Training Track: Clinical
Lab: PACCT Lab
Advisor: Noni Gaylord-Harden, Ph.D.
Office: Coffey Hall 248
youth mental health, coping with traumatic events and stress, involvement in the juvenile justice system, and the implementation of evidence based services for underserved children and families
Masters Thesis Title
The Longitundinal Relationship Between Callous-Unemotional
Traits and Exposure to Community Violence: Examining Primary and Secondary Psychopathy in Serious Adolescent Offenders
Masters Thesis Abstract
Background: Psychopathic traits (e.g. lack of remorse, superficial charm, and antisocial behaviors) are more common in juvenile justice populations compared to community samples (Castellana et al., 2014). Two variants of psychopathy have been proposed: primary psychopathy describes individuals with an inherent inability to feel empathy, while secondary psychopathy describes traits that develop in response to environmental causes (Karpman 1941, 1948). In youth samples, psychopathy is often represented by callous-unemotional (CU) traits, a dimension of psychopathy characterized by a lack of guilt and empathy. In youth, high levels of CU traits and low levels of anxiety are proposed to represent primary psychopathy, whereas high CU traits and high anxiety represent secondary psychopathy (Sharf et al., 2014). For secondary variants, CU traits may result from exposure to traumatic events (Kerig, et al., 2012). Community violence exposure (ECV), especially common in justice-involved youth (Kerig, et al., 2009), may be one such traumatic experience leading to high CU traits. Previous research has made connections between CU traits and ECV (Howard, et al., 2012), but there remains a need to better understand how traumatic experiences and CU traits may change and influence one another over time in serious adolescent offenders (Bennett, 2013).
Methods: Data from the Pathways to Desistance study (Mulvey, 2004) were used to examine primary and secondary pathways between ECV and CU traits. Participants were 1,170 male adolescent offenders ages 14-20 (M =16.56, SD = 1.16 at Time 1), convicted of a serious offense prior to adulthood. Participants completed the Exposure to Violence Inventory (ETV; Selner-O’Hagan et al., 1998), the Youth Psychopathic Traits Inventory (YPI, Andershed et al., 2002), and the Brief Symptom Inventory (BSI, Derogatis & Melisaratos, 1983) at Times 1, 2, 3 and 4.
Results: Results of the latent profile analysis revealed that a 3-profile model provided superior fit for the data when compared to 1-, 2-, and 4-profile models (LMRT = 598.12, p = .03; BIC values 2-class = 10183.49, 3-class = 9611.03). The majority of the sample (N = 702, 83% of sample) fell in a “Low Distress” group, with uniformly lower scores on measures of distress. A “Medium Distress” group was the second largest (N = 118, 14% of sample) and showed mid-range scores in psychological distress. Finally, a “High Distress” group only consisted of 26 participants (3% of sample) and had the highest scores on the BSI subscales. An initial cross-lag panel model showed significant associations between ECV and CU traits over time. Separate cross-lag panel models were conducted for each of the two larger profiles. For the Low Distress group, only one significant cross path was found in which Time 2 callous-unemotional traits predicted Time 3 exposure to community violence (B = .03, p < .05). Additionally, within-variable paths and within-timepoint paths were statistically significant for this Low Distress group. For the Medium Distress group, which was comprised of 118 participants, there was also one significant cross path in which exposure to community violence at Time 2 predicted callous-unemotional traits at Time 3 (B = .81, p < .05). In this model, exposure to community violence and callous-unemotional traits predicted future levels of the each respectively, but exposure to community violence and callous-unemotional traits were not significantly associated within timepoints.
Conclusion: Results of the current study highlight a longitudinal relationship between CU traits and ECV. Results for the current study reflect previous research showing differences in psychological distress may predict directional influences in the relationship between ECV and CU traits. Further research with on this topic may inspire future directions for timing of and approaches to intervention.
Masters Thesis Committee
Noni Gaylord-Harden, PhD (Chair) & Dr. James Garbarino PhD