The relationship between early maladaptive schemas and childhood adversity
When I first became interested in schema therapy, I was initially in awe of the volume of theorised relationships between types of specific early life adversity and predictable patterns of later-life thinking, emoting and behaving. I felt embarrassed that I had ignored the books and workshops containing the holy grail linking developmental history to presenting problems and mental state for so long. Then the killjoy-psychologist in me kicked in and started wondering “where is the evidence for all of these relationships?” There seemed to be more hypotheses than references in the schema therapy manuals. So, I was genuinely excited when Chris Hayes and Lisa Chantler both independently put me on to this week’s article: a meta-analysis of the existing literature on the relationship between adverse childhood events and early maladaptive schemas in adults, by Australian psychologist, Pamela Pilkington and colleagues.
Schema therapy theoretical basics
Schema therapy assumes that humans have fundamental physical and psychological needs, including safety, stability, nurturance, acceptance, autonomy, competence, a sense of identity, freedom to express needs and emotions, spontaneity and play, and realistic limits. It is thought that childhood environments that fail to meet these needs stunt or pervert the individual’s psychological development. For example, a child growing up with a physically abusive father without protection from their violent outbursts has not had their need for safety met; a child who grows up being allowed to do whatever they want whenever they want has not had their need for realistic limits met. These early childhood environments teach the child what to expect: from the world, others and themselves. Depending on the nature of childhood adversity, the child may learn to expect that they will never get their needs met, they will always get their needs met, or that they have to go to unusual lengths to get their needs met. The child with a physically abusive father is likely to form a Mistrust/Abuse schema, expecting that others will hurt or abuse them. The child allowed to do whatever they want might develop an Entitlement schema (“I am superior”, “ordinary rules don’t apply to me”) if they are simultaneously taught that they need not be polite or exercise self-control because they are special. Alternatively, they might as well or instead develop an Insufficient Self-Control/Self-Discipline schema, a pattern of refusal to exercise self-control or frustration tolerance which might emerge if the adults in the child’s life fail to model and teach self-discipline.
These experiences and resulting expectations are rarely a good preparation for interacting with mainstream adult society. A dysfunctional environment is likely to produce dysfunctional schemas, although the schemas that result also depend on the interaction between the child’s temperament and the environment. A child high in neuroticism of a distant, aloof mother is likely to form deep lifelong expectations that others will be never provide the love and nurturance they crave (Emotional Deprivation schema), whereas a child lower in neuroticism given the same parenting may not. A key way Schema therapy differs from present-focused CBT is in its consistent and greater attention to Early Maladaptive Schemas: problematic patterns of belief and emoting in adulthood, that explain current problems, and that likely had their origins in childhood or adolescence because fundamental needs were not adequately met.
The study
Pilkington and colleagues analysed 33 studies across 11 countries. All studies measured Early Maladaptive Schemas using one of the several versions of the Young Schema Questionnaire (YSQ) and all but one used a retrospective design (one used a longitudinal design). Indicators of adverse childhood events were drawn from 9 different questionnaires across studies (notably, the commonly used ACE checklist was not among them). Childhood adversity categories were created for maternal, paternal and general emotional neglect, general physical neglect, maternal, paternal and general emotional abuse, general physical abuse and general sexual abuse. Meta-analyses were then performed on the (available) correlations between each of the 18 schemas and each of the childhood adversity categories, resulting in 124 aggregated correlations. The number of studies informing the aggregate correlations varied from 5 to 14.
It is difficult to adequately summarise 124 correlations – certainly in a blog I aim to keep brief. Dedicated schema therapists will want to review the correlation tables in the original article as all will be of interest. Apparently there wasn’t sufficient consistent information across the original studies to perform moderation analyses and there was substantial inter-study variability in 69% of the aggregate correlations (i.e., I2 > 50%), so there is a lot more left to unpack from this literature. However, given the mammoth undertaking that this study would have entailed, and how useful it is to have this summary of the existing literature, the study still makes a huge contribution. So, in lieu of more robust ways to analyse the quality of the correlations, you’ll have to put up with my hack/sports commentator level of commentary.
The strongest correlation was between maternal emotional neglect and Emotional Deprivation, r = .51, based on 9 studies. Although the strength of relationship varied between studies (I2 = 60%), the precision of this aggregate correlation was also better than most, with a 95% confidence interval of .42 to .59. The emotional deprivation schema is measured on the YSQ by items like “I don’t have people to give me warmth, holding and affection” – notice the present focus. Emotional neglect was measured by low scores on items like “my parents showed a lot of interest in me”, “my mother spoke to me in a warm and loving voice” – notice the historical focus. The 26% shared variance represents good support for schema therapy theory: the hypothesis that maternal disengagement is associated with emotional deprivation schemas is well documented throughout early schema therapy texts. Also consistent with schema texts (e.g., Reinventing Your Life), paternal emotional neglect was associated with Emotional Deprivation but with a weaker relationship (r = .36) although the overlap in confidence intervals indicates this is probably not a statistically significant difference. In fact, overall, there was no real evidence that a schema would have a stronger relationship with parenting depending on the parent’s gender. The precision of the estimate for paternal neglect was considerably lower than for maternal neglect; even though based on 8 studies, there was 94% variability between studies and the 95% confidence interval ranged from .08 (very weak) to .58 (very strong). What this means is there were probably important unmeasured moderators - factors that determine whether paternal emotional neglect will lead to an emotional deprivation schema in a given sample. These could be boring methodological issues, but they might also be due to substantive differences in study sample profiles (e.g., some may have more protective factors than others). An obvious candidate would be whether some samples differed systematically in the extent to which the mother was also emotionally neglectful or not; presumably the potentially detrimental effect of emotional neglect from the father might be protected by sufficient nurturance from the mother.
It may be perhaps surprising that the correlations between abuse and the Mistrust/Abuse schema weren’t higher, with r = .25 [.14, .35], I2 = 85% for physical abuse and r = .25 [.13, .38], I2 = 77% for sexual abuse. The high between-study variability again hints at important moderator effects. The different places and processes by which samples are recruited across studies could lead to some studies selecting for more or less resilient samples. Nevertheless, the upper limit of the confidence intervals also suggests that even with retrospective designs – where people are rating both their historical antecedents and current effects at the same time, potentially inflating the correlation as both could be influenced by current mood, stressors etc – the relationship between physical or sexual abuse and Mistrust/Abuse schema is moderate at most. Hardly deterministic. Of course, schema therapy itself doesn’t suggest it would be. The quality of an individual’s coping strategies, and other protective or resilience-building factors (e.g., other supportive family or community members, strong schools or recreational activities, physical and temperamental factors and more) can mitigate whether these experiences translate into schemas. By the way, Tim Fischer – a former Master’s student supervisee – did a small study showing that parental behaviour as measured by the Young Parenting Inventory was not associated with early maladaptive schemas once psychological flexibility -according to Acceptance and Commitment Therapy – was included as a predictor (Fischer et al., 2016). The clinical implication is clear: we should not assume which schemas will affect a client based on developmental adversity alone.
Other correlations further illustrate the diverse impacts of different types of childhood adversity. Physical abuse had similar strength of relationship with Emotional Deprivation, Social Isolation (“I’m different to everyone else; I don’t belong anywhere”), Defectiveness (“I’m defective/worthless/unlovable”), Vulnerability to Harm (“something terrible could happen any at moment and I won’t cope”) and Subjugation (“I have to do what others want”) as it did to Mistrust/Abuse. Sexual abuse had similar strength of relationship to Emotional Deprivation, Abandonment (“others won’t continue to love, support and protect me”), Social Isolation, Enmeshment (lack of own identity combined with excessive closeness and dependence on another) and Subjugation as it did to Mistrust/Abuse. Emotional abuse included being ridiculed, insulted, shamed and ‘destructively’ criticised. Emotional abuse was especially associated with the Emotional Deprivation schema (r = .44 [.35, .51]), with moderate correlations of similar strength (~.3 - .35) with Mistrust Abuse, Social Isolation, Defectiveness/Shame, Failure (“I’ll never be able to achieve what normal people can achieve”) , Vulnerability to Harm and Subjugation. In addition to Emotional Deprivation, Emotional neglect was moderately correlated with Mistrust/Abuse, Social Isolation, Defectiveness/Shame and Emotional Inhibition (excessive tendency to suppress emotional expression).
Otherwise, remaining schemas had either no relation or weak relationship to childhood adversity. Entitlement had a weak relationship with emotional abuse. Unrelenting Standards (one must strive to attain very high standards) had weak relationships with paternal emotional neglect and emotional abuse. Self-Sacrifice (excessive voluntary focus on meeting others’ needs) had weak relationships with emotional, physical and sexual abuse. Dependence/Incompetence (“I can’t handle everyday responsibilities”) had weak relationships with emotional, physical and sexual abuse and neglect.
As meta-analysis is only as strong as the original studies, there were some limitations. The sample was predominantly female (15 studies recruited only females, 2 recruited males only). Sample sizes varied between 30 – 1102 participants. Thirteen studies used clinical samples and 5 of these were of people with eating disorders. Future studies exploring adversity-schema relationships could seek to fill gaps in representativeness. It would have been nice to see the meta-analysis summarise the % of participants who had actually experienced each of the adverse childhood events. Obviously, longitudinal designs would be a stronger test of the theory but such studies would be incredibly expensive and effortful to conduct.
Pilkington’s study represents a great attempt to cover the breadth of schema therapy’s hypotheses about the relationship between parenting and specific early maladaptive schemas. It will certainly be cited as a seminal collation of research for many years to come. However, for all the effort, it still only represents a portion of the theory, and a complete account would also need to incorporate measures of temperament, coping and protective factors. For now though, these findings should reinforce clinicians’ confidence that clients’ adverse parenting experiences remain relevant influences on their expectations of themselves and others.
Clinical implications
· The strongest relationship between a schema and parenting experience was: Emotional Deprivation strongly related to maternal emotional neglect. This finding was relatively precise.
· Emotional abuse had moderate associations with a number of schemas: Emotional Deprivation, Abandonment, Mistrust/Abuse, Social Isolation, Defectiveness/Shame, Vulnerability to Harm, Subjugation.
· Physical and sexual abuse were weakly associated with a diverse range of schemas
· Clinicians should take care not to prematurely assume the most problematic schema for a client based on their developmental history; similar experiences can have a myriad of impacts. Schema therapy encourages attending to the individual’s temperament and coping styles, and strengths/protective factors should also be considered when assessing which schemas account for most of the client’s presenting problems and obstruct their goals.
· Beginning schema therapists could use these findings to guide their process of hypothesis-testing: looking for those with a stronger ‘on average’ relationship with the client’s experienced adversity.
For the original article go to: https://onlinelibrary.wiley.com/doi/10.1002/cpp.2533
REFERENCE
Pilkington, P.D., Bishop, A., & Younan, R. (2020). Adverse childhood experiences and early maladaptive schemas in adulthood: A systematic review and meta-analysis. Clinical Psychology and Psychotherapy. 1-16.
Young, J.E., & Klosko, J.S. (1994). Reinventing Your Life: How to Break Free From Negative Life Patterns. New York: Plume.