«by Amy Lynn Byrd, Ph.D. B.S. in Psychology, College of Charleston, 2006 M.S. in Clinical Psychology, University of Pittsburgh, 2010 Submitted to the ...»
Importantly, the current dissertation differs from past research in three important ways. First, it utilizes a basic paradigm that isolates receipt of reward and receipt of punishment, allowing for the separate characterization of neural response following monetary cues of reward and punishment. Second, it proposes the examination of the BOLD response in pre-adolescent youth, prior to noted maturation of the neural circuits that are of primary interest. Lastly, it examines responsivity to reward and punishment among subgroups of CP youth (with and without CU;
with and without psychopathic features) relative to HC and represents the first imaging study to examine reward and punishment processing among these subgroups.
The extent to which differential responsivity to reward and punishment has meaningful implications that extend beyond our understanding of etiology is explored in the next section.
Specifically, research has suggested that aberrant reward/punishment processing may not only impact the development and persistence of CP but could also influence responsiveness to intervention. The following section explores this notion and provides an overview of the intervention literature among youth with CP, focusing on the effectiveness of the multi-modal empirically-validated intervention that was used in the current dissertation.
Better characterizing potential abnormalities in the responsivity to reward and punishment may not only help us to better understand the development of CP but could also be particularly important with regard to intervention. Noteworthy, the majority of interventions targeting CP are grounded in social learning theory and place strong emphasis on behavioral strategies associated with reward and punishment contingencies. While burgeoning empirical research suggests that intensive, empirical-validated interventions function to reduce these problem behaviors over time (Eyberg, Nelson, & Boggs, 2008; Hawes & Dadds, 2007; Kolko, et al., 2009; Kolko & Pardini, 2010; McDonald, Dodson, Rosenfield, & Jouriles, 2011; Somech & Elizur, 2012; White, Frick, Lawing, & Bauer, in press), it is also well-documented that interventions for CP youth are not effective for everyone (Hawes, Price, & Dadds, 2014; Matthys, et al., 2012b; Webster-Stratton & Hammond, 1997).
Along these lines, researchers have investigated myriad moderators that may undermine the effectiveness of interventions, the majority of which have focused on demographic risk factors (e.g., poverty), parental psychopathology (e.g., depression) and other related stressors (family conflict; for review see Shelleby & Shaw, 2013). There is also some suggestion that child factors may influence the effectiveness of intervention (Matthys, et al., 2012b), with more recent focus on the presence of CU traits or psychopathic features (Hawes, et al., 2014). As described above, these features are believed to demarcate a more homogenous subgroup of youth that are most severe and seemingly resistant to traditional interventions (Frick, Ray, Thornton, & Kahn, 2013; Pardini & Frick, 2013). Despite noted clinical pessimism, research in this area remains mixed, with emerging studies suggesting that intensive, multimodal interventions may
Hyde, 2013). Nonetheless, treatment resistance within this population has been linked to abnormalities in reward and punishment processing (Dadds & Salmon, 2003; Matthys, et al., 2012b). Along these lines, research suggests that youth with increased reward seeking and punishment insensitivity may respond well to the reward-based components of parent training (e.g., praise, token reinforcement) yet appear insensitive to the disciplinary components of treatment (e.g., time-out). For example, treatment studies have found CP youth with CU traits to respond significantly worse to punishment focused techniques (e.g., "time-out"; Haas, et al., 2011; Miller, et al., 2014) while reward-oriented strategies have been found to work equally well across CP youth with high and low levels of CU traits (Hawes & Dadds, 2005). While this implies that individual differences in reward and punishment processing may influence responsiveness to intervention, this notion has yet to be empirically tested.
2.7.1 Overview of Treatment Literature Multifaceted, multimodal intervention efforts appear to be particularly effective for children exhibiting heightened levels of CP. Specifically, research has suggested that child-focused CBT and skills training coupled with parent management training (PMT) serves to reduce CP in young children over time (Eyberg, et al., 2008; Kazdin, Siegel, & Bass, 1992; Webster-Stratton, Reid, & Hammond, 2004). Child-focused components rely heavily on CBT principles and are designed to enhance problem-solving techniques. Specifically, these interventions teach strategies that aim to interrupt impulsive and perseverative, reward-focused action by encouraging youth to think about the consequences of their behaviors and develop a socially appropriate plan of action. At the same time, prominent PMT interventions, strongly founded in behaviorist principles, target a
inconsistent and harsh punishment, which inadvertently increases the frequency of these behaviors over time while simultaneously failing to reinforce prosocial behaviors (Patterson, DeBaryshe, & Ramsey, 1989; Patterson, et al., 1992). PMT breaks this coercive cycle by shifting behavioral contingencies and teaching parents to positively reinforce prosocial behaviors, while setting clear and consistent consequences for disruptive behaviors (Serketich & Dumas, 1996).
Meta-analytic reviews examining the effectiveness of each of the aforementioned intervention modalities in quality coded, randomized control trials (RCT) have demonstrated promising effects. For example, an examination of 29 child-focused interventions aimed at increasing self-control over impulsive, reward-seeking behaviors in CP youth showed significant reductions in subsequent delinquency (Piquero, Jennings, & Farrington, 2010). This review focused on early interventions for youth aged 10 and below that utilized primarily CBT and skills training. While these studies only examined outcomes up to age 12, they reported increased self-control in CP youth across informants (i.e., teacher-report, self-report, direct observation; ES range=.28-.61) and more importantly a reduction in CP over time, though this was specific to teacher-reported behaviors (ES=.30). At the same time, meta-analytic reviews of intervention efforts focused primarily on PMT have also demonstrated significant reductions in CP over time (Farrington & Welsh, 2003; Serketich & Dumas, 1996). More recently, Lundahl and colleagues (2006) conducted an extensive meta-analysis examining 63 peer-reviewed studies and found behaviorally focused PMT interventions to show decreases in CP both immediately following the intervention (ES=.47) as well as at later follow-up (ES=.25), though these effects are notably within the small to moderate range. Taken as a whole, the literature suggests that these types of early interventions may be particularly beneficial for at risk youth.
and parent behavioral training modalities to produce greater changes in CP over time relative to the implementation of a single modality in isolation (Burke, Loeber, & Birmaher, 2002; Kazdin, et al., 1992; Webster-Stratton & Hammond, 1997), with some suggestion that a multimodal approach produces more sustained treatment gains over time (Kazdin, et al., 1992; WebsterStratton, et al., 2004). Along these lines, a variety of early intervention programs designed to include both child and parenting components have been widely disseminated (e.g., Problem Solving Skills Training/PMT; The Incredible Years; Coping Power) and demonstrate fairly consistent success, showing reductions in problem behavior over time (e.g., Eyberg, et al., 2008;
Kazdin, 2010). The current dissertation focuses on a similar multimodal early intervention designed to reduce CP in at-risk youth. As is detailed below, the Stop-Now-And-Plan (SNAP) Under 12 Outreach Project is an empirically supported, manualized program that incorporates child-focused CBT and skills training as well as PMT.
2.7.2 Stop-Now-And-Plan Intervention SNAP was created in 1985 in Toronto, Canada and targets children under the age of 12 who have had previous contact with the juvenile justice system or are at serious risk for police contact. The SNAP program takes a multimodal approach, comprised of two primary components, each of which is developed with gender-specific programming. The first is child-focused and teaches children cognitive-behavioral self-control skills and problem-solving techniques within a group setting. Youth are taught strategies (e.g., deep breathing, counting to 10) that aid in interruption of potential reward-focused actions, helping them to “STOP” and think about the positive and negative consequences of their behaviors. Next, they are taught to “PLAN” effective solutions to
themselves and others safe. In addition, groups are designed to provide youth with structured ‘practice experiences’ during which they are able to apply their skills under a variety of different circumstances (e.g., stealing, anger management, peer pressure). The second component is focused on teaching parents effective child management strategies. Specifically, parents are taught behavioral strategies that focus on consistent reward and punishment implementation designed to improve the quality and consistency of their response to negative (e.g., defiance) and positive (e.g., compliance) child behavior. Groups include of modeling, behavioral rehearsal/role plays and parents are given home practice exercises. Parents are also informed of the cognitive and behavioral self-control techniques that their children are learning. Both of these core components are offered simultaneously within a weekly 90-minute group setting for 12 consecutive weeks. Youth and their families are also offered additional components based on need and preference. These include family therapy, individual befriending designed to connect youth with positive structured activities in their community, and academic tutoring for youth who are not performing at their grade level.
Initial studies evaluating the effectiveness of SNAP found that youth who attended the program had a significant reduction in CP from pre to post treatment that was maintained at 6and 12-month follow-ups (Hrynkiw-Augimeri, Pepler, & Goldberg, 1993). In addition, more than half of youth enrolled in SNAP successfully refrained from future contact with the criminal justice system for up to 11 years post-treatment (Day, 1998). Bolstering early work, more rigorous examinations of the SNAP program have utilized RCT or matched control designs and allow for stronger conclusions about the overall effectiveness of the program. Augimeri and colleagues (2007) found 16 youth who were randomly assigned to SNAP to exhibit reductions
activities. Significant reductions in CP were evident post-treatment and gains were maintained at 6-, 12- and 18-month follow-ups. Expanding on this study, an additional 50 SNAP participants matched on age, sex and level of CP were included in analyses and compared to the original groups examined by Augimeri and colleagues (2007). Results indicated reductions in delinquency, major aggression, and minor aggression for youth completing SNAP relative to controls (Koegl, Farrington, Augimeri, & Day, 2008). Moreover, youth participating in SNAP who attended 9 or more group sessions had significantly fewer convictions up to age 18 (36% vs.
68%) relative to those who attended fewer than 8 sessions. In a larger sample, youth enrolled in SNAP (n=223) demonstrated significantly greater decreases in parent-reported problem behaviors, including rule-breaking, aggression and conduct problems when assessed 6-months post-treatment (Lipman, et al., 2008).
More recently, Burke and Loeber (2014) evaluated the effectiveness of SNAP in the first RCT to date (n=252). This study is particularly relevant because participants in the current dissertation were recruited from this sample. Results indicated that children enrolled in SNAP evidenced a greater reduction in conduct problems, aggressive behavior and overall externalizing behavior when compared to youth receiving treatment as usual (TAU). Immediately following intervention, children receiving SNAP had average scores below the ‘clinical’ cutoff (T-score 70 on the CBC-L) and these effects were maintained across three follow-up assessments up to 15-months from baseline. Effect sizes were small to moderate (Cohen’s d=.25-.31) and remained even after controlling for age, race, income, IQ and prior police contact.
Taken together, these results provide support for the effectiveness of this program for children at-risk for persistent CP. However, what remains unclear is how individual differences
there has been some suggestion that aberrant reward/punishment processing underlies treatment resistance in a subgroup of youth with CP (e.g., Hawes & Dadds, 2005), it is also possible that emphasis on consistent reward/punishment contingencies could serve to modify problem behaviors for these youth who are at-risk for more chronic forms of CP. A better understanding of how abnormalities in reward/punishment processing may impact the success of treatment (in either direction) would be particularly important with regard to further elucidating additional moderating mechanisms. Moreover, this knowledge could aid in the tailoring of intervention techniques, perhaps at a more individual level, to achieve optimal effectiveness (Dadds & Salmon, 2003). As such, the current dissertation takes an exploratory approach and seeks to investigate associations between baseline reward/punishment responsivity and post-treatment levels of CP as well as the extent to which potential deficits moderate treatment effectiveness.