Dispositions, Context, and Antisocial Behavior 1
Running head: ADOLESCENT DISPOSITIONS
Adolescent Dispositions for Antisocial Behavior in Context: The Roles of Neighborhood
Dangerousness and Parental Knowledge
Christopher J. Trentacosta
Wayne State University
Luke W. Hyde, Daniel S. Shaw, and JeeWon Cheong
University of Pittsburgh
Dispositions, Context, and Antisocial Behavior 2
This study examined an ecological perspective on the development of antisocial behavior during adolescence, examining direct, additive, and interactive effects of child and both parenting and community factors in relation to youth problem behavior. To address this goal, early adolescent dispositional qualities were examined as predictors of boys' antisocial behavior within the context of parents' knowledge of adolescent activities and neighborhood dangerousness. Antisocial behavior was examined using a multi-method latent construct that included self-reported delinquency, symptoms of conduct disorder, and court petitions in a sample of 289 boys from lower socioeconomic status backgrounds who were followed longitudinally from early childhood through adolescence. Results demonstrated direct and additive findings for child prosociality, daring, and negative emotionality that were qualified by interactions between daring and neighborhood dangerousness, and between prosociality and parental knowledge. The findings have implications for preventive intervention approaches that address the interplay of dispositional and contextual factors to prevent delinquent behavior in adolescence.
Dispositions, Context, and Antisocial Behavior 3
Adolescent Dispositions for Antisocial Behavior in Context: The Roles of Neighborhood
Dangerousness and Parental Knowledge
Antisocial behavior (AB) in adolescence is predictive of numerous problems in adulthood including crime, mental health concerns, substance dependence, and work problems (Moffitt, Caspi, Harrington, & Milne, 2002). Due in part to the personal, economic, and social toll of AB, extensive attention has been directed toward elucidating factors that increase risk for engaging in AB during adolescence. For example, child attributes and contextual variables, including parenting and the broader family ecology, have received much support as factors in the emergence of conduct problems during middle childhood and the subsequent development of more serious AB in adolescence (Campbell, Shaw, & Gilliom, 2000; Dishion & Patterson, 2006; Lahey & Waldman, 2003). In line with a focus on child attributes and contextual mechanisms and the development of AB, the present study examined child dispositions, including sensation seeking, prosociality, and negative emotionality, and contextual factors, including parental knowledge of adolescent activities and neighborhood dangerousness, as predictors of AB from early to mid adolescence.
Furthermore, a transactional perspective suggests that deviations from normal behavior are not solely related to factors within the individual or context but rather interactions between child attributes and context (Sameroff, 2000; Sameroff & Mackenzie, 2003). Depending on the context, child attributes may serve as sources of either vulnerability or resilience in the development of psychopathology (Nigg, 2006). This interactive interplay is particularly salient for AB during the transition to adolescence because both time outside the home and the seriousness of AB increase during adolescence (Dodge, Coie, & Lynam, 2006). Child dispositions associated with AB, particularly those that may be linked to risky early child
Dispositions, Context, and Antisocial Behavior 4
temperament and/or later adult personality traits, may be exacerbated when a young adolescent lives in a relatively dangerous neighborhood or when parents have little knowledge of their adolescent's activities and whereabouts (Lynam et al., 2000). Therefore, the central goal of the present study was to examine contextual risk factors as moderators of relations between youth dispositions and antisocial outcomes in adolescence.
Child Characteristics and Risk of Antisocial Behavior
A vast amount of research has linked child dispositional traits, including early temperamental attributes and later personality traits, with an array of outcomes including psychopathology. Temperament theories emphasize early-appearing, relatively stable differences in children‟s behavioral styles and regulation of emotion in response to affectively significant stimuli (Wachs & Bates, 2001; Rothbart, Posner, & Hershey, 1995; Lahey, Waldman, & McBurnett, 1999). Theories of personality similarly emphasize relatively stable global differences in behavior and response to the environment (McCrae & Costa, 1997). While temperament contributes to later personality development and many temperament traits map onto adult theories of personality (Shiner & Caspi, 2003; Wachs & Bates, 2001), most research on temperament has been confined to infancy and childhood. However, temperamental attributes also have been studied during adolescence and adulthood (John, Caspi, Robins, Moffitt, & Stouthamer-Loeber, 1994; Rothbart, Ahadi, & Evans, 2000). Dispositional characteristics during late childhood and adolescence should play a role in the development of psychopathology, but research has been lacking that links dispositions during this key transitional period to increased risk for later AB. Note that the term “disposition” will be used to refer to broad and general trait-
like characteristics during late childhood and early adolescence that likely reflect earlier temperamental traits and/or later personality.
Dispositions, Context, and Antisocial Behavior 5
Beginning with examinations of difficult temperament as a predictor of early behavior problems (e.g., Bates, Maslin, & Frankel, 1985) and dimensions of personality as predictors of crime and delinquency (see Eysenck, 1996), extensive research has supported an association between a wide range of dispositional characteristics and externalizing problems. For example, fearlessness observed at age 2 predicted trajectories of elevated conduct problems across early and middle childhood in models controlling for other factors including parenting, maternal depression, child IQ, and family demographics (Shaw, Gilliom, Ingolsby, & Nagin, 2003). Given empirical support for the role of temperament and personality in the development and maintenance of AB, multiple theories addressing externalizing problems across the lifespan highlight temperament-related constructs as key risk factors. For example, Eisenberg and colleagues (e.g., Eisenberg, Hofer, & Vaughan, 2007) suggest that undercontrolled (externalizing) behavior is predicted by low levels of effortful control, high levels of impulsivity, and high levels of negative emotionality, as demonstrated by their longitudinal research (e.g., Eisenberg et al., 2001). Frick and Morris (2004; see also Nigg, 2006) suggest that developmental precursors to callous-unemotional traits including fearlessness and impaired conscience development may have unique associations with covert AB and psychopathy in late childhood and adolescence.
A recently proposed developmental propensity model of AB examines dispositions that fit with the theoretical perspectives described above (Lahey & Waldman, 2003). This propensity model is an attempt to integrate older propensity theories (Hirshi, 1969; Gottfredson & Hirshi, 1990; Farrington, 1995) with recent theory and empirical research on temperament, personality, and developmental theories of AB. Research on this model has yielded three factors: daring, negative emotionality, and prosociality (Lahey et al., 2008). Daring was central to Farrington and West's (1993) description of predictors of crime and encompasses high-energy activities and
Dispositions, Context, and Antisocial Behavior 6
other risk-taking opportunities that are theoretically related to traits such as sensation seeking and novelty seeking. Furthermore, daring may be inversely related to Kagan's temperamental dimension of behavioral inhibition (see Kagan, Reznick, & Snidman, 1988). Negative emotionality is the second factor in Lahey and colleagues‟ model and is proposed to relate to the
Big Five factor of neuroticism, as reflected in frequent and intense experience of negative emotions. Numerous forms of psychopathology are associated with high levels of negative emotionality (e.g., Clark & Watson, 1991). Furthermore, oppositional traits including irritability and defiance are indicative of high levels of negative emotionality and frequently precede more serious forms of AB (Lahey, Waldman, & McBurnett, 1999). Prosociality is the third dimension and includes empathy, dispositional sympathy, respect for rules, and guilt in response to misdeeds. Conceptually, prosociality shares some common ground with the Five-Factor Model's (McCrae & Costa, 1997) dimension of agreeableness. High levels of empathy, guilt in response to misdeeds, and other aspects of agreeableness should protect children against involvement in antisocial activities. Furthermore, prosociality is inversely related to callous/unemotional traits (Lahey & Waldman, 2003). Both callous/unemotional traits and low levels of prosociality show a similar pattern of modest yet reliable correlations with AB suggesting that prosociality and AB are similar but not synonymous constructs (Barry et al., 2000; Lahey et al., 2008).
All three factors are related to AB and associated problems in concurrent and prospective longitudinal studies (see Lahey & Waldman, 2003 for a review). For example, constructs associated with the daring dimension, such as sensation seeking, are correlated with conduct problems (e.g., Arnett, 1996; Daderman, 1999). Negative emotionality predicts AB and other forms of psychopathology in adults (e.g., Krueger, 1999), as well as externalizing problems in children and adolescents (e.g., Eisenberg, Fabes et al. 1996), although other studies have failed to
Dispositions, Context, and Antisocial Behavior 7
find a robust association during adolescence (e.g., John et al., 1994). Negative emotionality has also been linked to internalizing problems (Eisenberg et al., 2001) and may be particularly salient in the development of comorbid pathologies. Lastly, a low level of prosociality is associated with AB across childhood and adolescence (e.g., Haemaelaeinen & Pulkkinen, 1996). Given past research linking both temperament and personality to AB and the role these dispositions should theoretically play in the development of AB during adolescence, a first goal of the present study was to examine these characteristics in early adolescence as direct and additive predictors of later AB in an ethnically diverse, at-risk sample.
Moderators of the Disposition-Antisocial Behavior Link
Although there are robust links between dispositional traits at various ages and later adjustment, the field has begun to examine the more complex moderated link between dispositions (particularly early temperament) and psychopathology (Nigg, 2006). As a result, a second goal of the present study was to test contextual factors as moderators of relations between daring, negative emotionality, and prosociality, and AB from early to middle adolescence. One key implication of a transactional approach is that risky dispositions should lead to higher levels of AB in riskier contexts (Sameroff, 2000). In examining a transactional perspective on the prediction of AB in adolescence, parental knowledge of adolescent activities and neighborhood risk are two key contextual factors that may interact with adolescent dispositional traits in predicting AB.
Parenting has held a central place in multiple developmental models of antisocial behavior (Aguilar, Sroufe, Egeland, & Carlson, 2000; Greenberg & Speltz, 1988; Patterson, Reid, & Dishion, 1992; Shaw & Bell, 1993), and parental behavior has been consistently related to early conduct problems and later AB (Owens & Shaw, 2003; Shaw et al., 2003). In early
Dispositions, Context, and Antisocial Behavior 8
adolescence, parental knowledge of the adolescent‟s activities and whereabouts becomes
increasingly important (Dishion & McMahon, 1998). Previously, the term “parental monitoring”
was equated with parental knowledge. Active parental monitoring attempts are not necessarily helpful, and adolescent disclosure of activities is a more robust correlate of adolescent behavior (Stattin & Kerr, 2000). Furthermore, adolescents may have a more accurate representation of their parent's knowledge (Laird, Pettit, Bates, & Dodge, 2003), and parents may overestimate their actual level of knowledge (Smetana, 2008). As a result, the present study examined adolescent‟s reports of their parent‟s knowledge of activities rather than active monitoring
Because parental knowledge is a robust contextual factor associated with adolescent involvement in delinquent activities (e.g., Laird et al., 2003), it is a likely candidate to moderate effects of risky dispositions on AB. Adolescents with predispositions to daring behaviors whose parents have little knowledge of their activities may have greater opportunities to engage in AB. When parents are not as aware of their adolescent's whereabouts and do not channel their adolescent's interests into prosocial expressions of their propensity toward sensation seeking,
daring adolescents may spend time with deviant peers and involve themselves in high-excitement antisocial behaviors. In the scenario of low parental knowledge and low levels of prosociality, an adolescent's unempathic and callous traits would be more likely to be channeled into delinquent activities if parents are not consistently aware of their adolescent's choices regarding peer relationships and use of free time. Alternately, when parents and adolescents openly discuss information on the adolescent's whereabouts, parental involvement in their adolescent's daily life may reduce the negative consequences of low levels of prosociality. For example, parents' awareness of their young adolescent‟s activities might lead to discussion of the impact of AB on
Dispositions, Context, and Antisocial Behavior 9
others following initial minor transgressions and increased attention to the adolescent's choices, thus preventing the development of future, more serious AB. Although somewhat speculative, it is also possible that more frequent parent-child communication of activities may reduce the likelihood of involvement in high-risk situations when adolescents have a tendency toward irritability and other intense negative emotions that could lead to serious conflict or violence.
Neighborhood dangerousness is a second contextual factor that has the potential to moderate the relation between dispositional traits and AB. Neighborhood environments characterized by crime and exposure to deviant peers may be powerful negative learning environments during childhood, and neighborhood risk often predicts increased AB in adolescence (Leventhal & Brooks-Gunn, 2000). For example, exposure to community violence was associated with later AB among school-age children and adolescents (McCabe, et. al., 2005; Ng-Mak, Salzinger, Feldman, & Stueve, 2004). Moreover, the association between exposure to community violence and antisocial outcomes remains after accounting for the effects of other confounding variables such as child maltreatment, socioeconomic status, and intimate partner violence (McCabe et al., 2005).
Adolescents with risky dispositions who live in dangerous neighborhoods may be especially likely to engage in AB across adolescence. For example, in two separate reports, trait impulsivity was more strongly related to delinquency in poorer or lower quality neighborhoods (Lynam et al., 2000; Meier, Slutske, Arndt, & Cadoret, 2008). Similar findings exist for the relations among neighborhood quality, temperamental traits and AB in early childhood such that low neighborhood quality was associated with increased AB in the context of temperamental risk (Colder, Lengua, Fite, Mott, & Bush, 2006). We hypothesized a similar interaction in the present study, using parental report of familial exposure to neighborhood violence as the moderator. For
Dispositions, Context, and Antisocial Behavior 10
example, daring adolescents in high-risk neighborhoods may find an outlet for their high-excitement interests by becoming involved in gangs or other criminal activities, whereas daring adolescents in low-risk neighborhoods may have more opportunities to channel their energy into prosocial activities. Furthermore, adolescents low in prosociality may be more likely to engage in conflicts with peers or neighbors in areas where crime is more rampant, and their lack of guilt over misdeeds may lead to more serious AB in high-risk environments that reinforce delinquent behavior. Conversely, in low-risk neighborhoods the peer group culture may disapprove of callous delinquent behaviors, thus ameliorating the likelihood of peer deviancy training that might exist for adolescents in high-risk neighborhoods. Although speculative, deviant peers in a high-risk neighborhood may more likely encourage an adolescent's propensity toward negative emotions to be channeled into delinquent behaviors such as fighting and bullying. In contrast, low-risk neighborhoods offer comparatively fewer opportunities for involvement in delinquent activities as an outlet for expression of negative emotion.
The Current Study
A central goal of the present study was to examine risky dispositions measured in early adolescence as predictors of AB later in adolescence. A latent adolescent AB construct was created from court records, symptom counts of conduct disorder from a psychiatric interview, and youth reports of delinquency. Risky dispositions of negative emotionality, prosociality, and daring were expected to have direct and independent relationships with AB, and we also considered these relations between dispositions and AB while accounting for the youth‟s
previous history of AB across middle childhood. The present study also examined interactions between child and contextual factors, exploring whether parental knowledge of adolescent's activities and neighborhood dangerousness moderated associations between dispositional