Predictors of Reactive and Instrumental Aggression in Jail Detainees: An Initial Examination

Abstract

Antisocial personality disorder (APD) and psychopathy have been extensively researched in relationship to criminal behavior and aggression. Comparatively few studies have evaluated these clinical constructs for subtypes of aggression, specifically reactive and instrumental aggression. Moreover, published studies tend to focus on extremes, either prisoners in maximum-security facilities or college samples. The current study utilized jail detainees on a minimum-security unit to retrospectively examine their self-reported reactive and instrumental aggression. Because conduct disorder (CD) is required for APD, the combined APD-CD symptomatology was used to predict subtypes of aggression. As a comparative analysis, facet scores on the Psychopathy Checklist-Revised were also used as predictors. Results of the current study have implications for forensic practice by utilizing aggression as a multi-dimensional construct in conjunction with personality traits (e.g., facets of psychopathy) in risk assessments.

Keywords:

• aggression
• reactive
• instrumental
• psychopathy
• antisocial

Notes

¹For reactive aggression, PCL-R Facets 1 and 2 were significant predictors when entered first (R² = .09). Adding APD-CD as a second step significantly improved the entire model (ΔR² = .19); however, Factor 1 no longer contributed significantly to the model.

²As a follow-up, the same analysis was conducted, restricting participants to include only the upper quartile of PCL-R Factor 1 scorers (n = 27). This procedure significantly increased the average PCL-R total score to 20.99 (SD = 6.27). However, this restriction did not significantly improve PCL-R Factor 1 R² for instrumental aggression (ΔR² = .01).

³For instrumental aggression, PCL-R Facets 3 and 4 were significant predictors when entered first (R² = .45). When APD-CD severity was added as a second step, it improved the overall model (ΔR² = .10), but only Facet 4 remained a significant predictor.

⁴This comparison was at the 40% (APD) standard; the 75% (PCL-R) standard could not be tested because too few participants reached this level with APD severity.

⁵A much larger sample would be needed to examine the PCL-R facets and APD-CD components.

⁶The percentage of truly innocent offenders in jail settings—awaiting trial or wrongly convicted—is unknown.