Do insecure adult attachment styles mediate the relationship between childhood maltreatment and violent behavior?

Abstract

Attachment theory has played an important role in attempts to understand the "cycle of violence," where maltreated children are at increased risk for perpetrating violence later in life. However, little is known empirically about whether adult attachment insecurity in close relationships may partly explain the link between childhood maltreatment and violent behavior. This study aimed to address this gap using data from a prospective longitudinal study of documented childhood abuse and neglect cases and demographically matched controls (ages 0-11 years), who were followed into adulthood and interviewed (N = 892). Participants completed the Relationship Scales Questionnaire assessing adult attachment styles at mean age 39.54. Criminal arrest data were used to determine arrests for violence after the assessment of attachment through mean age 50.54. There were significant direct paths from childhood maltreatment and adult attachment insecurity to violent arrests after attachment measurement. Attachment insecurity partly explained the higher levels of violence in individuals with maltreatment histories. Analyses of maltreatment subtypes and attachment styles revealed that attachment anxiety appeared to mediate paths between neglect and physical abuse and later violence. There were no significant indirect paths from neglect or physical abuse to violence via attachment avoidance. Implications and future directions are discussed.

Keywords: adult attachment styles; child maltreatment; neglect; physical abuse; prospective longitudinal; violence; violent offending.

Figure 1.

Structural equation model showing paths from maltreatment to arrests for violence through adult attachment insecurity. Note. Standardized betas are presented. All lines are solid, reflecting significant paths. Analyses control for age, sex, and race. Model fit indices: RMSEA = 0.02, CFI/TLI = 0.99/0.97. R² = 0.05 (attachment insecurity) and 0.22 (arrests for violence). *p < .05, * *p < .01, ***p < .001.

Figure 2.

Structural equation model showing paths from childhood neglect and physical abuse to arrests for violence through adult attachment anxiety. Note. Standardized betas are presented. Solid lines are significant paths, dotted lines are not significant. Analyses control for age, sex, and race. Model fit indices: RMSEA = 0.05, CFI/TLI = 0.89/0.88. R² = 0.03 (attachment anxiety) and 0.21 (arrests for violence). *p < .05, **p < .01, ***p < .001.

Figure 3.

Structural equation model showing paths from childhood neglect and physical abuse to arrests for violence through adult attachment avoidance. Note. Standardized betas are presented. Solid lines are significant paths, dotted lines are not significant. Analyses control for age, sex, and race. Model fit indices: RMSEA = 0.05, CFI/TLI = 0.86/0.80. R² = 0.03 (attachment avoidance) and 0.20 (arrests for violence). *p < .05, **p < .01, ***p < .001.