Degree Name

Doctor of Philosophy (Clinical Psychology)


School of Psychology


Background: Borderline personality disorder (BPD) is a mental health disorder characterised by impairments in affect, behaviour, identity, and interpersonal relationships. Personal agency, operationalised by locus of control, refers to the extent that individuals feel that they have control over their outcomes. Low personal agency is understood as the tendency for individuals to perceive outcomes as a result of external factors, such as luck, chance, or fate. High personal agency is the extent that individuals perceive outcomes as dependent on their own personal characteristics, such as effort or ability. Low personal agency has been shown to be associated with greater BPD symptoms; however, no known studies have assessed these constructs together over time. Adult attachment styles consist of secure, preoccupied, fearful, and dismissive, and previous research has shown a relationship between BPD symptoms and insecure attachment. The association between low personal agency and adult attachment styles has been less researched, though initial evidence suggested a link particularly in BPD populations. Three studies were conducted to examine the constructs of personal agency, BPD, adult attachment styles, and pathological personality traits.

Method: Study One (N = 192) examined a sample of sub-clinical BPD participants compared with healthy controls. Study Two (N = 57) replicated the design of Study One by using a clinical sample of treatment-seeking BPD outpatients currently meeting criteria for BPD compared with those not meeting criteria. Study Three (N = 57) assessed these outpatients over 12 months across three time points and used personal agency measures at intake to predict treatment outcomes. All participants completed a Dialectical Behaviour Therapy (DBT) group and were followed up for 12 months. Validated measures of agency, psychopathology, and attachment were collected and written informed consent was obtained following ethics approval for each study.

Results: Study One found that individuals with sub-clinical BPD displayed lower personal agency and higher fearful and preoccupied attachment styles in their close relationships, compared with controls. Lower personal agency was associated with insecure adult attachment styles. In addition, using multiple mediation modelling, an indirect effect of personal agency on BPD symptoms was found for preoccupied, fearful, and secure attachment, but was not significant for dismissive attachment. Study Two aimed to replicate these results in a clinical sample and found that low personal agency was again associated with BPD symptoms. Results also showed that the BPD group had significantly poorer personal agency than the clinical control group. Additionally, the BPD group had greater levels of general psychopathology and pathological personality traits. Further, low personal agency was significantly associated with greater preoccupied adult attachment. Using multiple mediation modelling for the full sample, the direct effect of personal agency on BPD symptoms was significant. Study Three demonstrated that personal agency was a stable variable across time in this sample and may be resistant to change despite effective group DBT treatment. In addition, results illustrated that personal agency was able to predict BPD symptom severity over 12 months. Moreover, BPD symptoms significantly reduced over the course of the 12-week treatment program and reductions were generally stable 12 months later. Despite significant reductions in BPD symptom severity, over three-quarters (78%) of participants who initially met criteria for BPD continued to meet criteria at follow-up. In addition, the majority of participants who originally met criteria for BPD continued to meet criteria at follow-up. Lastly, low personal agency was able to predict high rates of negative affect and disinhibition over time. As a result, suggestions were made to target the therapeutic alliance and attachment style in order to influence personal agency. Overall, these studies used a combination of a sub-clinical sample (Study One), cross-sectional clinical sample (Study Two), and longitudinal clinical sample (Study Three) and all paved the way to further understand the relationship between personal agency and BPD.

Conclusions: The three studies highlighted the role of personal agency within BPD. Low personal agency was associated with higher BPD symptoms, more insecure attachment, greater psychopathology, and pathological personality traits. Despite BPD symptoms significantly reducing as a result of the 12-week DBT program, personal agency remained stable over time. This suggests that DBT as delivered in this study may not be as effective in modifying agency. Increasing insight, autonomy, and personal responsibility of one’s emotions and behaviours may be hard to modify within a group skills-based treatment. Taken together, these studies put a spotlight on personal agency within BPD. The association found with severity of symptoms and insecure relational patterns suggests that modification of agency over time may be difficult and require additional focus on (a) relationship styles and assessing personal agency throughout treatment, (b) monitoring changes in attachment style and addressing relationship insecurities, (c) focusing on the therapeutic relationship to increase personal agency, and (d) practical strategies in targeting levels of personal agency. The results of these studies may advance research and inform therapeutic interventions for BPD by illustrating the importance of assessing and increasing personal agency.


Thesis by publication

FoR codes (2020)

520302 Clinical psychology



Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.