Transcript the full sized image

Childhood Trauma as a Mediator of the Relationship between
Borderline Personality Disorder and Emotion Recognition
Julie A. Meinert
Faculty Advisor: Kenneth N. Levy, Graduate Advisor: Benjamin N. Johnson
Introduction
Method Cont.
Results Cont.
• Borderline Personality Disorder (BPD) is a serious and highly
prevalent psychological disorder characterized by emotional
lability, unstable self-image, impulsivity, chaotic interpersonal
relationships, and high levels of suicidality, among other
features (Levy, 2013; Skodol et el., 2002).
• The literature is currently mixed on BPD’s relationship with
emotion recognition abilities. However, a recent meta-analysis
found significant deficits in emotion recognition in BPD
compared to healthy controls, especially in the context of
displays of anger and disgust (Daros, Zakanis, & Ruocco, 2012).
• Additionally, those with BPD often report high levels of
traumatic childhood experiences.
• Maltreated children have been found to do worse on emotion
recognition tasks overall. Further, deficits in interpreting certain
emotional displays have been found to be associated with
certain types of trauma experienced, such as sexual abuse.
Hypotheses:
1. Increases in BPD symptoms will be associated with
increased trauma severity.
2. Increases in both BPD symptoms and trauma severity will be
associated with decreased emotion recognition ability.
3. Trauma severity will mediate the relationship between BPD
and emotion recognition.
• Reading the Mind in the Eyes (RME; Baron-Cohen et al., 2001);
evaluates emotion recognition abilities. It contains 36 pictures
of the eye region of the face. Participants select the emotion
being displayed from four given options. Three emotional
subcategories were identified (positive, negative, and neutral).
Statistical Analysis
• One-way analysis of variance was used to examine the
differences between groups across RME and CTQ scores.
• Spearman rank order correlation analyses determined
associations between BPD symptoms and RME subscales.
• To address our second hypothesis, a mediation analysis
(Preacher & Hayes, 2004) was conducted to assess if trauma
severity mediated the relationship between BPD symptoms and
RME scores. The SPSS PROCESS macro (Hayes, 2012) with 5,000
bootstrapped samples was utilized to assess:
1. the effect of BPD symptoms on CTQ scores (“a path”);
2. the effect of CTQ scores on RME scores, controlling for
BPD symptoms (“b path”);
3. the direct effect of BPD symptoms on RME scores (“c’
path”);
4. the indirect effect of BPD on RME through CTQ (i.e.,
mediation).
• There was trend level difference between the clinically diagnosed
BPD group and the control group on negative, F(1,53) = 3.81, p =
.056, η2 = .068, and positive RME scores, F(1,53) = 3.90, p =
.054, η2 = .070.
• BPD symptom severity was positively associated with overall
trauma severity, r = .71, p <.001.
• Dimensional BPD was negatively associated with negative (r = .29, p = .025), trend level for positive (r = -.25, p = .061), but not
neutral (r = .19, p = .16) or overall (r = -.15, p = .25) RME scores.
• Emotional neglect (r = -.34, p = .009) and overall trauma (r = -.28,
p = .033) were negatively associated with negative RME scores.
• Contrary to hypothesis, the relationship between dimensional
BPD severity and negatively valenced RME scores was not
mediated by overall trauma, β = -.0009, SE = .0035, 95% CI [.0086, .0048] (Figure 1), or emotional neglect β = -.0040, SE =
.0029, 95% CI [-.0097, .0016] (Figure 2).
Method
Participants
• Sixty women were included in this study.
• 22 had a clinical diagnosis of BPD
• 6 were subthreshold for BPD, meeting < 5 BPD symptoms.
• 32 were control participants: 11 were temperamentally
matched (on impulsivity and negative emotionality), 21 were
non-temperamentally matched both meeting < 3 BPD
symptoms.
Measures
• International Personality Disorder Examination (IPDE; Loranger,
1999); a clinician rater diagnostic semi-structured interview. The
9-item BPD subscale (scored 0-2) determines severity of BPD
symptoms.
• Childhood Trauma Questionnaire (CTQ-SF; Bernstein et al.,
2003); a self-report measure that assesses for type and severity
of traumatic experiences in childhood.
Results
Figure 1. Mediation of BPD-Negative Emotion Recognition Relationship by Overall Trauma
Overall Trauma
b: -.0003
a: 2.55*
Indirect effect: -.0009
BPD Symptom
Severity
Negative Emotion
Recognition Deficit
c’: -.0062
Figure 2. Mediation of BPD-Negative Emotion Recognition Relationship by Emotional Neglect
BPD Symptom
Severity
Negative Emotion
Recognition Deficit
c’: -.0030
Indirect effect: -.0040
a: .703*
b: -.0057
Emotional Neglect
Note: Data shown are parameter estimates from a mediation analysis following the recommendations of Preacher and Hayes (2004)
and utilizing the SPSS PROCESS macro for indirect effects testing with 5,000 bootstrapped samples (Hayes, 2012).
* p < .05
Discussion
• Dimensional BPD symptoms were associated with deficits in
recognizing negative emotional displays, suggesting that
problems with recognizing others emotions may contribute to the
symptoms of BPD, including emotional lability and interpersonal
difficulties.
• Childhood emotional neglect was associated with difficulty
recognizing negative emotions, suggesting that early emotionally
void environments may indicate a lack of coherent emotionspecific mental representations later in life.
• Since trauma severity did not explain the relationship between
BPD severity and emotion recognition ability, both trauma and
BPD symptoms may be important independent contributors to
emotion recognition difficulties.
• This finding provides an explanation for the mixed findings in the
field regarding BPD and emotion recognition abilities as trauma
severity may have an independent effect on emotion recognition
outcomes and should therefore be assessed in future research.
• The study was limited by a small sample size and a possible threat
to ecological validity due to the laboratory setting in which the
RME task was conducted.
References
References available upon request. Please email Julie Meinert at
[email protected].