Direct and Indirect Effects from Aggression to Internalizing Symptoms

• In Event: Poster Session 03
  In Poster Session: PS 03 Topic - Developmental Psychopathology

Abstract

The failure model (Patterson and Capaldi (1990) is an influential conceptual model that posits a longitudinal link from aggression to internalizing symptoms, via peer rejection. Children who exhibit aggression are thought to experience psychosocial failures that contribute to the development of internalizing symptoms. However, empirical support for the model remains inconclusive because it has not been subject to a longitudinal test of associations (Gooren et al., 2011; Kiesner, 2002). Further, although the model examines associations between variables known to be genetically influenced (e.g., Boivin et al., 201), it has not been tested in a genetically controlled design to isolate nonshared environmental effects. The current study is the first to describe longitudinal associations between reactive aggression and internalizing symptoms, as mediated by peer rejection, independent of genetic and family-wide (i.e., shared) environmental effects.
Participants included 257 monozygotic twin pairs (128 male twin pairs, 129 female twin pairs) obtained from the Quebec Newborn Twin Study (Boivin, et al., 2013). Data were collected when children were in Kindergarten (M=6.04 years, SD=0.27), Grade 1 (M=7.07 years, SD = 0.27), and Grade 4 (M=10.00 years, SD= 0.28). At each wave of data collection, teachers reported on children’s reactive aggression and internalizing symptoms. Peer rejection was assessed through peer nominations.
A series of full longitudinal mediational models (Fritz & MacKinnon, 2012) examined the indirect effect of peer rejection in the association between reactive aggression and internalizing symptoms. Analyses were conducted twice, with traditional analyses (using one member of the twin pair) and again with a genetically controlled design (using the twin difference method, see Vitaro et al., 2012), to examine whether indirect effects remain after removing influence from genetic influences.
Results from traditional analyses revealed significant longitudinal associations between reactive aggression and peer rejection. Higher levels of reactive aggression predicted increases in peer rejection. Higher levels of peer rejection predicted increases in reactive aggression. There was also an association between peer rejection and internalizing symptoms. Higher levels of peer rejection predicted increases in internalizing symptoms, but not the reverse. Results from genetically controlled analyses failed to reveal statistically significant direct or indirect effects. There were no statistically significant associations between peer rejection and either reactive aggression or internalizing symptoms. Neither was there an indirect effect of peer rejection on the association between reactive aggression and internalizing symptoms. The results indicate that shared genetic influences are responsible for all associations identified in the traditional analyses.

Together, the findings suggest that the causal paths previously identified as support for the failure model may be, in large part, an artifact of common genetic factors and gene-environment interplay. Underlying genetic factors behind reactive aggression may evoke maltreatment from peers, which ultimately, may contribute to a rise in internalizing symptoms. The findings suggest that the mechanisms underlying the failure model may have a genetic origin during childhood.

Authors

• Olivia Valdes, Florida Atlantic University
  Presenting Author

• Sharon Faur, Florida Atlantic University
  Non-Presenting Author

• Frank Vitaro, Université de Montréal
  Non-Presenting Author

• Mara Brendgen, L'Université du Québec à Montréal (UQAM)
  Non-Presenting Author

• Michel Boivin, Universite Laval
  Non-Presenting Author

• Ginette Dionne, Universite Laval
  Non-Presenting Author

• Brett Laursen, Florida Atlantic University
  Non-Presenting Author