Although future expectations are consistently linked to juvenile delinquency, whether or not these expectations change contingent on behavior remains unclear. In addition, few studies have considered the role an official arrest plays in changing the expectations an adolescent holds for his future. The current study (1) examines the reciprocal relations between self-reported delinquency and adolescent future expectations to graduate from college, to have a successful job or career, and to stay out of trouble with the law, and (2) evaluates the reciprocal relations between re-arrest and future expectations. To address these questions, a sample of 1,166 male juvenile offenders were recruited after their first arrest and followed for 12 months. We find partial support for our prediction that educational, occupational and behavioral expectations will influence behavior, but also that behavior will influence future expectations. In addition, our results suggest that re-arrest predicts changes in expectations for staying out of trouble, but has no effect on educational and occupational expectations.

Compared to children and adults, adolescents make riskier choices and do so more often when in the presence of peers. Traditional cognitive explanations for adolescent behavior have failed to account for increases in risk-taking during this developmental period. More recent biopsychosocial models of adolescent risk-taking have emerged, highlighting the importance of not just cognitive but social and biological factors that contribute to adolescent risk-taking. Nonetheless, one biological system- the adolescent physiological stress system- has been understudied and may add to our understanding of adolescent risk-taking. More specifically, it may be that physiological stress makes adolescents vulnerable to making risky decisions by increasing their self-conscious affective states. These effects were hypothesized to be more pronounced after a stressful encounter with a peer, while being dampened after a stressful encounter with an adult.

Sixty male adolescents aged 12 to 16 were randomly assigned to one of two Trier Social Stress Test (TSST) conditions. In the first condition, adolescents were evaluated by same aged peers, and in the second, adolescents were evaluated by adults. The manipulation of the age of evaluators in these two conditions was effective, with adolescents in the peer condition perceiving evaluators to be around 17 years old and adolescents in the adult condition perceiving evaluators to be around 31 years old. Throughout the experimental session, adolescents provided 4 whole saliva samples which were assayed for cortisol and alpha-amylase as markers of physiological stress response.

No differences were found between the two TSST conditions regarding physiological stress response and risky decision-making. However, adolescents who were evaluated by adults reported more self-conscious affect compared to adolescents who were evaluated by peers. Additionally, adolescents who were more self-conscious experienced larger changes in salivary alpha-amylase. Although adolescence is a time of social orientation towards peers, the results of the current study illustrate that adults’ negative evaluations are powerful and influence adolescents’ emotions and physiology. These findings suggest the potential iatrogenic effects of negative adult evaluations in environments like classrooms and juvenile courtrooms.

Adolescent and young adult substance use is associated with a wide range of poor mental and physical health outcomes and continues to be an area of substantial public health concern. Substance use has been linked to higher rates of long-term substance abuse and dependence, worse academic and occupational outcomes, impaired cognitive function, poorer physical and mental health, and greater criminal behavior. Therefore, the goal of this two-part dissertation was to better understand the risk and protective factors linked to substance use among adolescents and young adults. Employing a combined sample of 2,386 justice-system-involved youth from the longitudinal Pathways to Desistance and Crossroads studies, Study One examined the associations between a broad range of risk and protective factors and within-person changes in substance use from adolescence through young adulthood, whether any links between predictors and substance use were asymmetric, and whether associations varied by age. The results indicate that offending and peer substance use were the factors more strongly related to different types of substance use. While most predictors did not have asymmetric effects, offending was asymmetrically associated with alcohol use and future expectations were asymmetrically related to marijuana use. Further, decreased offending was more strongly related to lower rates of alcohol use in young adulthood than in adolescence, while having more positive future expectations was related to decreases in marijuana use throughout adolescence and young adulthood. Study Two identified whether the COVID-19 pandemic was related to changes in substance use and whether depressive symptoms, anxiety symptoms, sensation seeking, future expectations, and motivation to succeed moderated those changes. Findings show that alcohol and marijuana use increased during the COVID-19 pandemic and sensation seeking and future expectations moderated the increases in alcohol use, while depressive symptoms, anxiety symptoms, and sensation seeking moderated the increases in marijuana use. Overall, the results from Study One suggest that intervention and treatment efforts should focus on reducing criminal behavior and discouraging friendships with substance-using peers to help youth reduce or desist from using substances. Reducing offending during young adulthood in particular may be effective in discouraging problematic alcohol use. The results from Study Two highlight the importance of providing mental health services, helping high sensation seeking youth find healthy outlets, and providing resources to encourage more optimistic expectations of life after the pandemic to reduce costly iatrogenic effects such as problematic substance use that could have lasting impacts on youth long after the pandemic is over.

Callous-unemotional (CU) traits (e.g., lack of empathy, deficient guilt/remorse, and shallow affect) are a risk factor for delinquent behavior. Youth high in CU traits exhibit a range of cognitive and emotional deficits, such as fearlessness, insensitivity to punishment, and reward dependency, that predispose them to antisocial behavior and justice system involvement. This two-part dissertation study sought to better understand how CU traits predispose youth to delinquent behavior by examining the hormones thought to underlie these emotional and cognitive deficits. Employing a sample of 55 justice-involved male youth recruited from an ongoing longitudinal study of first-time juvenile offenders, Study One examined the association between CU traits, adverse experiences (prior exposure to violence and hostile parent-child relationships), and cortisol and alpha-amylase reactivity. The results indicate that individuals with elevated levels of CU traits exhibit lower cortisol reactivity than their counterparts. However, they exhibited the same pattern of alpha-amylase reactivity. In addition, prior adversity did not moderate the association between CU traits and either indicator of stress reactivity. Study Two tested whether cortisol and testosterone mediated the association between CU traits, reward and punishment sensitivity, and risk taking. Unfortunately, the study was underpowered and could not determine whether there was a direct or indirect association between CU traits, risk taking, or reward/punishment sensitivity. However, the study provided evidence that individuals with high CU traits exhibited the same pattern of testosterone reactivity as their counterparts. Taken together, the results of the two studies suggest youth with elevated levels of CU traits exhibit a pattern of stress reactivity that may contribute to their behavior. Future research should continue to explore the relation between stress reactivity, CU traits, and other aspects of antisocial behavior, such as aggression and retaliation.

Teenagers who hold optimistic expectations for their future are less likely to engage in crime and risk-taking. This dissertation examined three questions related to future expectations among two samples of adolescent males who had all been arrested (Crossroads Study, N = 1216; Pathways to Desistance, N = 1170). The Crossroads study recruited males between the ages of 13-17 who had experienced their first arrest after a low level offense (i.e., misdemeanor), and interviewed them for four additional years. The Pathways study consists of males between the ages of 14-18 who were arrested for serious felony-level offenses and were interviewed multiple times over the course of seven years.

Study one used data from Crossroads and differentiated future expectations (the perceived likelihood of achieving one’s goals) from future orientation (the tendency to think about the long-term consequences of one’s decisions). Results supported the analytic distinction between expectations and orientation, and identified a portion of adolescents who displayed high expectations, but low orientation. This imbalance related to crime, substance use and casual sex. Study two assessed four mechanisms (impulse control, the perceived personal and social rewards of crime, and the perceived social costs of crime) to explain the link between future expectations and offending among youth in Pathways. Only the perceived social rewards of crime was a significant mediator, and suggested that youth with low expectations perceive more social rewards from crime, which relates to their decision to commit crime. Study three used data from Crossroads to understand the relation between estimated life expectancy and crime across development. The results revealed that youth become more optimistic about their chances for survival as they age, and also supported significant within-person associations with crime: as adolescents increase their life expectancy, they report lower levels of delinquency. The findings from the three studies underscore the importance of fostering positive expectations among adolescent males, and highlight the diversity of expectations among high-risk youth. Despite all participants experiencing at least one arrest, many continued to report positive future goals, and these goals played an important role in deterring continued risk-taking and crime.

Laws legalizing recreational marijuana use have recently been passed in several states across the U.S., sparking debate about their impacts on marijuana use among adolescents and young adults. Critics are concerned that legalization will lead to increased use due to greater access to marijuana. Advocates, on the other hand, argue that legalization will reduce the “forbidden fruit” effect, leading to decreased use. To shed greater light on this debate, the present study investigates the impacts of legalization on marijuana use among young adults who have had prior experiences with the justice system. Importantly, young adults are at higher risk of marijuana use than any other age group and those who are involved in the justice system are at even greater risk. As such, we compare justice-system-involved (JSI) young adults in California (N = 396), where recreational marijuana use was recently legalized, to those in Pennsylvania (N = 372), where recreational use is still prohibited. Results indicate that rates of marijuana use increased significantly in both states, regardless of whether participants lived in a legalized state. While these findings are surprising, they suggest that increases in rates of marijuana use might be better explained by factors beyond legalization laws. In fact, these findings may indicate that increased state legalization of recreational marijuana use is giving rise to a culture of acceptance among JSI young adults.

It has long been a concern that young adults are not as civically engaged as other age groups. This historical pattern held true for the 2016 and 2020 presidential elections, despite increased voter turnout overall. A group that might be particularly at risk for low voting turnouts is that of justice-involved young adults. Although several studies have assessed voting behaviors among community youth, it remains unclear to what degree justice-involved young adults are civically engaged, and how contact with the justice system affects their voting behaviors. This is concerning, as this population is arguably more directly affected by voting outcomes because of their close association with the law, which dictated how they were processed, sentenced, and their conditions of release. Using data collected after the 2016 general election, this study aimed to understand the effect of justice system involvement on young voters. Additionally, it served to introduce a profile of the justice-involved youth voter, assessing basic demographic characteristics as predictors of voting behavior and providing a descriptive overview of the political attitudes of participants. Research was conducted within a racially diverse sample of young adult men (N = 927) who were part of the Crossroads Study, an ongoing longitudinal study following 1,216 participants (initially aged 13-17) for seven years after their initial arrest for a moderate offense. Results showed the high-risk young adults voted at lower rates than community young adults. Additionally, 46% of the sample reported being not at all interested in politics. Similar to the community sample, age and SES were found to be predictors of voting. Race and prior offending, however, were not significantly associated with voting behavior. Although the direct association between processing type and voting behavior was not significant in the fully adjusted model, the indirect effect between processing type and voting behavior (through arrest count) was significant. Formal processing was associated with more arrests, and higher arrest count was subsequently linked to a lower likelihood of voting. Number of re-arrests explained 59% of the association between processing type and behavior.

Legal socialization is the study of how individuals develop their perceptions of legal authorities and laws. Studies demonstrate that adolescents with negative perceptions of legal authorities tend to violate the law, yet the literature has not adequately addressed how adolescents develop their perceptions, why their perceptions affect behavior, and, despite being a critical component of legal socialization, how individuals develop their perceptions of law. This dissertation applied a developmental approach to filling these gaps in the literature. The first study demonstrated that the development of perceptions of the justice system through adolescence varies based on both race and experience. Specifically, Black youth report the most negative perceptions, followed by Latino, and then White youth, and racial differences increase as youth age. Further, attitudes become worse over time for youth who remain in the system. The second study showed that youth differentiate between legal authorities, such as law enforcement and the justice system, and socio-cultural authorities, such as schools and religious institutions. Further, across races/ethnicities youths’ perceptions of law enforcement have declined markedly in recent years. The third study demonstrated that the effect of perceptions on delinquency varies based on developmental level. Finally, the fourth study revealed that the obligation to obey the law declines through adolescence and how youth perceive law is strongly associated with how their parents view law.

According to Newman's response modulation hypothesis, psychopathy is underpinned by a cognitive processing deficit that interferes with the ability to attend to peripheral information. However, recent research and theory suggests that externalizing traits are also associated with performance deficits on response modulation tasks. This confusion over the specificity of response modulation deficits has been extended downward to adolescent populations during a time when normal development processes may account for observed deficits. In this study of 84 adult and 98 adolescent offenders, I examined whether the association between response modulation deficits and psychopathy is driven by specific underlying affective, interpersonal, and externalizing traits or an interaction between traits (assessed by the PCL-R/PCL:YV and PPI/YPI). Results indicate that affective features of psychopathy (PCL-R Affective and PPI Coldheartedness) were independently associated with response modulation deficits among adults. In contrast, externalizing traits (PCL:YV Antisocial; trend for YPI Lifestyle) were associated with enhanced response modulation among adolescents whereas scales assessing affective-interpersonal features of psychopathy manifested less of a relationship with response modulation deficits (trend for YPI Interpersonal). These findings raise questions about the generalizability of response modulation and the assessment of psychopathy across developmental phases.