.1007/s10826-019-01649-5

Zanarini, M. C., Vujanovic, A. A., Parachini, E. A., Boulanger, J. L., Frankenburg, F. R., & Hennen, J. (2003). A screening measure for BPD: The Mclean Screening Instrument for Borderline Personality Disorder (MSI-BPD). Journal of Personality Disorders. https://doi.org/10.1521/pedi.17.6.568.25355

Abstract

Body Dysmorphic Disorder (BDD) is a growing mental health disorder based around the relationship an individual has with their body, their perceived thoughts and images of their body, and the flaws they believe exist on them. Attachment security can play an impactful role in healthy relationships in adulthood, leading to the possibility that attachment security can affect the development of BDD symptoms. The Experiences in Close Relationships – Revised and the Body Dysmorphic Disorder Scale were both used to help measure our variables. It was hypothesized that there would be a positive, strong correlation between attachment-related anxiety and reported BDD symptoms. It was also hypothesized that there would be a positive, weak correlation between attachment-related avoidance and reported BDD symptoms. A total of 359 participants (Age: M= 27, SD= 9.6, Gender: Female (80.8%), Male (15.6%), Non-Binary (2.7%), Transgender Males (0.6%), and Transgender Females (0.3%)) voluntarily took part in a Qualtrics survey that contained our set of questionnaires amongst a series designed by student researchers. The results showed that ECR-R anxiety subscale scores correlated moderately, positively with BDDS scores while the ECR-R avoidance subscale scores correlated weakly, positively with BDDS scores. The implications of this survey are new information to help create more specific treatment programs for BDD, while also helping society better understand the difference between disordered eating contrasted with BDD. Nonetheless, the results show a promising future of research in the relationship between BDD symptoms and attachment insecurity.

The Link Between Attachment Insecurity and Reported Body Dysmorphic Disorder Symptoms

Introduction

The romanticized culture of the 2010’s surrounding being within the “normal” size range intensified the pressure for young men and women to fit in. With the creation of image-sharing apps such as Instagram and Snapchat paired with anonymous Q&A apps, individuals have been pressured to change their appearance due to social comparison and bullying (Matsakis, 2018). Additionally, Twitter and Tumblr have allowed for the easy development of pages, hashtags, and trends related to disordered eating (Tumblr Staff, 2012). These online communities have grown to accept similar body-related disorders, one of the more prevalent ones being Body Dysmorphic Disorder (BDD). While technically classified as an obsessive-compulsive disorder, many individuals have been misdiagnosed with having an eating disorder instead of BDD. Tags such as “#proana” and “thinspo” are often used to group posts together that vent internal emotions and struggles with body dysmorphia and disordered eating symptoms (Matsakis, 2018). There have also been so many affected individuals who mask their disorder, unable to be helped by the current practices and methods.

The lack of information surrounding BDD until the 2010s made it extremely difficult to shape treatments around this mental health disorder. BDD doesn’t have the same symptoms compared to those with an eating disorder that would be noticed by other people, such as dramatic changes in weight, changes in eating habits, weakened teeth, hair loss, and fear around food. For many, it is unrelated to weight and involves other flaws on the body leading to indictors more like lesions, severe depression, and possible forms of self-harm (Phillips, 2019). Many of these individuals will have been raised in dysfunctional families or possibly abusive relationships leading to insecure attachment development (Didie et al., 2006; Phillips, Menard, Fay, & Weisberg, 2005). Other forms of trauma can also emerge during childhood and adolescence that can significantly impact secure attachment development. While insecure attachment is a common factor in both BDD and eating disorders, BDD presents its own unique set of issues. The ability for almost any feature on an individual with BDD to become viewed as a defect leads to variability and complexity in treatment methods. When we take this into account with the underlying insecure attachments, there has been a lack of applicable methods in body dysmorphic disorder diagnosis and treatment to help those with insecure attachments adapt to a healthier outlook. Individuals with insecure attachments often devalue themselves in terms of self-concept and, for some, social-comparisons; this only builds on to the devaluing that takes place over flaws.

Literature Review

The intersection of negative body image and insecure attachment is not a newly discovered field, and the emergence of empirical research in the area has been evident since the 2000s. All of the relevant literature reviewed for this, in insecure attachment and negative body image, has found consensus in the fact that insecure attachment does seem to help predict psychopathology related to disordered eating and body dysmorphia. However, the methods used for this research does not directly test on BDD measures. So, when researchers are concluding a connection between insecure attachment and psychopathology-related body dysmorphia, it is an educated inference based on data tested with disordered eating scales such as the Eating Disorder Inventory (EDI; Garner, Olmstead, & Polivy, 1983) or body satisfaction scales. While this still supports the proposed correlation, it does leave gaps in the literature for more direct research on the connection between insecure attachment and BDD symptoms. Along with this, no clear mediators or moderators have been defined for the correlation between insecure attachment and body dysmorphic disorder.

The literature reviewed for this study has focused on Caucasian females in young adulthood, in the Western world (clinical and non-clinical); these demographics may show similar sociocultural values, and form patterns in developing, and thinking about body dysmorphia and negative body-image. Due to the amount of research showing the connection between insecure attachment and disordered eating, much of the relevant literature has chosen to focus on identifying moderators or mediators as a way to help explain how insecure attachment effects body-image and body-satisfaction. This has helped show the differences between how avoidant and anxious insecure-attachments. The relevant literature has shown that mediators often focused on interpersonal relationships and social comparison tend to only be associated with anxious-attachments because anxious-attached individuals tend to compare and devalue themselves amongst others while also seeking validation (Nandrino, Dodin, Cottencin, Doba, 2020). People with avoidant-attachments tend to not engage in validation-seeking behavior or inter-personal relationships, due to deactivating normal attachment reactions (Bamford & Halliwell, 2009). However, even with those distinct differences, there is evidence that self-concept and maladaptive perfectionism can be mediators between both types of insecure attachment styles and eating disorder psychopathology; this may be due to both insecure attachments, in general, involving extreme criticism of one’s self (Dakanalis et al., 2014).

Each literature article reviewed shows strong, positive correlations between insecure attachments and the presence of negative body image issues. They find people with anxious-insecure attachments could experience mediators more related to social-comparison and the relationships around them due to the consistent need for validation from peers (Bamford & Halliwell, 2009). Without this validation, the development of negative body image and possibly body dysmorphia symptoms can occur; anxiously-attached individuals may put more emphasis on popular figures such as models compared to avoidant individuals. Avoidant-attachments do not utilize more social or relationship-based mediators because much of their negative body image comes from a cycle of cynicism and devaluing (Dakanalis et al., 2014). There is a strong presence of empirical research that shows while there are base-level similarities, there are core differences between eating disorders and body dysmorphic disorder thinking patterns for people with these disorders diagnosed (Jefferies-Sewell, Chamberlain, Fineberg, & Laws, 2017). Even further research has been done to show the visual cognitive processing differences for individuals with diagnosed eating disorders and individuals with diagnosed BDD (Madsen, Bohon, & Feusner, 2013). For that reason, some inferences about shared moderators, target groups, and treatment methods can be made to help shape an idea of an improved support system for individuals with BDD. However, a downside of the disorders’ similarities is that body dysmorphic disorder has been looked over by researchers and media sources leading to a lack of direct research and adaptable treatment methods.

An important supporting aspect of the literature is the continued usage of John Bowlby’s and Mary Ainsworth’s attachment theories. Both of these theories have helped create the attachment styles of insecure-anxious, insecure-avoidant, secure, and later, disorganized. Bowlby and Ainsworth were the first key researchers who were able to categorize attachment styles based on observational cues and relationships with parents and caregivers (Bretherton, 1992). By understanding how secure and insecure attachments operate differently from each other helps understand how the cognitive, thought, and reward processes in the brain work. Those with insecure attachments may struggle with rewarding themselves or recognizing depressive cycles. Bowlby’s attachment styles have continued to have a large impact on understanding how we form relationships with trust, and how we grow to see and interact with others. Our research should be supported by and also help support Bowlby’s theory surrounding the long-term mental impacts of those with insecure attachments (anxious and avoidant).

Purpose of the Current Study

Previous literature suggests that there are continued gaps between understanding what role, if any, insecure attachment has on body dysmorphic disorder symptoms. This lack of direct measures and data can prevent a well-rounded view of body dysmorphia-related psychopathology and the long-term impacts of attachment style. As a more newly classified mental health disorder, literature has continued to suggest that there are unknown mediators or moderators between how insecure attachments lead to body dysmorphia, such as internal trauma or self-concept issues. Along with this, potential comorbidity with other illnesses can leave individuals with BDD still struggling even once in a standard treatment program, therapy, or medication. There are high rates of individuals with body dysmorphic disorder who also are diagnosed with other mental health disorders such as depression, anxiety, substance abuse, and mood-regulation disorders such as bipolar disorder (Gunstad, 2003). Individuals with body dysmorphic disorder may find it easier to overcome compulsive urges and thoughts about flaws if treatment and therapy programs were able to accurately identify underlying causes for the development of the disorder. Attachment continues to shape aspects of self-concept and interpersonal-relationships long after adolescence, and may play a much larger role in body-image than originally assumed (Ty & Francis, 2013).

Most current literature has focused on females in a clinical setting, with mediators and moderators geared towards inter-personal relationships. All of the studies have focused on eating disorder-specific symptoms instead of body dysmorphia related symptomology or general body image issues. Relevant literature often relies on broader terminology such as body-image, body satisfaction, and self-esteem to make connections between BDD and other related disorders. As much of the relevant literature has said, many people experience an insecure attachment, however, very few will develop an eating disorder (Bamford & Halliwell, 2009).

The current study is going to give a more focused look at individuals in a non-clinical setting, and their role attachment insecurity has on the presence and the number of body dysmorphic disorder symptoms present. Often being lumped in with others eating disorders, body dysmorphic disorder presents a complex set of challenges often involving overcoming mental difficulties to confront the negatively perceived flaws that exist to them. It is a part of the Obsessive-Compulsive Disorders in the DSM-V and has its own set of criteria apart of other eating disorders due to emphasis on the constant obsession over perceived flaws. These symptoms are often in connection with concepts like body image, self-concept, and peer pressure. Our research needs to separate anxiously-attached individuals from avoidantly-attached individuals because they do not share all the same qualities or possible mediators. While previous literature has shown inter-personal relationships can play a role in the severity of body image issues, it is more directed at the constant validation from peers that would help anxious-attached individuals. Further research would have to be done to see how much inter-personal relationships and peer validation plays a role in BDD symptoms because perceived flaws always exist to the individual.

Variables and Hypothesis

Two different tests were used to help create continuous variables for insecure attachment and BDD symptoms. The Experiences in Close Relationships – Revised (ECR-R; Fraley, Waller, & Brennan, 2000) was utilized to help gauge attachment-related anxiety or avoidance levels. The higher someone scores (or lower for reverse-keyed questions) means they experience stronger attachment-related anxiety or avoidance; the test is scored in two different sections to separate insecure-anxious and insecure-avoidant traits. The Body Dysmorphic Disorder Scale (BDDS; Buhlmann et al., 2010) was reformatted and utilized to help gauge body dysmorphic disorder symptoms. The higher someone scores (or lower for reverse-keyed questions) means they experience more BDD related symptoms.

I hypothesize that there will be a positive, strong correlation between attachment-related anxiety and reported body dysmorphic symptoms. I also hypothesized that there will be a positive, weak correlation between attachment-related avoidance and reported body dysmorphic symptoms.

Methods

Participants

The online survey was circulated by the researchers via email and social media. A total of 359 people participated in the online study. The ethnic/racial backgrounds the participants identified were: 77.9% White/Caucasian, 3.88% Multi-Ethnic, 1.18% Black/African-American, 7.08% Hispanic/Latino, 6.57% Asian/Asian-American, .88% Native American, .298% Hawaiian/Pacific Islander, .9% Middle Eastern, 1.18% Other). The majority of the sample identified as Female (80.8%), with other identities including Male (15.6%), Gender-Queer / Non-Binary (2.7%), Transgender Males (0.6%), and Transgender Females (0.3%).  Participants ranged in age from 18 to 82 years (M = 27, SD = 9.60). None of the participants were compensated for their participation.

Materials

This study consisted of two self-report questionnaires, one measuring attachment insecurity and the other measuring body dysmorphic disorder symptoms. Insecure attachment was evaluated using the Experiences in Close Relationships – Revised (ECR-R; Fraley et al., 2000), a 36-item measure that assesses insecure adult attachment, focusing on how avoidant or anxious an adult is. Example items include: “I am nervous when partners get too close to me” and “My desire to be very close sometimes scares people away”. The participants used a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree) to rate how individuals feel in emotionally intimate relationships. The first 18-items refer to the attachment-related anxiety scale whereas items 19-36 refer to the attachment-related avoidance scale. Participants can receive a score between 1 and 7. Scores closer to 1 designate higher security, and score towards 7 designate higher insecurity. The scores are averaged for each individual for both the attachment-related anxiety items and attachment-related avoidance items. While only items 9 and 11 are reverse-keyed for the anxiety scale, items 0, 22, 26, 27, 28, 29, 30, 31, 33, 34, 35, and 36 are reverse-keyed for the avoidance scale.

To measure self-reported body dysmorphic disorder symptoms, the Body Dysmorphic Disorder Scale was adapted (BDDS; Buhlmann et al., 2010). While originally using a yes/no format, researchers reformatted it to use a 7-point Likert scale ranging from strongly agree to strongly disagree to help individuals’ rate how many symptoms they may have related to body dysmorphic disorder. Example items include: “Do you think about your appearance concerns for at least one hour a day?” and “Are you primarily concerned about not being thin enough or becoming too fat?”. To rate the answers, strongly agree would equal 1 and strongly disagree would equal 7. The score range is from 8 to 56; with higher scores indicating a higher number of body dysmorphic symptoms affecting the individual.

These two questionnaires and demographic questions were included in a compilation of questionnaires created by the class researchers and then administered via one survey on the Qualtrics website.

Procedures

Researchers recruited participants by asking them to assist with a project for a psychology course by taking an online survey. This survey contained multiple questionaries’ that the class researchers designed. The link for the online survey in Qualtrics was e-mailed to participants or posted onto social media platforms. Participants opened or clicked on the link on their personal computers or mobile devices. This leads them directly to the survey where the first page asked participants to give informed consent. Informed consent is a feature directly on the Qualtrics website. After giving informed consent, participants would be led to the questionnaires. For our particular study, participants answered 36 items regarding insecure attachments via the Experiences in Close Relationships – Revised, 8 items regarding body dysmorphia via the Body Dysmorphic Disorder Scale, and 4 items on demographics. At the completion of the questionnaire, participants were debriefed about the research goals of the study. Along with this, resources were provided if participants felt distressed during the survey. Contact information was provided in case participants had any inquires related to any part of the survey.

Results

            The current study hypothesized that Experiences in Close Relationships – Revised scores would positively correlate with Body Dysmorphic Disorder Scale scores. More specifically, attachment-related anxiety scores from the ECR-R would positively, strongly correlate with BDDS scores while attachment-related avoidance scores from the ECR-R would positively, weakly correlate with BDDS scores. The finals results confirmed my hypothesis related to attachment-related avoidance scores, with statistically significant evidence. However, my hypothesis was incorrect for the attachment-related anxiety scores. Those scores ended up correlating only moderately, instead of strongly, with BDDS scores.

The ECR-R score subscales (anxiety and avoidance) were averaged separately to output an attachment-avoidance score and an attachment-anxiety score for each participant. The mean anxiety-related ECR-R score was 3.63 (SD= 1.27) with a range of 5.56. The reliability, Cronbach’s alpha, for the anxiety-related subscale items was high, α = .93. The mean avoidance-related ECR-R score was 2.87 (SD=1.29) with a range of 6.00. The reliability, Cronbach’s alpha, for the avoidance-related subscale items was high, α = 0.96.

The Body Dysmorphic Disorder Scale scores were determined by summing the scores together of each item for each participant. The mean BDDS score was 11.6 (SD= 13.5), with a range of 49. No reliability analysis, Cronbach’s alpha, was performed for this variable as the items on the questionnaire were not designed for direct consistency in answers; it was expected there would be variation in participants’ answers.

To study the associations between attachment-related anxiety scores, attachment-related avoidance scores, and body dysmorphic disorder symptom scores, a bivariate Pearson’s correlation was conducted. The bivariate Pearson correlation revealed a statistically significant, moderate, positive correlation between the ECR-R anxiety subscale scores and BDDS scores, r(268) = +0.33, p < .001. This indicates that when there is an increase in anxiety-related ECR-R scores, there is most likely an increase in body dysmorphic disorder scale scores. While originally hypothesizing that the strength of this subset of data to be strong, it was moderately correlated and still supported the general hypothesis that there would be a positive correlation present.

The bivariate Pearson correlation revealed a statistically significant, weak, positive correlation between the ECR-R avoidance subscale scores and BDDS scores, r(269) = +0.13, p = .03. This indicates that when there is an increase in avoidance-related ECR-R scores, there is most likely an increase in body dysmorphic disorder scale scores. This is consistent with our original hypothesis, to see a weak, positive increase in BDDS scores correlated with increased ECR-R avoidance subscale scores.

Discussion

Summary

It was hypothesized that there would be a positive, strong correlation between attachment-related anxiety and reported body dysmorphic symptoms. It was also hypothesized that there would be a positive, weak correlation between attachment-related avoidance and reported body dysmorphic symptoms. While the results supported the latter half of the hypothesis, it was not completely accurate for the first, anxiety-related, part. ECR-R avoidance subscales scores did show a weak, positive correlation to BDDS scores, like hypothesized. However, ECR-R anxiety subscale scores did not show a strong correlation, only a moderate one. The results showed that ECR-R anxiety subscale scores had a moderate, positive correlation with BDDS scores.

These results relay the finding that as attachment-related anxiety or attachment-related avoidance increased, so did having BDD related symptoms. This means that attachment security may have a larger role in the development of BDD symptoms and overall body image, however that cannot be concluded as an experimental study did not take place and appropriate constructs were not used to try and identify cause and effects. Rather, these results would directly support my original idea that attachment security (often developed in childhood) would correlate with the development of BDD symptoms (often in young adulthood). Statistically, my results aligned fairly accurately with my operationalized hypothesis, showing the weak, positive correlation between BDDS scores and ECR-R avoidance scores. While my hypothesis about the strength of the correlation between BDDS scores and ECR-R anxiety scores predicted strong, the results ended up in the moderate region of strength.

Compared to the previous literature reviewed, my results are closely aligned with their findings. An important note to make is that the literature review only encompassed studies that did not include measures or constructs specifically based on Body Dysmorphic Disorder; instead, these studies used disordered eating and body-image measures to help find patterns that could also correlate with BDD symptoms.  All of my studies showed a positive association between anxious-insecure attachment indicators and body image issues. Our results lend more support to previous findings while expanding the depth of the combined research to encompass Body Dysmorphic Disorder more. Melissa Ty and Andrew Francis’s research in emotional dysregulation as a mediator in the association between insecure attachment and disordered eating habits was partially supported by our results. While they focused strongly on their second hypothesis, social comparison, and emotional dysregulation are mediators, our results supported their first hypothesis that there is a positive correlation between insecure attachment and body image issues. Analysis of the results to certain questions also could support their findings that social comparison was a mediator for anxiously-attached individuals but not avoidantly-attached individuals (Ty & Francis, 2013). This also relates and supports the Bamford and Halliwell article which also explored social comparison as a mediator for anxiously-attached individuals, and not for avoidantly-attached individuals. They also found no significant connection for avoidantly-attached individuals and social comparison as a mediator. (Bamford & Halliwell, 2009)

Regarding the article about interpersonal emotional competences, our results would support their findings that anxiously or avoidantly-attached individuals with body image issues tend to face more interpersonal struggles as their body image issues become worse. This is because body image issues can push people away from wanting to socialize due to their perceived flaws (Nandrino et al, 2020). Our results greatly aligned with the journal article researching self-concept as a mediator in the relationship between attachment insecurity and identity of women with body image issues. They found that women had a harder time developing identity differentiation, which can make it hard to improve body satisfaction. Both anxiously-attached and avoidantly-attached individuals had negative effects when associated with lower self-concept (Demidenko et al., 2010). This helps builds onto the possibility of self-concept being a mediator for our current study. Partially tied into self-concept, the article on maladaptive perfectionism was partially supported. While our research did not specifically look for maladaptive perfectionism, a sign of BDD is trying to perfect perceived flaws. There can be broader connections made that maladaptive perfectionism can be a mediator between attachment insecurity and self-reported BDD symptoms, similar to eating disorder psychopathology like previously researched (Dakanalis et al., 2014).

Strengths

            The study conducted had several strengths outside of the results. Several methodological steps were taken to make sure we had accurate measures and results. Pseudo-random sampling was used as the most random sampling method possible for the decided route of administering the survey. Since researchers recruited participants by email or social media, true random sampling could not be guaranteed. However, pseudo-random sampling still provides most of the benefits and protection needed to eliminate the chance of bias or errors in data collection. By pseudo-random sampling, there were still some barriers in place to protect the results from possible tampering while also allowing the researchers to work with the data in the same methods as randomly sampled data.

The dependent variable was also assessed against two independent variables. Self-reported BDD symptoms were assessed using a bivariate correlation that compared it separately to attachment-related anxiety, and attachment-related avoidance. Expected results did come out of our operationalized variables and other measures, which also helps support the strength of our dependent variable (self-reported BDD symptoms) having good construct validity.

Lastly, while a larger sample size could have brought more diversity in background information such as race/ethnicity, age, and gender, a sample size of 359 was adequate to analyze for having the survey open for only the course of a few days. 359 participants allowed us to see significant correlations between attachment security and self-reported BDD symptoms. It also allowed us to verify that our ECR-R section of the survey measured the proper construct once the Cronbach’s alpha was completed.

Limitations

Sample Limitations. While our study provided significant results, there were still limitations preventing us from receiving a more well-rounded view of the correlation seen. Due to only having 359 participants, our majority were mostly women in their twenties. This is a common demographic to study, especially concerning body image because they often seek help the soonest; this leads to most clinical studies being mostly comprised of women. As seen in our demographic statistics, there were very few participants who identified as Non-Binary/Gender-Queer, Transgender Male, or Transgender Female. Along with this, only 15.6% of participants identified as male.  These small numerical figures led to non-significant correlations between each gender category, and our variables. This limits the generalizability of the study because our results may not be as applicable to males, Non-Binary/Gender-Queer, or Transgender individuals. These individuals may experience strong, positive correlations between BBD symptoms and attachment insecurity however, more information about those demographics would be needed.

Along with this, most of the sample identified as White or Caucasian (77.9%). This leads to gaps in information that would help us determine if racial or ethnic background led to influenced attachment security or the development of BDD symptoms. Many black, indigenous, and other people of color face elevated rates of mental health disorders due to the lack of adequate healthcare and mental health services for those communities, along with racial biases often making it harder for those communities to receive the diagnosis and treatment needed. The lack of data on racial and ethnic backgrounds also limits what conclusions can be made about intersectionality and the development of BDD symptoms.

Methodological Limitations. The most limiting factor of our study is that it cannot make or support any causal claims. The studies were constructed to be correlational and use variables designed to be continuous. These factors do limit the ability to make any conclusions on cause or effect, as the study was not experimental. As it was a survey, participants were not observed and there were not strong protectors in place to prevent participants from marking non-truthful answers on the individual questionnaires.

The questions asked in the survey also had to be narrowed down from many of the original tests because, researchers did not want the participants to get tired during the survey, possibly affecting the answers for some. Questions also had to be within certain guidelines to prevent participants from answering questions with information that would require a follow-up. For these reasons, graphic questions related to BDD, suicide, or personal safety were not asked. While necessary, these guidelines do technically limit some questions that could be asked about either of our variables, reducing the amount of direct information or key indicators about BDD and attachment insecurity; for example, self-harm habits which could be a possible indicator for the presence either variable (attachment insecurity or BDD symptoms).

The last limitation of the methodology is that there was not true random sampling. We could not verify that everything was randomly sampled because the researchers recruited participants by posting the survey online, and emailing participants asking them to take part in the survey. While no identifying information was requested, and we tried to control any ability for the researchers to identify any participants through demographic information, there is still no way to fully guarantee a perfectly random sample. For this reason, it is defined as a pseudo-random sample. While this is a methodological limitation, it should not be a highly prominent issue as we tried to measure the constructs, and administer the survey most anonymously and randomly possible.

Third Variable Limitations. The results of this study aligned very well with the other literature, leading to an educated inference that other possible mechanisms such as third variables may be involved in the association between BDD symptoms and attachment insecurity. Most of the mediators used in previous literature focused on aspects of the development of disordered eating, including inter-personal relationships and forms of peer pressure. However, those mediators would not be as applicable to this research due to avoidantly-attached individuals being very detached from social relationships and do not feel dependent on other’s opinions as much as anxiously-attached individuals. However, a possible mediator for the association between reported BDD symptoms and attachment insecurity is body satisfaction. One possible reason why attachment insecurity may be positively correlated with self-reported BDD symptoms is that when attachment insecurity is high, individuals may develop complex relationships based on anxiety or avoidance, including with their own body for some. The previous experiences of insecurely-attached individuals may have helped create the body satisfaction issues some of these individuals may experience. The poor experiences or mistreatment from family and partners that create attachment insecurity could be based on issues with body appearance; many families, peers, and partners are not “body-positive” or accepting of all-sizes. This in return, could lead to the development of Body Dysmorphic Disorder symptoms because, when body satisfaction is low, the perceived flaws that individuals with BDD experience may arise. This is because generally, those with low body satisfaction often want to change their appearance to become more satisfied, however, the change could be minor or major. BDD focuses on individuals with perceived flaws who will never be satisfied with their bodies (without some form of treatment or support) and possibly actively trying to change those flaws.

Other possible mediators could operate in almost an identical way outside of the constructs that would need to be used. This includes self-esteem which can be hard to develop healthily when in insecure relationships especially in youth with parents, decreased self-esteem can also lead to perceived flaws. This is also very similar for possible mediators like body image, self-concept, and confidence.

Lastly, there is the possibility that comorbidity of other mental health disorders acts as a mediator, where it helps create the correlation between attachment insecurity and BDD symptoms. Mental health disorders could also act more as a confound where it pushes the other two variables to be correlated. For example, insecurely-attached individuals may experience depression, that pushes forward the presence of self-reported BDD symptoms. Disordered eating could also work in this manner for those that experience it before developing their self-reported BDD symptoms.

These mediators and confounds do limit but, also help expand on the research conducted here. While they do lead to the possibility that our constructs could be improved, they also provide more areas for research. It is important to keep in mind that while these third variables could alternatively explain why we see the correlation between attachment insecurity and self-reported BDD symptoms, they do not invalidate or disprove the original correlation that has been found.

Implications

The findings of our study add new contributions to the information surrounding the presence of BDD symptoms, and possible associations it may have. As none of the reviewed literature directly focused on the connection between BBD symptoms and attachment security, especially using constructs designed for BDD, this study can add a lot of new data. Along with this, it also provided more information on the effectiveness of the ECR-R to measure attachment-related anxiety/avoidance, and of the BDDS to measure self-reported symptoms. Lastly, by viewing the questions on the scale, and comparing what answers were given to each, valuable new information can be found. Such as, the answers from individuals who had high avoidant scores, tended to score lower for self-reported BDDS symptoms. For example, individuals who responded to question 19 on the ECR-R, “I prefer not to show a partner how I feel deep down” with a high number indicating they were more avoidant, then they also tended to score low on question 4 of the BDDS, “Do the worries about your physical defect cause significant impairment in your everyday life (e.g. in you job or social life)?”. For individuals who marked low scores to questions 19 of the ECR-R, indicating they were more anxious, they also tended to score quite high on question 4 of the BDDS. This helps show how avoidantly-attached individuals disconnect themselves from social relationships and build onto how they develop BDD symptoms without inter-personal relationships or peer pressure. This helps contribute new information on more specific characteristics or situations that people with attachment-related anxiety or avoidance experience.

As discussed above, my results supported the previous findings, especially for the literature based on social comparison. My results directly matched their findings of anxious-attached insecurity positively correlating with body dissatisfaction or body image issues. This creates implications for new targeted treatment processes and opens up the field for researching mediators of BDD development. While a lot of their work focused on eating disorders, their broader findings, findings, and mediators could be indirectly applicable to BDD. There could be expanded treatment programs aimed at addressing attachment-security along with possible mediators such as maladaptive perfectionism, social comparison, self-concept, and body dissatisfaction. Along with this, expanded research about possible comorbidity with other mental health disorders that may encompass some of those mediators listed above, or even unidentified third variables yet to be found.

Our study’s findings have many implications for fields outside of psychology. Body dissatisfaction is a growing issue faced by many people, of all ages and backgrounds. Body Dysmorphic Disorder can often develop in youth to young-adulthood and often goes unrecognized or misdiagnosed. Many people are misdiagnosed as having only an eating disorder or are referred to as ineffective treatments for their BDD. Our findings help expand the information supporting an increased presence of other factors correlating with the development of BDD. In general, the research about BDD, a very specific mental health disorder, can more broadly be applied to those facing difficulties with perceived flaws, and the struggle to correct them, even if these individuals are not at a severity to be diagnosed with BDD.

Helping individuals overcome perceived flaws on their bodies can help build confidence, self-concept, and body satisfaction. This in return can boost mood, positive thinking, and healthy relationships. While many of these individuals may struggle with attachment insecurity, there are possibilities to gain trust and attachment skills through body acceptance. This is especially important for individuals who have emotional trauma related to their bodies, this information could help them understand better ways to seek help and cope.

Along with this, our research could apply to physicians, nutritionists, and personal trainers, who may all come across needing to work with individuals who show symptoms of BDD. BDD is rooted in perceived flaws, and a constant unacceptance of the body; it does not stem in the same places an eating disorder would yet individuals with BDD may find themselves sharing the same habits as restrictive food-intake, and trying to harm themselves to correct perceived appearance flaws.

Future Directions

Now that some core research has been conducted on the correlation between attachment insecurity and self-reported BDD symptoms, there are a lot of new directions for future research. While our study focused on attachment insecurity as a correlating factor with self-reported BDD symptoms, there is the possibility to construct a clinical experiment to obtain more longitudinal, casual data. Our constructs, measures, variables were not developed to be used on an experiment, however, that pathway would lead to more direct cause and effect data. There could also be more treatment-focused clinical experiments on the types of effective BDD treatment, and the possible incorporation of treatment methods used to address attachment insecurity.

Another possible direction includes using new measures. Our questionnaire had to be brief, yet encompassing because it was being compiled into a much larger survey to be spread to participants. If the survey was only for our study, there could have been longer tests utilized or even tests included that addressed possible mediators. One option instead of the ECR-R that is often used to address adult attachment is the Adult Attachment Interview (AAI) which uses an interview-style format while asking questions based on childhood recollection to determine attachment. There are also many alternative tests to measure BDD symptoms, such as the Body Dysmorphic Disorder Symptoms Measure or the Body Dysmorphic Disorder Symptoms Scale, which is a much more encompassing survey at 52 questions. There is also the Body Image Disturbance Questionnaire and the BDD subscales of the Eating Disorder Inventories.

To help expand on the current research, a very beneficial direction would be to look into potential third variables (confounds, moderators, mediators, etc.) that may help better explain the relationship between attachment insecurity and self-reported BDD symptoms. Potential variables include self-concept, body satisfaction, and maladaptive perfectionism. The Body-Image Acceptance Scale and Body Satisfaction Scale are both tests that are commonly used to assess body image and satisfaction which are potential mediators to explain why attachment insecurity is connected with BDD symptoms.

There is also the ability to focus more specifically either on anxiously-attached or avoidantly-attached individuals and conduct possible comparisons to other disorders. There was a weak correlation between avoidantly-attached individuals and self-reported symptoms which could lead to the possible research question, do we find that avoidant-attached people experience body dysmorphia more than disordered eating habits do because it relies more on self-concept rather than direct validation from peers?

With this research topic, there are a variety of directions to go to expand on the information already collected. By identifying more aspects of the relationship that creates the positive correlation viewed in our study, there will be more able to adapt our results to applicable policy and treatment methods.

References

Bamford, B., & Halliwell, E. (2009). Investigating the role of attachment in social comparison theories of eating disorders within a non-clinical female population. European Eating Disorders Review, 17(5), 371–379. https://doi.org/10.1002/erv.951

Bretherton, I. (1992). The origins of attachment theory: John Bowlby and Mary Ainsworth. Developmental psychology, 28(5), 759.

Buhlmann, U., Glaesmer, H., Mewes, R., Fama, J. M., Wilhelm, S., Brähler, E., & Rief, W. (2010). Body Dysmorphic Disorder Scale [Database record]. Retrieved from PsycTESTS. doi: https://dx.doi.org/10.1037/t29823-000

Dakanalis, A., Timko, C. A., Zanetti, M. A., Rinaldi, L., Prunas, A., Carrà, G., … Clerici, M.       (2014). Attachment insecurities, maladaptive perfectionism, and eating disorder symptoms: A latent mediated and moderated structural equation modeling analysis across diagnostic groups. Psychiatry Research, 215(1), 176–184.         https://doi.org/10.1016/j.psychres.2013.10.039

Demidenko, N., Tasca, G. A., Kennedy, N., & Bissada, H. (2010). The mediating role of self-concept in the relationship between attachment insecurity and identity differentiation among women with an eating disorder. Journal of Social and Clinical Psychology, 29(10), 1131–1152. https://doi.org/10.1521/jscp.2010.29.10.1131

Didie, E. R., Tortolani, C. C., Pope, C. G., Menard, W., Fay, C., & Phillips, K. A. (2006). Childhood abuse and neglect in body dysmorphic disorder. Child Abuse and Neglect, 30(10), 1105–1115. https://doi.org/10.1016/j.chiabu.2006.03.007

Fraley, R. C., Waller, N. G., & Brennan, K. A. (2000). An item-response theory analysis of self-report measures of adult attachment. Journal of Personality and Social Psychology, 78,    350-365.

Garner, D. M., Olmstead, M. P., & Polivy, J. (1983). Development and validation of a multidimensional eating disorder inventory for anorexia nervosa and bulimia. International Journal of Eating Disorders, 2(2), 15–34. https://doi-org.libproxy.uoregon.edu/10.1002/1098-108X(198321)2:2<15::AID-EAT2260020203>3.0.CO;2-6

Gunstad, J., & Phillips, K. A. (2003). Axis I comorbidity in body dysmorphic disorder. Comprehensive Psychiatry, 44(4), 270–276. https://doi.org/10.1016/S0010-  440X(03)00088-9

Jefferies-Sewell, K., Chamberlain, S. R., Fineberg, N. A., & Laws, K. R. (2017). Cognitive dysfunction in body dysmorphic disorder: new implications for nosological systems and neurobiological models. CNS spectrums, 22(1), 51–60. https://doi.org/10.1017/S1092852916000468

Matsakis, L. (2018, June 13). How Pro-Eating Disorder Posts Evade Filters on Social Media. Retrieved August 03, 2020, from https://www.wired.com/story/how-pro-eating-disorderposts-evade-social-media-filters/

Nandrino, J. L., Dodin, V., Cottencin, O., & Doba, K. (2020). Effect of intrapersonal emotional competencies on the relationship between attachment insecurity and severity of eating disorder symptoms in patients with restrictive anorexia. Journal of Clinical Psychology, 76(3), 476–492. https://doi.org/10.1002/jclp.22877

Phillips, K. A. (2019, January 10). Signs & Symptoms of BDD. Retrieved August 03, 2020, from  https://bdd.iocdf.org/professionals/signs-symptoms/

Phillips, K. A., Menard, W., Fay, C., & Weisberg, R. (2005). Demographic characteristics, phenomenology, comorbidity, and family history in 2001 individuals with body dysmorphic disorder. Psychosomatics, 46(4), 317–325. https://doi.org/10.1176/appi.psy.46.4.317

Tumblr Staff. (2012, February 23). A New Policy Against Self-Harm Blogs. Retrieved August 03, 2020, from https://staff.tumblr.com/post/18132624829/self-harm-blogs

Ty, M., & Francis, A. J. P. (2013). Insecure Attachment and Disordered Eating in Women: The Mediating Processes of Social Comparison and Emotion Dysregulation. Eating Disorders, 21(2), 154–174. https://doi.org/10.1080/10640266.2013.761089