Article Review Assignment? . Author names for all authors, and journal info (title of article, year published, title of journal it was published in) ?? 2. General issue and topic being in

Early Relational Trauma and Self Representations: Misattributing Externally Derived Representations as Internally Generated

Chui-De Chiu and Hoi Lam Ho The Chinese University of Hong Kong

Marieke S. Tollenaar and Bernet M. Elzinga Leiden University

Tianyang Zhang Soochow University

Objective: Early relational trauma has been posited to be responsible for dysfunctional self schema as negative feedback derived from abusive close others may influence the development of self-evaluation. However, the association between early relational trauma and negative self-evaluation has proven inconsistent. In addition to the evaluative aspect, early relational trauma may impact on the procedural aspect of self schema, with a difficulty in differentiating mental representations derived from others from those generated internally by the self. Method: To test this hypothesis, the authors adopted a source attribution paradigm tapping on the distinction between mental representations generated by the self or derived from another person in a nonclinical sample, together with scales measuring self-evaluation and early relational experiences. Results: The results showed that individuals with early relational trauma tended to attribute the representations externally derived as internally generated, although there were no associations between early relational trauma and self-evaluation. Importantly, early relational trauma had unique contribution to source misattribution independent from common covariates including early nonrelational trauma, parental dysfunction, general memory function, and negative affect states. Con- clusions: Erroneously identifying information derived from other people as self-generated may be a specific sociocognitive propensity linked to early relational trauma and may impact upon the develop- ment of self schema.

Clinical Impact Statement With a tendency to misattribute representation derived from other people as self-generated, adult survivors with early relational trauma may be susceptible towards other’s evaluation and have instable sense of self. Meanwhile, this source misattribution tendency may extend to other types of representations, such as the memory of imaginary experiences which are brought to mind by other people. Clinicians who work with clients with a history of early relational trauma should be cautious to adopt approach that could be suggestive, and interpret the reported source of memory carefully.

Keywords: child abuse, self-esteem, shame, source monitoring, trauma

Self schema, the inner organization of the experience, evalua- tion, and knowledge pertinent to the self, is a critical determinant of mental health and psychological life in adulthood (Kim, Thi- bodeau, & Jorgensen, 2011; Sowislo & Orth, 2013). This inner structure of self-related mental representations is considered to be shaped by feedback from other people about the self, especially

through dyadic interactions with close others, such as caregivers, at the early developmental stage (Fonagy & Target, 1997). Unfor- tunately, early adverse relational experiences are common in the general population (Edwards, Holden, Felitti, & Anda, 2003), including those that elicit extreme stress and cannot be readily assimilated by ordinary experience, such as emotional, physical, or

This article was published Online First April 26, 2018. Chui-De Chiu and Hoi Lam Ho, Department of Psychology, The

Chinese University of Hong Kong; Marieke S. Tollenaar and Bernet M. Elzinga, Clinical Psychology Unit, Institute of Psychology, Leiden University; Tianyang Zhang, School of Public Health, Soochow Uni- versity.

This research was supported in part by a start-up grant from Department of Psychology, The Chinese University of Hong Kong, a grant of Research Committee Funding from The Chinese University of Hong Kong (4052099), and a grant from the Research Grants Council, Early Career

Scheme (24618115) to Chui-De Chiu. We thank Hiu Ching Chui, Pik Ka Li, Pak Kwan Lo, Ho Yeung Lam, and Ka Wai Un, for their assistance in data collection and data cleaning, and the students who kindly participated in the study. Parts of the data have been presented at the XV Congress of the International Society for the Study of Personality Disorders, Germany, Heidelberg, September 25–28, 2017.

Correspondence concerning this article should be addressed to Chui-De Chiu, Department of Psychology, The Chinese University of Hong Kong, Hong Kong S.A.R., The People’s Republic of China. E-mail: [email protected]

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

Psychological Trauma: Theory, Research, Practice, and Policy © 2018 American Psychological Association 2019, Vol. 11, No. 1, 64–72 1942-9681/19/$12.00 http://dx.doi.org/10.1037/tra0000369

64

sexual maltreatment (Terr, 1991; Van der Kolk & van der Hart, 1989). The humiliation and other negative messages from abusive caregivers may be incorporated into the representation of the self, leading to a dysfunctional self schema. Early relational trauma hence may impact upon the formation of self schema and increase the susceptibility to mental illness (Follette, Polusny, Bechtle, & Naugle, 1996).

The relationship between early relational trauma, dysfunctional self schema, and mental illness has been vigorously investigated in individuals with a history of childhood maltreatment. Results, however, have been inconsistent. Some studies reported that early relational trauma was associated with the evaluation of self schema, including lower self-esteem (e.g., Bolger, Patterson, & Ku- persmidt, 1998; Briere & Runtz, 1990; Fergusson, McLeod, & Horwood, 2013; Silvern et al., 1995; Turner, Finkelhor, & Orm- rod, 2010; Wong et al., 2009), a tendency to blame oneself for being the target of abuse or to attribute the cause of a negative event to the self (e.g., Alessandri & Lewis, 1996; Andrews, 1995; Andrews, Brewin, Rose, & Kirk, 2000; Feiring, Cleland, & Simon, 2009; Filipas & Ullman, 2006; Gold, 1986; Kim, Talbot, & Cic- chetti, 2009; Negrao, Bonanno, Noll, Putnam, & Trickett, 2005), and a propensity to associate the self with negatively valenced words (van Harmelen, de Jong, et al., 2010). Dysfunctional self schema in terms of elevated abuse-related shame (Feiring, Taska, & Chen, 2002) mediated the link between early relational trauma and psychiatric symptoms. In contrast, others did not find that early relational trauma correlated reliably with self-esteem (e.g., Brayden, Deitrich-MacLean, Dietrich, Sherrod, & Altemeier, 1995; Feiring et al., 2009), the dispositional tendency to experi- ence shame (i.e., shame proneness; Feiring, Taska, & Lewis, 2002; Feiring et al., 2009), and the propensity to associate the self with negative words (Rüsch et al., 2011). Early relational trauma may not be necessarily linked to negative self-evaluation, at least in the nonclinical population (see, for reviews, Jumper, 1995; Rind, Tromovitch, & Bauserman, 1998).

Several methodological problems can be noted in these studies. First, victim-abuser relationship was not specified at times. Abuse perpetrated by trusted people is expected to be more impactful than abuse inflicted by nonclose others on the development of self schema (Cole & Putnam, 1992; Freyd, 1994; Terr, 1991). The inconsistent association between self-evaluation and early rela- tional trauma may arise because of inclusion of maltreatment inflicted by nonclose others such as extrafamilial people. Second, covariates were frequently ignored. Early relational trauma was found to correlate with anxiety and depression (Spinhoven et al., 2010), which can bias information processing (Dalgleish & Watts, 1990) and may distort the appraisal of early experience and the self in clinical patients. This may account for the absence of the link between early relational trauma and negative self-evaluation in nonclinical populations (Jumper, 1995; Rind et al., 1998). Further- more, maltreating caregivers were found to be less sensitive and responsive to their offspring (Bousha & Twentyman, 1984). How- ever, parental dysfunction has usually been neglected in the study of early relational trauma (Kagan, 2013). This neglect introduces difficulty to partial out the putative confounding effect of the low sensitivity and responsibility due to parental dysfunction.

In addition to the evaluative aspect of the self (e.g., I am to blame and I am not worthy of any love), the impact of early relational trauma may also be observed on the procedural aspect of

the self, the cognitive processing of mental representations perti- nent to the self. An ability to differentiate the source of represen- tations between those derived from other people and those gener- ated by the self appears to be critical, as from the social feedback in early close relationships, children start to develop the sense of self and build up their own views toward the self (Cole & Putnam, 1992). Indeed, some evidence suggests that effective source attri- bution emerges early (Lindsay, Johnson, & Kwon, 1991) and it can already be observed in 6-year-old children (Foley, Johnson, & Raye, 1983). We postulate that the development of this cognitive ability may be hindered by early relational trauma. This source misattribution tendency may render individuals with early rela- tional trauma susceptible to the internalization of externally de- rived information, including those negative and humiliating ap- praisals toward the self from abusive close others (Schore, 2002). Clinical studies have shown that patients with several trauma- related disorders have a tendency to confuse the source of self- related representations and other representations derived from other people, including patients with posttraumatic stress disorder (Golier, Harvey, Steiner, & Yehuda, 1997) and pathological dis- sociation (Chiu et al., 2016). This is also found in a subgroup of patients with schizophrenic spectrum disorders (Fisher, McCoy, Poole, & Vinogradov, 2008; Keefe, Arnold, Bayen, McEvoy, & Wilson, 2002; Vinogradov et al., 1997; Vinogradov, Luks, Schul- man, & Simpson, 2008). It remains unclear, however, whether the findings in these patients result from early relational trauma or other factors pertinent to psychopathology.

Findings from neuro-cognitive studies also suggest that there may be a link between source misattribution and early relational trauma. Laboratory studies have indicated that the cortical midline structure (CMS; Northoff & Bermpohl, 2004), a neural substrate responsible for the processing of self-related information including the discrimination between representations derived internally by the self or externally from others (Allen et al., 2005; Simons, Henson, Gilbert, & Fletcher, 2008; Wang, Metzak, & Woodward, 2011), has been associated with early relational trauma. That is, structurally, a volume reduction in the CMS has been observed in individuals with early relational trauma (right medial prefrontal cortex in Tomoda et al., 2009; dorsomedial prefrontal cortex in van Harmelen, van Tol, et al., 2010). Functionally, child emotional abuse correlated with the reactivity in dorsomedial prefrontal cor- tex at an interpersonally stressful situation (van Harmelen, Hauber, et al., 2014) and was associated with CMS hypoactivation during encoding and recognition of emotional words (van Harmelen, van Tol, et al., 2014). These neuro-cognitive findings implied that there may be a link between early relational trauma and dysfunc- tional processing of self-related representations, which could lead to the misattribution of the source of information.

This study aimed to fill the gaps in the literature regarding the relationship between self schema and early relational trauma. First, to address the multidimensionality of self schema, both the eval- uative aspect (self-esteem and shame proneness) and the proce- dural aspect (the discrimination of self- and nonself representa- tions) were assessed. Second, two types of traumatizing events were differentiated. Relational trauma denotes abuse inflicted by trusted people (e.g., caregivers), which may impose conflict in forming stable attachment to them; nonrelational trauma denotes other traumatic experience including abuse inflicted by nonclose others or other extremely stressful events, which involve less

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

65TRAUMA AND SELF REPRESENTATIONS

interpersonal nature such as traffic accidents or natural disasters (Freyd, 1994). Finally, the potential confounding effects of nega- tive affect states and parental dysfunction were addressed. A nonclinical sample was used to exclude the potential effect of psychiatric status.

We hypothesized that there would be an association between self schema and early relational trauma. This association, however, may not emerge on both of the evaluative and procedural aspects of the self. We predicted an association between early relational trauma and a deficit in the procedural aspect of the self due to the structural or functional changes in the neuro-cognitive underpin- nings responsible for self-referential processing. In contrast, we did not expect that early relational trauma would correlate with the evaluative aspect of self schema. The associations with the eval- uative aspect, if any, may be small and possibly be explained by the confounding variables, as the associations in the nonclinical populations have appeared unreliable (Jumper, 1995; Rind et al., 1998).

Method

Participants

This study was part of a larger research project investigating the relationships between early relational experience, sociocognitive functioning, and psychosocial well-being. The research project was approved by the research ethics committee at the Faculty of Social Science, The Chinese University of Hong Kong. A total of 159 college students were recruited from the campus. Participants were eligible if they are Cantonese native speaker and able to read traditional Chinese. The average age of this sample was 20.9 years old (SD � 1.3, ranging from 18–24) and 43% were male. A compensation of HKD 70 (per hour) was remunerated after their participation.

Instruments

Evaluative aspect of self schema. Two scales were used to assess self-esteem and self-conscious emotions (e.g., shame and guilt proneness). The Rosenberg Self-Esteem Scale (Rosenberg, 1965), a 10-item instrument, was used to assess self-esteem. Par- ticipants were instructed to indicate their agreement with each statement on a 5-point Likert scale from 1 (strongly disagree) to 5 (strongly agree). The total score indicates the level of self-esteem. This scale has been adopted in a Chinese population and demon- strated good reliability (Martin, Thompson, & Chan, 2006). The Test of Self-Conscious Affect (TOSCA; Tangney, Wagner, & Gramzow, 1989) is used to measure shame and guilt proneness. The TOSCA consists of 15 daily life scenarios. Each scenario is followed with reaction styles pertinent to self-conscious emotions, including shame and guilt proneness, on a 5-point scale. Shame and guilt can be distinguished by their focus of self-evaluation, guilt focuses on specific behavior and shame focuses on the entire self (Lewis, 1971). Shame, rather than guilt proneness, has been found to relate to psychopathology (Tangney, Wagner, & Gramzow, 1992). The Chinese version of TOSCA has been developed and showed acceptable reliability (Gao, Qin, Qian, & Liu, 2013).

Early adversities. The Brief Betrayal Trauma Survey (BBTS; Goldberg & Freyd, 2006) was used to assess early relational and nonrelational trauma. The BBTS covers various potentially trau- matizing events in childhood including five items for those im- posing threat to attachment (e.g., emotional, physical, or sexual assaults by close people) and another five items for those without the relational feature (e.g., physical and sexual assaults by non- close people, accidents, natural disasters). Each item is scored on a 3-point scale ranging from 1 (never) to 3 (more than two times) and the scores are converted into a dichotomous scale for analysis (0 � no, 1 � yes). We deployed two scoring schemes for early relational trauma. First, we divided participants into two groups according to the presence of at least one type of early relational trauma. Second, a total score (from 0 to 5) was calculated as the measures of cumulated early relational and nonrelational trauma, respectively. The Measure of Parenting Style (Parker et al., 1997) was applied to assess the perceived parenting style of both parents before the age of 16. Two types of parental dysfunction, insuffi- cient parental care and strict discipline, were assessed by 10 items, with a 4-point scale ranging from 0 (not true at all) to 3 (extremely true). The two subscale scores were used respectively to represent parental indifference and overcontrol in the analysis.

Negative affect states. The levels of anxiety and depression experienced in the last week were assessed by the Hospital Anx- iety and Depression Scale (HADS; Zigmond & Snaith, 1983). The scale consisted of seven items for each of the two subscales, anxiety and depression. Participants were required to choose the description which describes their mood state the best (scores ranging from 0 to 3).

Procedure and Design

Source monitoring paradigm. Following the design of self- versus-other source monitoring tasks (Chiu et al., 2016; Fisher et al., 2008; Mitchell & Johnson, 2009) and participants’ feedback from a pilot test, we developed a computerized version of the source monitoring task. Test materials were simple sentences composed of a subject, a verb, and an object. Two categories of sentences were prepared, one pertinent to occupations and the other to animals. Some examples of the test materials were “Post- men deliver letters,” “Students read textbooks,” “Monkeys eat bananas,” and “Bees collect nectar.” All of the sentences fitted the common knowledge for Hong Kong residents and the linguistic features of Cantonese. The sentences were randomly divided into three sets for the experimenter-provided condition, the participant- generated condition, and new items for a recognition test.

In the study phase, the sentences of the first two sets were displayed one by one on the screen, and participants were asked to read the sentences aloud. In the experimenter-provided condition, the full sentence with the subject, verb, and object were presented (10 items for each category). Participants merely repeated the sentence. In the participant-generated condition, an incomplete subject with the first character and blank grids indicating the number of characters were presented together with the verb and object. Participants were asked to give a meaningful sentence by self-generating the subject according to the cues (10 items for each category). The two categories of sentences were presented sepa- rately in two blocks and the order of the two blocks were random-

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

66 CHIU, HO, TOLLENAAR, ELZINGA, AND ZHANG

ized across participants. The two conditions appeared intermit- tently within one category.

In the test phase, the verb and the object of each sentence were displayed on the screen, but the subject was missing. Participants were asked first whether the sentence had been studied in the study phase. Subsequently, participants were asked about the source of the sentence when they had answered that it had been learnt in the study phase. Again, the two categories of sentences were presented in two separate blocks and the order was randomized across participants. Within each category, the three types of sentences were presented randomly (for each category, 10 experimenter- provided items, 10 participant-generated items, and 20 new items). Five indices were formed by calculating the accuracy rates of item recognition of the three experimental conditions (experimenter- provided, self-generated, and new) and the accuracy rates of source attribution of the two experimental conditions (experimenter-provided and self-generated).

A preliminary analysis was conducted to examine the validity of this task. The results showed a higher accuracy rate for item recognition of self-generated items than items provided by the experimenter in the full sample (M � .28, SD � .15, paired t � 23.99, p � .0001), which is a standard self-generation effect on item-recognition (Bertsch, Pesta, Wiscott, & McDaniel, 2007). The result lent support for the success of our experimental manip- ulation.

Procedure

Participants were assessed individually in our laboratory. Upon arrival, the study was explained, and the written informed consent was obtained. Then, behavioral experiments including the source monitoring task were administered. Participants performed the study phase first. After a distractions phase with irrelevant percep- tual tasks for 25 min, the memory task assessing item recognition and source attribution was given. Finally, the self-report scales were administered. Participants were debriefed after their partici- pation.

Results

Early Relational Trauma, Nonrelational Trauma, and Parental Dysfunction

On the basis of the retrospective report of trauma history, participants were divided into two groups, one reporting early relational trauma (n � 45) and the other without early relational trauma (n � 111).1 This prevalence rate (28.8%) is comparable to the rate found in a recent review (ranged from 8.7% to 26.6%) in Chinese nonclinical population (Fang et al., 2015). We tested whether the two groups showed differences in other early adver- sities including nonrelational trauma and parental dysfunction. Indeed, significant differences were found in the scores of early nonrelational trauma, F(1, 154) � 25.51, p � .001, �p

2 � .14; parental indifference, F(1, 154) � 56.30, p � .001, �p

2 � .27; and parental overcontrol, F(1, 154) � 14.94, p � .001, �p

2 � .09 (see Table 1). The significant between-groups differences reflected the positive correlations of the measures for other early adversities with the scale of cumulated relational trauma; for nonrelational

trauma, parental indifference, and parental overcontrol, r(n � 156)s � .48, .56, and .28, ps � .001, and ds � 1.09, 1.35, and 0.58.

Early Relational Trauma and Self Schema

The descriptive statistics of the measures for self schema were also summarized in Table 1. First, with respect to the evaluative aspect of self schema, the two groups did not differ significantly in terms of self-esteem and levels of shame and guilt proneness (ps � .4). Thus, nonclinical individuals with early relational trauma did not report low self-esteem and elevated shame and guilt prone- ness.2

Regarding the procedural aspect of self schema, a significant between-groups difference was found in source attribution for experimenter-provided items, F(1, 154) � 10.34, p � .002, �p

2 � .06, but not for self-generated items, F(1, 154) � 0.03, p � .85, �p

2 � .0002. Individuals with early relational trauma made more errors on attributing the source of experimenter-provided items in comparison with individuals without this history. The between- groups difference could not be accounted for by general memory function as the two groups performed equally well on item recog- nition for self-generated and experimenter-provided items (ps � .11). Also the two groups did not differ in the rate of successful rejection of new items (p � .60).

Covariate Analysis

To eliminate the potential confounding effects from the com- mon covariates on the association between early relational trauma and source misattribution, a hierarchical regression model was constructed (Neter, Kutner, Nachtsheim, & Wasserman, 1999). The accuracy rate of source attribution for experimenter-provided items served as the dependent variable. Predictors including gen- der and negative affect states (state depression and anxiety), gen- eral memory function (item recognition for self-generated and experimenter-provided items), other early adversities (nonrelational trauma and parental dysfunction), and early relational trauma were incorporated into the model step by step. The increase of the coeffi- cient of determination between a current model and its preceding model was used to index the unique contribution of the new vari- able(s) in the current model. The results of the regression analysis were summarized in Table 2.

At the baseline model, we placed gender, anxiety, and depres- sion into the model. The model was not significant and none of these three variables reached significance. Second, we added two

1 Eight participants reported that they had a history of early nonrelational trauma but not early relational trauma. Because of the small sample size, we combined them with those without a history of early relational trauma. Similar between-subject differences were found when using this sample that included the eight participants and another sample without them.

2 A similar pattern of correlation has been demonstrated between neg- ative self-evaluation and negative affect states across the two groups. Low self-esteem (rs � �.34 to �.60, ds � 0.72 to 1.50) and high shame proneness (rs � .26 to .33, ds � 0.54 to 0.70) were associated with anxiety and depression in both of the groups, as documented (Tangney et al., 1992; Zeigler-Hill, 2011).

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

Collepals.com Plagiarism Free Papers

Are you looking for custom essay writing service or even dissertation writing services? Just request for our write my paper service, and we'll match you with the best essay writer in your subject! With an exceptional team of professional academic experts in a wide range of subjects, we can guarantee you an unrivaled quality of custom-written papers.

Get ZERO PLAGIARISM, HUMAN WRITTEN ESSAYS

Why Hire Collepals.com writers to do your paper?

Quality- We are experienced and have access to ample research materials.

We write plagiarism Free Content

Confidential- We never share or sell your personal information to third parties.

Support-Chat with us today! We are always waiting to answer all your questions.