Using Biosocial Criminology to Understand and Improve Treatment Outcomes

Biological Outcomes of EBPs in Infancy and Early Childhood

Treatment and prevention programs are believed to affect antisocial behavior by reducing the number and severity of risk factors and/or increasing individuals’ protective factors. During infancy and early childhood, some risk factors for later antisocial behavior include prenatal substance use, birth complications, low birth weight, nutritional deficits, genetics, exposure to higher levels of testosterone, maternal stress, maternal depression, poor parent–child interactions, and parental involvement in criminal behavior (Moffitt, 2005; Raine, Brennan, & Mednick, 1997). It is important to note that many of these risk factors co-occur and are related to each other, so that a change in one risk factor may affect the emergence and exacerbation of other risk factors. For instance, stress and depression among pregnant mothers can lead to elevated cortisol activity in mothers, birth complications, brain damage in infants, and abnormal cortisol activity in infants (Diego et al., 2004; Ponirakis, Susman, & Stifter, 1998). Also, mothers who have high levels of stress and depression often differ from nondepressed mothers in their perceptions of their child’s behavior, their use of critical or negative statements, and the quality of their attachments to children (Webster-Stratton & Hammond, 1988), and youth reared in these types of homes are often at-risk for antisocial behavior across the life course (Campbell, Shaw, & Gilliom, 2000).

Given the relationship between the above early childhood risk factors and later antisocial behavior, a number of interventions have been developed that target such risk factors, such as prenatal stress reduction programs, the Nurse Family Partnership (Olds, Hill, & Rumsey, 1998), parent training programs, the Perry Preschool Project, and other enriched preschool models. Many of these programs have not examined how these interventions would affect children at a biopsychological level, but there is reason to believe that such interventions will have biological implications. Several prenatal and perinatal stress-reduction programs have been shown to reduce mothers’ stress level, as well as normalize the biological functioning of mothers and infants (Urizar & Muñoz, 2011; Wesley, 2006). Urizar and colleagues (2004) instructed 41 pregnant, low income Spanish-speaking mothers to “reduce or eliminate things that create stress in your life,” and found that this simple intervention resulted in a decrease in perceived stress, negative affect, depression symptoms, and abnormal cortisol levels among mothers. Field, Pickens, Prodromidis, and Malphurs (2000) randomly assigned adolescent depressed mothers into either a control condition or a treatment that included free day care, relaxation therapy, music mood induction, massage therapy, infant massage, mother–infant coaching, and vocational high school for mothers. At the 6- and 12-month follow-ups, the researchers found that mothers in the treatment condition had more frequent positive interaction and less negative interaction with their infants. The biochemical activity (dopamine, epinephrine, serotonin, cortisol) of both mothers and infants was also normalized during the follow-up period, and infants in the treatment condition had fewer pediatric complications and higher birth weights than infants of depressed mothers in the control group.

Other researchers have found that providing enriched foster care and parent training programs to parents can normalize the cortisol activity of foster care children. Fisher, Gunnar, Chamberlain, and Reid (2000) found that participation in the Early Intervention Foster Care (EIFC) program led to improvements in foster parents’ parental stress and use of discipline, monitoring, and positive reinforcement strategies that paralleled those of parents of healthy comparison youth. Foster EIFC children also showed normalizations in weekly basal cortisol activity, diurnal cortisol activity, and child behavioral problems relative to youth in regular foster care (RFC). Similarly, later studies by Fisher and colleagues (Fisher, Stoolmiller, Gunnar, & Burraston, 2007; Fisher, Van Ryzin, & Gunnar, 2011) found that a program focused on contingency management and attachment called Multidimensional Treatment Foster Care (MTFC) prevented the development of atypical diurnal cortisol activity among maltreated foster children ages 3 to 6. Youth in RFC had lower morning cortisol levels and less of a change in cortisol from morning to evening than MTFC and healthy comparison youth while in foster care, and RFC youth showed abnormal diurnal cortisol patterns when placed in another foster home or in a permanent home. Dozier et al.’s (2006) assessment of an attachment-focused program showed that youth in the intervention group had morning and evening basal cortisol levels that were similar to the nonfostered comparison youth at the 1-month follow-up, while control youth showed signs of elevated cortisol levels. Together, these studies suggest that enriched foster care programs can prevent the development of atypical biological and behavioral patterns, and perhaps bring youth biological and behavioral functioning into normal range.

In addition to foster care children, other at-risk groups may also benefit from enhanced parent management training (PMT) programs. A randomized control trial study of siblings (ages 33-66 months) of adjudicated youth found that a combination of PMT, youth social competence training, and home visits—a version of the Incredible Years Series—resulted in the normalization of cortisol reactivity in response to a social stressor among youth in the intervention group (Brotman et al., 2007). A follow-up study found that the Incredible Years Series intervention was effective in increasing cortisol responses to a stressor and in reducing observer-reported aggression among youth whose parents exhibited lower amounts of warmth (i.e., praising, verbally responding, physical affection) toward their children (O’Neal et al., 2010). Further, increased cortisol activity mediated 69% of effects of the intervention on child aggression; thus, the Incredible Years Series program was partially effective in reducing aggression because it normalized youth biopsychological functioning.

Finally, there is evidence that enriched nursery schools may improve youth biological and behavioral functioning. Raine and colleagues (2003; Raine et al., 2001) assessed the long-term outcomes of youths who completed a holistic, enriched preschool program in Mauritius. The program included a host of components that focused on nutrition, cognitive development, health and exercise, hygiene, social skills, and emotional development. The program had a low staff to student ratio (between 1/5.5 to 1/10) and included teacher visits to families’ homes, referral to social services, counseling sessions for parents, and parents’ support groups. They found that children who were randomly assigned to an enriched nursery school at ages 3 to 5 had increased physiological arousal activity at age 11 and lower levels of self-reported offending at age 26 compared with youth in the control group. In particular, youth in the treatment condition had increased skin conductance at rest, greater skin conductance reactivity (in terms of time and amplitude), faster skin conductance recovery after presentation of a provoking stimulus, and less slow wave (theta and delta) electroencephalography (EEG) activity at rest and during the Continuous Performance Task (CPT).

The above studies suggest that early intervention programs may affect antisocial behavior later in life because such programs affect the physiological arousal of participants. Low physiological arousal has been linked to higher levels of antisocial behavior across the life course (Jennings, Piquero, & Farrington, 2013; Ortiz & Raine, 2004; Portnoy, Chen, & Raine, 2013), and is often indexed by lower activity in the hypothalamic–pituitary–adrenal (HPA) axis (such as low cortisol levels), reduced startle reflex, increased slow waves (delta and theta) in EEG, lower resting heart rate, lower heart rate reactivity and variability, lower vagal tone, low respiratory sinus arrhythmia (RSA), and lower skin conductance activity and reactivity. One of the leading explanations for the association between low physiological arousal and antisocial behavior is the fearlessness hypothesis. The fearlessness hypothesis states that low physiological activity is linked with antisocial behavior because individuals who exhibit low physiological arousal report lower levels of fear (especially in response to punishment), experience less shame and guilt from engaging in antisocial behavior, underestimate the likelihood of punishment, and have difficulty learning the association between their behavior and punishment (Armstrong & Boutwell, 2012; van Goozen & Fairchild, 2008). Thus, prenatal/perinatal stress-management programs, enriched foster care programs, early PMT programs, and enriched preschool programs may help regulate the physiological systems related to stress and fear response of both mothers and children. In turn, children with normalized fear responses may be less likely to engage in antisocial behavior later in life.

Biological Moderators of EBPs in Infancy and Early Childhood

While EBPs may be powerful enough to impact individuals on both a psycho-physiological and behavioral level, it still must be recognized that some participants benefit from intervention efforts more so than other participants. For instance, Schweinhart’s (2004) review of the Perry Preschool Project showed that 28% of the preschool participants were sentenced to prison or jail by age 40, and 36% had been arrested 5 or more times. Similar results have been garnered by other EBPs in infancy and early childhood, with 15% to 30% of youth not responding to early interventions (Raine et al., 2003; Reynolds, Temple, Robertson, & Mann, 2001). One explanation for these null effects is that those who are not responding to treatment tend to be higher risk, and that even if they and their families complete the intervention, the treatment effects are masked or washed out by the concentration of risk factors. In contrast, an alternative explanation is that the interventions are effective for the highest risk individuals, and that low-risk individuals are harmed by the intensive nature of the interventions. This latter explanation is in line with the risk principle found in corrections (Andrews, Bonta, & Hoge, 1990), which expects positive treatment effects for high-risk respondents and iatrogenic effects for low-risk participants. Other perspectives, such as the differential susceptibility hypothesis (Belsky, Bakermans-Kranenburg, & Van IJzendoorn, 2007) and the vantage sensitivity hypothesis (Pluess & Belsky, 2013), anticipate that high risk individuals will benefit the most from interventions, but contrary to the risk principle, these perspectives expect that interventions will have no effect on lower risk participants. With these frameworks in mind, it is important to assess what types of individuals (i.e., low risk vs. high risk) benefit the most from EBPs in infancy and early childhood.

Risk has traditionally been conceptualized as psychological (i.e., difficult temperament) or sociological (i.e., low socioeconomic status), but it may also include biological risks such as carrying certain genetic variants, physiological underarousal, nutritional deficits, and deficits in brain functioning. It is likely that these biological factors may moderate the success of EBPs in infancy and early childhood (Brett et al., 2015). Indeed, a recent meta-analysis of 12 randomly controlled trials found that individuals’ genetic profile was related to the success of EBPs in reducing behavioral problems (Bakermans-Kranenburg & Van IJzendoorn, 2015). In particular, the authors found that interventions reduced externalizing behavior for youth who had “risk conferring” genetic variants related to the dopamine and serotonin systems, but EBPs did not affect externalizing behavior for youth who did not have the genetic risks (except see Brett et al., 2015). Previous studies had suggested that one of the reasons why EBPs may be effective with the genetically “at-risk” individuals is that these interventions normalize the physiological functioning or cortisol activity of children (Bakermans-Kranenburg, Van IJzendoorn, Mesman, Alink, & Juffer, 2008).

Other studies have also found that early interventions may have the greatest benefit for higher risk children. Bagner and colleagues (2012) examined whether low RSA—an indicator of low physiological arousal—conditioned the impact of the Parent–Child Interaction Therapy on disruptive behavioral problems among a sample of children (average age 3) who were born prematurely. The results showed that intervention children with low RSA had lower levels of parent-reported disruptive behavioral problems at the 4-month follow-up than intervention children with high RSA levels. Similarly, Raine et al. (2003) found that malnourished children in an enriched preschool program had greater improvements in interpersonal skills/deficits, schizotypal personality, cognitive disorganization, excessive motor movements, and conduct disorder in emerging adulthood (ages 17-23) than children who were not malnourished. The authors, however, did not find that skin conductance moderated the effects of the intervention on schizotypal personality or any of the behavioral problems, and malnourishment did not interact with the intervention to influence criminal behavior, thus suggesting that the moderation of treatment effects by physiological/biological factors may be more complicated than originally assumed.

Recommendations Based on EBPs in Infancy and Early Childhood

In sum, there is a growing body of research that suggests that biological factors may predict treatment response to EBPs in infancy and early childhood. The research indicates that individuals who are biologically “at-risk” may benefit more from early interventions than children who do not have biological risk factors. In particular, the interventions generally tended to benefit children with biological risk factors, and had little to no effect on biologically lower risk children. Theoretically, these results align with the differential susceptibility and vantage sensitivity hypotheses, and suggest that intervention efforts may have the greatest impact when they are concentrated on biologically higher risk children. It is important to note that the intensive interventions do not seem to be making biologically low-risk children worse but that we should expect lower treatment effect sizes when these interventions are applied to biologically lower risk groups. This is not to say that we should only target the highest risk children for early intervention. The majority of the above studies focused on at-risk families and children (i.e., premature infants, children with clinical levels of externalizing behavior), and while there is little to no change in antisocial outcomes for the biologically lower risk children, it is quite possible that these interventions are preventing the onset or escalation of antisocial behavior for at-risk children.

Furthermore, it is encouraging to see that prenatal stress programs, enriched preschool, and PMT programs grounded in attachment and behavioral principles can normalize the physiological functioning and behavior of both parents and infants. A large body of literature has shown that neglectful and abusive parenting can lead to low physiological arousal and alterations in the dopamine system within children, and that these biopsychological changes increase the risk of substance abuse later in life (Andersen & Teicher, 2009). Evidence that prenatal and early intervention programs affect these same biopsychological systems suggests that such interventions could potentially reverse the effects of parental risk factors, especially for the most biologically at-risk youth (i.e., those with low RSA/low physiological arousal and genetic variants correlating with low dopamine activity). Thus, it is recommended that we should continue to invest in early intervention programs for at-risk populations due to their long reaching effects, and to consider that youths may benefit the most from treatments that focus on parent–child attachment and sensitive discipline (i.e., reward based; Bakermans-Kranenburg et al., 2008).

Biological Outcomes of EBPs in Childhood

EBPs in childhood typically include PMT programs and CBT. These interventions have been shown to be effective at impacting a range of psychological and behavioral outcomes, such as conduct disorder, physical aggression, delinquency, and other externalizing behaviors (McCart, Priester, Davies, & Azen, 2006). In addition, PMT and CBT may have biological implications for children. Motamedi and colleagues (2008) examined the effectiveness of an 8-week PMT on cortisol levels and behavioral problems for 19 children, ages 8 to 12, who had been diagnosed with disruptive behavior disorder. The authors found that PMT was associated with a 25% increase in youth morning salivary cortisol levels, and that higher post-treatment cortisol levels were correlated with lower parent-reported disruptive behavior scores at the 2-month follow-up. Similarly, Dorn, Kolko, Shenk, Susman, and Bukstein (2011) found that boys with disruptive behavior disorder (DBD) experienced a significant increase in their mean levels of cortisol after completing a treatment that combined PMT, family therapy, school consultation, a peer intervention, and psychiatric consultation as needed. These effects, however, did not extend to females with DBD.

Other studies have assessed whether the combination of PMT with CBT would produce changes in brain activity and behavioral outcomes. Woltering, Granic, Lamm, and Lewis’s (2011) analysis of 71 children (ages 8-12) with clinical levels of externalizing behavior and 24 healthy controls found that completing a child-focused CBT and PMT led to the normalization of N2-event-related potentials (ERP) in the medial prefrontal cortex, the ventral prefrontal cortex (VPFC), and the anterior medial temporal lobe during the go/no go psychological task. The N2 ERP within the frontal cortex is believed to indicate self-regulation, suggesting that the combined PMT and CBT intervention led to a strengthening and an increased efficiency of youth self-regulatory capacities. The authors also found that changes in N2 values within the VPFC were correlated with clinical improvements in youth psychosocial functioning, and that changes in the N2 ERPs differentiated treatment improvers from non-improvers. A follow-up study from Woltering, Liao, Liu, and Granic (2015) found that the neural changes from PMT and CBT persisted for at least a year post-treatment, indicating that some of the changes may be long-lasting. The authors also found that internalizing and externalizing youth who benefitted from treatment showed greater efficiency in the neural networks underlying self-regulation than youth who did not improve from the program. In line with Woltering and colleagues’ work, a study of youth comorbid for internalizing and externalizing disorders also found that VPFC activity was normalized among treatment improvers after a 14-week combined PMT and child CBT program (Lewis et al., 2008), suggesting that such programs may lead to behavioral changes because they normalize functioning in brain regions that are associated with self-regulation.

Biological Moderators of EBPs in Childhood

CBT, PMT programs, and school-based interventions may have greater effects with some individuals than others. For instance, studies have shown that approximately 55% of youth experience significant improvements in behavioral problems after completing childhood EBPs, while the remaining children exhibit little to no response to treatment (Lewis et al., 2008; Shenk et al., 2012; Stadler et al., 2008). A growing body of research has found that genetic factors, hormones, and physiology may be important to explaining the differential effectiveness of childhood EBPs. In particular, genetic factors related to the reward (i.e., dopamine) and stress response (i.e., glucocorticoid) systems have been shown to enhance the effectiveness of childhood EBPs. A study of children ages 4 to 12 with attention-deficit hyperactivity disorder (ADHD) revealed that youth with 1 or more copies of the 9-repeat (9R) allele of the dopamine transporter gene (DAT1) benefited the most from a combination of PMT and routine clinical care (PMT + RCC), compared with youth with the 9R who just received RCC and youth with the 10-repeat allele who received PMT + RCC (van den Hoofdakker et al., 2012). The 9R allele has been associated with greater sensitivity to reinforcements or rewards, and thus, such youth may be especially responsive to the incentive components of PMT. Similarly, an evaluation of the Fast Track intervention—a longitudinal program with individual, school-based, and parent training components—showed that European American youth with the A allele of the NR3C1 glucocorticoid receptor gene were less likely to exhibit externalizing behavior problems at age 25 (18%) after completing the intervention, compared with control youth with the A allele (75%) and intervention youth without the A allele (57%; Albert, Belsky, Crowley, Latendresse, et al., 2015). A follow-up study showed that intervention youth with the A allele exhibited lower externalizing problems in adulthood because the intervention helped curtail childhood externalizing problems (Albert, Belsky, Crowley, Bates, et al., 2015). These studies suggest that treatment effectiveness was enhanced among individuals with these particular “risk” alleles, and that youth without these alleles experienced fewer benefits from treatment.

Other studies have shown that childhood EBPs may prevent the escalation of maladaptive behavior among youth who carry certain genetic variants. Albert, Belsky, Crowley, Latendresse, et al. (2015) found that Fast Track youth with the NR3C1 A allele showed less of an increase in alcohol and marijuana use during adolescence than intervention youth without the A allele. Similar results have been garnered from evaluations of the Strong African American Families (SAAF) intervention, a program targeting youth during late childhood (i.e., around age 11) and adolescence that focuses on skill building among youth and that includes a family curriculum. Results have shown that substance use significantly increased over a 2-year follow-up for control group youth with the 7-repeat allele of DRD4 (a dopamine receptor gene) and intervention youth without the 7-repeat allele, with little to no change in substance use for SAAF youth with the 7-repeat allele (Beach, Brody, Lei, & Philibert, 2010; Brody et al., 2014). The 7R allele has been associated with greater activity in the reward pathway upon presentation of drug-related stimuli (Goldman et al., 2013), and thus, the SAAF program may protect youth from developing substance use dependence disorders. Finally, Cleveland and colleagues (2015) evaluated the effectiveness of a multi-agency intervention that provided an in-home program (i.e., Strengthening Families Program: For Parents and Youth) during sixth grade and a school-based program during seventh grade in 14 communities. The authors found that the likelihood of alcohol use at ninth grade was approximately 2 times lower among intervention youth who carried the 7-repeat DRD4 allele and who had high levels of maternal involvement, compared with control youth with the 7-repeat allele and high levels of maternal involvement. These studies indicate that EBPs may prevent the escalation of some antisocial behaviors, especially among youth who are genetically susceptible to antisocial behavior.

While genetic studies show that higher risk youth benefit the most from childhood EBPs, studies of other biological markers—namely those related to stress response and to the autonomic nervous system—report that lower risk youth experience the greatest treatment gains from childhood EBPs while higher risk youth (i.e., those with low stress response and low autonomic nervous system activity and reactivity) receive fewer or no benefits from EBPs (Cornet et al., 2014). Stadler and colleague’s (2008) analysis of 23 children with DBDs found that youth with a high resting heart rate exhibited significant decreases in delinquency and aggression (~13% decrease) after completing an intensive child behavioral management program and a parent training program, while there was little change (4%-5% decrease) in the outcomes for intervention youth with a low resting heart rate. Similarly, Shenk et al. (2012) found that youth with lower testosterone experienced significant decreases (~70%) in the likelihood of a conduct disorder (CD) or oppositional defiant disorder (ODD) diagnosis at a 3-year follow-up after completing an intervention program that integrated PMT, child CBT, and family therapy. Youth with higher levels of testosterone experienced smaller reductions (~10%) in conduct disorder or ODD symptoms. Other studies have shown that problem behaviors actually increase after completing childhood EBPs among higher risk youth (Beauchaine, Gartner, & Hagen, 2000). For instance, a study of 22 children with DBD who completed child CBT and PMT reported that youth with high cortisol reactivity exhibited decreases in the mean levels of overt aggression, ODD, and externalizing behavior. In contrast, youth with low cortisol reactivity showed an increase in these problem behaviors post-intervention (van de Wiel, van Goozen, Matthys, Snoek, & van Engeland, 2004).

Recommendations Based on EBPs in Childhood

The above studies suggest that treatment outcomes may be worse for youth with lower stress response, low autonomic system activity, and higher testosterone. These findings beg the question of why these youth may not benefit from childhood EBPs. One hypothesis is that such youth may be hypersensitive to rewards and quick to perceive hostility or threats from others (sometimes inaccurately), yet they are not sensitive to punishment, exhibit lower levels of fearfulness, and show higher levels of callous–unemotional (CU) traits than other children (van Goozen & Fairchild, 2008). Thus, contingency management systems that emphasize sanctions or punishment do not deter such children (Raine, Venables, & Williams, 1990). van Honk, Harmon-Jones, Morgan, and Schutter (2010) argue that individuals may exhibit this high approach/low avoidance phenotype because of an imbalance in the ratio of testosterone and cortisol. Increased levels of testosterone and lower levels of cortisol lead to less coupling between cortical and subcortical regions of the brain, so that the amygdala cannot effectively transfer information to the ventromedial prefrontal cortex (VMPFC)/oribitofrontal cortex (OFC), leading to ineffective decision making and aggressive behaviors. Individuals with an imbalance between testosterone and cortisol, however, exhibit a heightened sensitivity to rewards, suggesting that incentive-based interventions may be more effective with this population.

Indeed, research has shown that a subset of delinquent youth (~20%-50%) display high levels of CU traits, low punishment sensitivity, and greater sensitivity to rewards, and that these youth exhibit significant improvements in antisocial behavior when interventions focus on incentives rather than sanctions (Frick, Ray, Thornton, & Kahn, 2014; Matthys, Vanderschuren, Schutter, & Lochman, 2012). For example, an evaluation of a 10-week PMT program found that sanctions (i.e., time outs) were generally regarded as ineffective for conduct disordered boys who exhibited high levels of CU traits, but that reward-based strategies were effective for both high and low CU conduct-disordered youth (Hawes & Dadds, 2005). Along similar lines, Caldwell, Skeem, Salekin, and Van Rybroek (2006) assessed the effectiveness of an intensive behavioral modification program for incarcerated juvenile delinquents that emphasized rewarding youth for short periods of compliant and prosocial behavior. Compared with a propensity score matched group of youth from traditional juvenile detention facilities, youth from the intensive reward-based behavioral modification program were 2.7 times less likely to be charged with a violent offense.

The above findings indicate that approximately 40% of youth may not respond to traditional childhood EBPs, and that such EBPs may need to be modified to increase the effectiveness of such interventions. The question of whether higher or lower risk children benefit most from EBPs is mixed as some studies provide support for either side. Yet a review of the research shows that a particular profile of youth appears to be resistant to childhood EBPs. Specifically, this subset of youth are CU, reward driven, and insensitive to punishment. These psychological factors may arise due to genetic factors that have been linked to stress response (NR3C1) and increased sensitivity to rewards (DAT1 and DRD4), an imbalance of testosterone and cortisol, and lower heart rate activity and reactivity, which may partially be influenced by the imbalance of testosterone and cortisol. As such, clinicians may assess youth on CU traits and a range of biomarkers (such as heart rate and stress response) to determine whether a potential client fits within the profile emerging from the literature. If it appears that the client fits the profile described above, the clinician may recommend services that focus on incentives and the client’s strengths, rather than traditional criminal justice practices that emphasize discipline and sanctions.

Clinicians may also consider integrating nutritional supplements into traditional EBPs to provide a more holistic approach to treatment. A number of studies have found that a daily dose of 500 to 1,000 milligrams of omega 3s is related to a reduction of attention problems, conduct problems, oppositional defiance disorder symptoms, impulsivity, aggression, and internalizing disorder in children (Gustafsson et al., 2010; Itomura et al., 2005; Meyer et al., 2015; Raine, Portnoy, Liu, Mahoomed, & Hibbeln, 2015; Schoenthaler, Amos, Doraz, & Kelly, 1997; Stevens et al., 2003, except see Kirby, Woodward, Jackson, Wang, & Crawford, 2010). The mechanism underlying omega 3 supplementation and behavior problems is currently unknown. One hypothesis, however, is that low omega 3 levels are linked with lower levels of serotonin, and low serotonin (especially in the prefrontal cortex) has been associated with reduced heart rate variability and low autonomic nervous system activity (Hamazaki & Hamazaki, 2008; Hibbeln, Ferguson, & Blasbalg, 2006). Studies have shown that increasing omega 3s through diet increases autonomic nervous system activity (Christensen, Christensen, Dyerberg, & Schmidt, 1999); thus increased autonomic nervous system activity may be one mechanism by which omega 3 supplementation reduces antisocial behavior. Clinicians who are considering using omega 3 supplements for their clients are encouraged to (a) examine the baseline level of omega 3s via blood or cheek samples to establish that a client has a lower level of omega 3s as some aggressive individuals have exhibited high levels of omega 3s prior to supplementation (Meyer et al., 2015); and (b) consider prescription-based omega 3 supplements (such as Lovaza or Vascepa), to research the purity of the omega 3 supplement as these supplements are not regulated by the Food and Drug Administration, or to consider food sources of omega 3s such as fatty fish (i.e., salmon, herring, tuna). Despite these considerations, it is believed that the combination of omega 3 supplementation and a reward-based behavioral management program may be effective in improving treatment responses for youth with low autonomic nervous system activity.

Biological Outcomes of EBPs in Adolescence

Despite adolescence being a biologically dynamic point in the life course, few studies, to our knowledge, have investigated whether adolescent EBPs lead to changes in youth biological functioning. Our review of the literature only identified two such studies, and both studies show that adolescent EBPs affect the biological stress response system. The first study—McCraty, Atkinson, Tomasino, Goelitz, and Mayrovitz (1999)—assessed whether an emotion self-management program delivered during the seventh- and eighth-grade years affected interpersonal skills, aggressive behavior, heart rate variability, and various psychological risk factors among at-risk youth. The authors found that youth who completed the program experienced increases in focusing on their work, feeling energized, feeling comfortable with teachers, having a positive attitude about school, being assertive, having an internal locus of control, stress management, self-satisfaction, and being resistant to peer influence 6 months after completing the program. Intervention youth also exhibited increases in heart rate variability and heart rhythm coherence when recovering from a stressful situation (relative to control youth), which reflected greater parasympathetic activity. The parasympathetic branch of the autonomic nervous system is responsible for decelerating emotional and physiological arousal after a stressful event, and thus, activation of this system shows that intervention youth were able to calm down more effectively and efficiently than youth who had not received the self-management program.

Similarly, Nickel et al. (2006) examined the effectiveness of a 12-week brief strategic family therapy (BSFT) among 72 males between the ages of 14 and 15 who had self-reported that they engaged in bullying behavior. The BSFT focused on helping youth and their families identify their communication styles and improve their conflict-resolution skills. Thirty-six youth received the BSFT, and 36 youth were placed in the control group. At the end of the 12-week intervention, youth in the BSFT group had a larger increase in morning cortisol secretion levels and a greater reduction in anger indices than the control group. Low cortisol levels have been consistently linked to greater involvement in externalizing behaviors in children and adolescents (McBurnett, Lahey, Rathouz, & Loeber, 2000; Shirtcliff, Granger, Booth, & Johnson, 2005), and thus, BSFT reduced youth biological risk for antisocial behavior. Nickel and colleagues also found that increases in cortisol levels among the BSFT youth were correlated with reductions in anger (especially anger directed at others) and improvements in social functioning, control of one’s anger, and mental health. This study is unique in that it not only shows that the intervention affects a biological risk factor (i.e., cortisol), but that changes in biological functioning is linked with psychological and behavioral improvements.

Biological Moderators of EBPs in Adolescence

Overall, the research shows that adolescent-based EBPs may be less effective with youth who show problems with attention and concentration, have lower cognitive efficiency, have difficulty identifying others’ emotions, perform poorly on risky decision-making tasks, are less sensitive to subtle rewards and punishments, and have higher levels of testosterone and abnormal cortisol levels (Cornet et al., 2014; Fishbein et al., 2006). For instance, Fishbein, Hyde, Coe, and Paschall (2004) found that youth who were not responsive to an effective prevention program (~50% of participants) made more omission errors on the Conners’ CPT (i.e., attention deficits), had slower reaction times on the CPT (i.e., cognitive efficiency), engaged in riskier decisions and won fewer points on a gambling task, and had lower skin conductance rates during the more tedious psychological tasks. Interestingly, however, the authors found greater skin conductance during the gambling tasks, suggesting that youth may be sensitive to rewards and punishments during risky activities or “sneaky thrills.” Similarly, Ryan and colleagues’ (2013) analysis of 112 male adolescents in a multi-systemic therapy (MST) program found that the presence of delinquent peers hindered the ability of MST to significantly reduce aggression and delinquency for boys with high levels of testosterone. Males who had a high level of testosterone and above-average numbers of delinquent peers experienced the lowest amounts of change in aggression and delinquency, suggesting that testosterone may enhance the rewarding properties of deviant peers and hinder the effectiveness of adolescent EBPs such as MST. Finally, another evaluation of MST found that two types of youth experienced little to no change in externalizing behavior after completing MST: (a) males with high morning cortisol levels and who reported high levels of daily stress, and (b) males with low afternoon cortisol levels and a high exposure of lifetime stress (Schechter, Brennan, Cunningham, Foster, & Whitmore, 2012).

Recommendations Based on EBPs in Adolescence

Together, the above studies suggest that traditional adolescent EBPs may need to be modified or supplemented with additional interventions to increase treatment effectiveness for a subset of youth. Similar to the results from the child EBP studies, some youth who perform poorly in treatment are not sensitive to subtle rewards. In particular, rewards evoked a positive biological response when they were delivered in a gambling format (Fishbein et al., 2004). This idea may be integrated into current programs by using incentive-based approaches such as the fishbowl, which is currently used in problem-solving courts. When clients follow their treatment and supervision requirements for the week, their names are entered into a drawing or “the fishbowl” for a prize, such as a gift card. The anticipation of receiving a reward may produce a larger neurobiological response than the actual receiving of the reward (Schultz, Dayan, & Montague, 1997), and thus, programs may enhance compliance and treatment effectiveness while being fiscally conscientious.

Also, the research shows that some youth who do not respond to treatment exhibit problems in paying attention, concentrating, and inhibiting responses. Youth who demonstrate deficits in executive functioning may benefit by completing cognitive remediation or enhancement programs prior to beginning traditional EBPs, to strengthen executive functioning skills. Cognitive remediation programs have been developed and extensively used with individuals diagnosed with schizophrenia (Eack, 2012), but there is evidence that these programs may benefit other at-risk populations. Studies of individuals with ADHD have shown that cognitive remediation programs reduce attention deficits, reduce impulsivity, and improve the ability to organize and manage daily tasks (Stevenson, Whitmont, Bornholt, Livesey, & Stevenson, 2002). These programs may be delivered in a therapeutic face-to-face or videogame format (Anguera et al., 2013), and they may be as simple as drill and practice exercises that are completed on a computer or as broad as strategy coaching techniques that are completed in a group setting. Indeed, a recent study evaluated the effect of practicing a simple task of identifying the direction of an arrow amid other arrows (→ → ← → → vs. → → → → →) three times a day for 6 days (Cohen et al., 2016). The authors found that practicing this task lead to increased connectivity between the frontal cortex and limbic system, suggesting that such practices may be beneficial for improving both nonemotional (i.e., attention) and emotional skills.

Youth with executive functioning deficits, high testosterone, and/or low cortisol activity may also benefit from neurofeedback. Neurofeedback is a user-directed EEG program in which clients train their brains to exert particular brain waves through a series of operant and classical conditioning exercises. With youth and adolescents, a sensory cap is placed on their scalp and a particular brain frequency causes activity within a videogame such as driving a car or a rocket ship. Activity within the videogame (via brain activity) helps the youth accumulate points and “win” the game. Neurofeedback has been shown to be an effective intervention in reducing ADHD and other externalizing behaviors, especially for treatment resistant or difficult to treat clients (Huang-Storms, Bodenhamer-Davis, Davis, & Dunn, 2006; Joyce & Siever, 2000; Martin & Johnson, 2005; Smith & Sams, 2006; van Outsem, 2011).

Finally, one additional intervention that may be used to supplement adolescent EBPs could be mindfulness training or yoga (Fishbein et al., 2015). Both practices have been shown to increase activation of higher order brain regions that are responsible for regulating lower order limbic regions, and they have been linked with improvements in attention and self-regulation (Hölzel et al., 2011). A number of studies have shown that yoga or mindfulness training programs that are 8 to 10 weeks long have been effective in reducing problems in attention, self-regulation, executive functioning, stress reactivity, and behavioral problems among youth with ADHD (van de Weijer-Bergsma, Formsma, Bruin, & Bögels, 2012) and delinquent or at-risk youth (Fishbein et al., 2015; Himelstein, Hastings, Shapiro, & Heery, 2012). While the treatment effects may begin to disappear 3 to 4 months after the intervention has ended, it should be noted that mindfulness training or yoga were the primary modes of intervention in listed studies, and that these practices may be used as a supplementary intervention for adolescent EBPs.

Biological Outcomes of EBPs in Adulthood

To date, only one study has examined whether EBPs in adulthood is linked with changes in biological functioning. Cornet (2015a) assessed whether completion of a cognitive skills program normalized heart rate activity. Her analysis of 121 Dutch prisoners found that individuals who completed the cognitive skills program did not significantly differ from controls on changes in resting heart rate, heart rate reactivity to a stimulus (i.e., the D2 cancelation task), resting respiration sinus arrhythmia (RSA), or changes in RSA in response to the D2 cancelation task. Also, she did not find that the intervention correlated with changes in executive functioning. The latter finding is in line with E. H. Ross’ (2012) results that completion of cognitive skills programs had little to no effect on inmates’ executive functioning. These findings indicate that if cognitive skills programs are effective, they may not be affecting heart rate and executive functioning at a psychological task level. Other studies have argued that CBT programs may lead to changes in various brain regions, which may be detected through EEG or functional magnetic resonance imaging (fMRI)-level analysis (Vaske, Galyean, & Cullen, 2011). This hypothesis, however, has not been empirically tested in a criminal justice population to date.

Biological Moderators of EBPs in Adulthood

Similar to the biological moderators for EBPs in adolescence, research generally shows that adults who fail to complete treatment or receive fewer benefits from treatment tend to have difficulty paying attention, inhibiting their behavior, and updating and reorienting their response after committing an error than adults who respond to EBPs (except see Mullin & Simpson, 2007). Cornet, van der Laan, Nijman, Tollenaar, and de Kogel’s (2015) analysis of 121 Dutch incarcerated males in a cognitive skills program found that inmates who performed worse on a concentration/sustained attention task (i.e., D2 cancelation task) self-reported receiving fewer benefits from treatment and were more likely to drop out of treatment than clients who did well on the task. The D2 cancelation task was the strongest correlate of treatment dropout, and it explained approximately 13% of the variation in dropout (Cornet, van der Laan, et al., 2015). Similarly, Fishbein et al.’s (2009) examination of 197 male inmates in the Thinking for a Change program found that clients who exhibited problems inhibiting responses and shifting responses on a psychological task (i.e., Stop/Change Task) were assessed as gaining less and being less responsive to treatment by social workers, were more likely to drop out of treatment, and were less likely to report improvement in aggressive reactions to provocation. Their analyses also found that clients who self-reported receiving fewer benefits from treatment exhibited low levels of cortisol reactivity in response to a social stressor. This parallels the findings from childhood EBP studies, which also showed that youth with low stress response had worse treatment outcomes than clients with normal or higher stress response.

EEG studies also suggest that individuals with attention, decision-making, and response-inhibition problems are less likely to complete treatment. An EEG study of 88 incarcerated adults in a 12-week cognitive-behavioral substance-abuse program found that treatment noncompleters showed smaller event-related potentials that are believed to measure attention (P2) during a go/no go task (Steele et al., 2014). Also, noncompleters differed from treatment completers on two ERPs that are thought to reflect the processing of errors in the cingulate cortex and regions of the frontal cortex (Overbeek, Nieuwenhuis, & Ridderinkhof, 2005). Similarly, another study found that P3a ERPs in response to an auditory tone were lower among individuals who failed to complete relapse prevention and life skills treatment at a residential drug and alcohol treatment facility (Anderson, Baldridge, & Stanford, 2011). P3a amplitude successfully identified 76.9% of individuals who did not complete treatment, and predicted treatment completion after controlling for age, severity of substance-use problem, IQ, and education. P3a amplitude is thought to reflect one’s attention to novel stimuli, and it has been shown to be positively correlated with executive functioning (Fjell, Walhovd, Fischl, & Reinvang, 2007). Thus, one could speculate that individuals who are at-risk for failing to complete treatment or gaining fewer benefits from treatment may exhibit problems learning from one’s errors and show less attention to novel (yet ordinary or everyday) stimuli.

Finally, results from fMRI studies also converge with those from EEG and psychological tasks to show that individuals who are at-risk for relapse show less activity in brain regions responsible for attention and optimal decision-making than those who remain drug free. Gowin and colleagues’ (2014) analysis of 68 methamphetamine-dependent individuals in a 28-day treatment program found that individuals who relapsed showed less activity in the VMPFC, right frontal gyrus, and insula during the Risky Gains Task than those who abstained from drug use. Similar results have been found by Clarke et al. (2014) and Paulus, Tapert, and Schuckit (2005), which found that individuals who relapsed had less activity in the posterior cingulate cortex (PCC), insula, VMPFC, and middle temporal gyrus during psychological tasks. These regions are implicated in directing attention to emotionally salient cues (PCC), consciously recognizing emotions and physiological sensations or “hunches” (insula), choosing optimal outcomes in decision making (VMPFC), and voluntary attentional control (middle temporal gyrus and frontal regions; Bechara, 2002; Hopfinger, Buonocore, & Mangun, 2000). While these studies may suggest the neural correlates of attention and decision making are underaroused during psychological tasks among relapsers, it is important to note that a study of cocaine-dependent individuals who relapsed found that the PCC and regions of the temporal cortex exhibited greater activity when respondents were exposed to drug-related cues (i.e., videos of individuals using cocaine). This latter finding builds on Raine et al. (1990) and Fishbein et al.’s (2004) argument that at-risk clients may show greater sensitivity to “risky” situations than the average client but that they exhibit patterns of underarousal to the typical or routine psychological task.

Recommendations Based on EBPs in Adulthood

The above evidence suggests that, similar to adolescents, adults who benefit less from EBPs tend to be those who have difficulty paying attention, inhibiting responses, and making efficient and effective decisions. Similar to the previous recommendations, clinicians may consider screening adult clients for executive functioning problems, and then recommending cognitive remediation training prior to EBP treatment or supplementing the EBP with mindfulness, meditation, yoga, and/or nutritional programs. All of these alternatives have been shown to be effective in reducing antisocial behavior in both institutional and community-based offender populations (Gesch, Hammond, Hampson, Eves, & Crowder, 2002; Himelstein, 2011; Orme-Johnson, 2011; Wupperman et al., 2012; Zaalberg, Nijman, Bulten, Stroosma, & van der Staak, 2010). In addition, clinicians and researchers may consider the use of incentive-based programs to increase treatment compliance and reduce recidivism. The Operation Peacemaker Fellowship program in Richmond, California, has provided monetary incentives for treatment-resistant offenders to engage in prosocial behaviors (i.e., get a GED, become employed) and refrain from criminal behavior. A 2014 evaluation of the program showed that of the 40 chronic offenders enrolled in the program, 84% did not have any new firearm-related arrests at the end of the program (Office of Neighborhood Safety, 2014). Similarly, a growing body of research shows that contingency management programs—which provide a graduated scale of monetary vouchers for each clean drug screen—are as effective as cognitive-behavioral interventions in reducing short- and long-term drug use (Rawson et al., 2006). These incentive-based programs may align with the neurobiological findings that at-risk clients may be especially sensitive to rewards, and thus one may increase compliance by integrating a reward-based program into one’s current treatment plan.