Using Developmental Science to Reorient Our Thinking About Criminal Offending in Adolescence

Abstract

The concept of adolescent development has become a relevant consideration for researchers interested in juvenile delinquency. However, the integration of constructs from developmental psychology into delinquency research is still in its early stages. This article argues that it is time to move beyond description of adolescent antisocial activities and to integrate developmental activities into delinquency research as mediators or moderators. Relevant examples of such an approach are presented.

Keywords

juvenile delinquency deviance crime

Adolescent development has cachet these days in juvenile justice research, practice, and policy. It is now generally accepted that adolescents are qualitatively different from adults (both neuropsychologically and psychosocially) and that these differences matter for a range of juvenile justice issues, including culpability and amenability to interventions (Bonnie et al. 2013; Scott and Steinberg 2010). As a result, there is a renewed energy among researchers for thinking about how antisocial behavior and other capacities and behaviors change systematically over adolescence as well as the implications of these patterns of development (Woolard 2012).

This focus on development and delinquency presents a set of challenges for researchers. First of all, it complicates, rather than clarifies, conceptualizations of delinquency. The current emphasis on developmental processes is another step in the shift, going on for some time now, away from the logic of the medical model that underpinned many of the frameworks of early juvenile court interventions (Rothman 1980). It reflects an orientation away from the premise that delinquency is a singular “pathology” that has a uniform set of interventions that will reduce or eliminate it. There is no yet to be discovered or implemented “cure” for delinquency, since it is a complex outcome produced by a multitude of influences, many of which shift as an adolescent grows up. Introducing the ideas of shifting capacities and influences over different stages of development makes the idea of a comprehensive theory of delinquency all that more chimeral.

At the same time, adoption of a developmental perspective holds out substantial promise. In this perspective, much of delinquency is seen as normative for a certain age or as a passing (but illegal) indicator of difficulties of development during a particular stage of life. Consequently, it is posited that understanding the developmental patterns of antisocial behavior could promote more targeted and theoretically informed assessment and intervention. If we can unravel the developmental phases of delinquency, we can possibly do the right intervention at the right time for the right adolescent. The vision is that more solid developmental knowledge could help identify adolescents with likely long-standing behavioral problems and point toward age-appropriate interventions to slow the development of antisocial behavior and/or promote prosocial behavior.

The posited potential of a developmental perspective on delinquency gives it a certain “mom and apple pie” quality. It is hard to be against the idea; it is attractive on a number of levels. Anyone who has raised a child usually readily agrees that their child’s judgment and emotional stability often seemed impaired during mid-adolescence. Practitioners working with adolescent offenders generally note that, for the vast majority of adolescents in the juvenile justice system, there is a good chance of positive change in the future, even for those with rather dismal histories. Who these current offenders are is no guarantee of what they could or might be in the future. Adopting the idea of development reinforces hope and raises the possibility of more focused problem identification and effective interventions. The challenge is not necessarily convincing people that the developmental perspective makes sense broadly. The current challenge is instead to figure out the specifics of constructively integrating knowledge about adolescent development into our conceptions of delinquency and juvenile justice.

This article provides a short examination of the potential of a developmental perspective to promote useful research regarding delinquency. Understanding the connections between development and criminal involvement may improve risk assessment, increase the effectiveness of interventions, or advance theory about delinquency, but the path to these outcomes is far from clear. The literature on adolescent development and delinquency is somewhat disjointed and still evolving. A consideration of what it may mean to integrate development into delinquency research is probably in order.

Moving beyond Description

Much of the work on adolescent development and delinquency has focused on documenting patterns of adolescent offending over childhood, adolescence, and early adulthood. The fact that numerous longitudinal databases (e.g., the Dunedin study, Moffitt 1993; the Causes and Correlates study, Thornberry and Krohn 2003) were available during the same period when advances were being made in the statistical modeling of change over time (Muthen and Muthen 2000; Nagin 2005) certainly contributed to advances in this area. A sizable body of studies has now documented that many children identified with behavior disorders at an early age may be diverted from adolescent delinquency (Loeber, Burke, and Pardini 2009) and that most samples of at-risk and justice-involved adolescents generally follow decreasing patterns of arrest and self-reported offending as they enter early adulthood (Piquero 2008) and that there appear to be identifiable groups of adolescents who follow different paths of offending over adolescence and early adulthood (Broidy et al. 2003).

Perhaps the best-known theoretical formulation arising from this type of work is that of “adolescent-limited” and “life course persistent” offenders based on the work of Moffitt (1993). One implication of this work is that there are adolescents for whom involvement in delinquency is part of normal adolescent processes, while another group of adolescents have intense needs and deficits that keep them involved in antisocial behavior across successive developmental periods. There is considerable ongoing work to investigate the processes of delinquency involvement in adolescents likely to follow one pattern of involvement or the other.

Despite these advances, there is still a great deal of work to be done to enrich theoretical formulations regarding delinquency based on “trajectory group” studies. Trajectory analyses of offending patterns provide new possible ways to group cases that follow the same patterns of offending over time, but this post hoc description is only the first step toward theoretical formulations of how adolescents become or stay involved with offending at different developmental periods. These studies have been valuable descriptive investigations, but the theoretical refinements from them have been rather limited. There has been some work (but really a rather limited number of studies) trying to identify “turning points” (Abbott 1997), where the expected pattern of offending is disrupted in either a negative or a positive direction (e.g., Haviland et al. 2008). There are also a limited number of studies that test how changes in behavior or developmental indicators affect subsequent levels or types of offending. It seems that the field is poised to move beyond description to more of a focus on explanation, as challenging as that might be. It would be an unfortunate diversion if the field becomes ensconced in extended exchanges about the boundaries of these hypothetical groups rather than the potential dynamics of their processes of change (Sampson and Laub 2005).

Whether we take up this challenge to do more explanation and less description depends at least somewhat on our ability to be knowledgeable and creative analytically. Most analyses of longitudinal data sets have examined the relative strength of certain risk indicators (e.g., peer associations, parental monitoring) to an outcome measured at some later time point. These types of regression-based “horse races” are valuable for identifying risk markers of cases that might be most appropriate to enroll in certain types of interventions, but they do not greatly enrich our theoretical understanding of how risk evolves or exerts an influence on the developmental course of these adolescents. There are alternative strategies and approaches in other areas (e.g., developmental psychopathology, Cicchetti and Cohen 2006; substance use careers, Chassin, Hussong, and Beltran 2009) that could provide templates for richer theory based on developmental analyses. Currently, delinquency research has documented patterns of change over time, but has not done extensive work on what produces variability in the patterns of change or how patterns of change in one area mirror patterns of change in another area (e.g., delinquency and substance use; for an illustrative exception, see Sullivan and Hamilton 2007).

Refining our analytic approaches is not an easy task, but it is necessary for fully integrating the idea of development and delinquency. Improved methods exist for looking at fixed and random effects in series of observations, and application to existing data sets could produce a more refined picture of how life changes in adolescents are related to offending behaviors. There are certainly some notable examples of analyses that address the impact of particular developmentally relevant events on offending; analyses of the differential effects of working on offending in adolescents with different histories provide one example (Apel et al. 2007). It is somewhat surprising, however, that so little work of this type has been done. The field could capitalize on existing longitudinal data and enrich our notions of development by looking to other areas of inquiry regarding development as well as moving beyond description and toward more systematic testing of dynamic effects.

Starting with Knowledge about Adolescent Development

One conceptualization of development is the unfolding of underlying internal processes over time, usually measured in terms of age. Age itself is not the driver or an explanatory factor; it is simply a marker of most individuals’ progress along an expected path. This framework is especially appropriate when thinking about adolescent neurobiological and psychosocial development.

In neurobiology, studies using functional magnetic neuroimaging have documented changes in structure and functioning in brain systems as well as changes in the connections among neural circuitry (pruning) that occur during adolescence and into the early 20s. These changes occur mainly in areas of the brain related to executive functioning and impulse control, providing evidence that adolescence is a period when an individual’s more emotionally based incentive processing system overtaxes the capabilities of the self-regulatory system (Steinberg 2010). Given this, adolescence can be thought of as a transitory developmental state where sensation seeking, risk taking, and impulsive behaviors are more likely to occur. From this perspective, heightened risk taking and impulsiveness in adolescence is normative and may exhibit itself in any of a range of behaviors during this period (e.g., binge drinking, unprotected sex, or criminal involvement). The end result of this line of logic is that adolescents in general, by virtue of simply being adolescents, have poorer social judgment and need time to gain psychosocial maturity.

This conception of development has been influential in the courts because it has provided a “scientific” framework for drawing a line between adolescence and adulthood. These types of findings illustrate that the period of adolescence is both distinct from adulthood and transitory; adolescents are qualitatively different and they will not stay that way. This body of developmental research has undergirded the series of recent Supreme Court decisions regarding the need to treat adolescents differently in terms of culpability (Graham v. Florida 2010; Miller v. Alabama 2012; Roper v. Simmons 2005). Because adolescents are naturally, developmentally, and temporarily different than either children or adults in their ability to make certain judgments related to criminal involvement, adult penalties are not seen as proportional when applied to adolescents as a group.

There are also sociological processes that unfold over the period of adolescence. Adolescence is a period of changes in one’s social world and social role, and how these events occur for an individual affects their likelihood of involvement in criminal activities (Mulvey et al. 2004). Adolescence involves numerous predictable shifts in areas like reliance on peers and relations with family members, initiation into the world of work, and establishment of romantic relationships. Some adolescents may also experience changes in schooling, home, or neighborhood environments.

These changes in social environments are also interconnected with the adoption of different social roles, some of which might promote involvement with criminal activity and some of which might reduce the likelihood of this behavior. Integration into a strong deviant peer group, like a gang, might call for more criminal involvement to demonstrate commitment and belonging. Alternatively, increased expectations to work or engage in a serious romantic relationship might reduce the chances of criminal involvement. Individuals who spend their daytime hours in a supervised workplace, their evening hours with their girlfriend, and their nighttime hours sleeping in order to rest for the next workday have limited opportunity to commit much crime. Research regarding substance use indicates that adoption of more adult roles is related to decreases in alcohol and drug use (Kandel and Yamaguchi 1999) and the same mechanisms are likely at work regarding criminal activity. An adolescent’s social world provides a framework in which criminal opportunities may be more or less prevalent and criminal involvement more or less normative.

It is easy to see that these intraindividual and social context changes can often work synergistically in development and these interactions might promote or discourage criminal involvement over the period of adolescence. For example, an aspect of psychosocial maturity, like reduced impulsivity, might be promoted by a structured and supportive work environment in which manufacturing procedures and attendance requirements must be followed. Alternatively, unfolding increases in the capacity to plan ahead might allow for more stable romantic relationships and less time spent in groups with opportunities for crime. Obviously, there are numerous possible, but yet unexplored, mediating and moderating relations between the intraindividual and social context variables that occur over the course of adolescence.

Obviously, the range of possibly influential interactions among these variables affecting criminal involvement in adolescence can be vast and complex. The effects of peer associations on desistance from crime in late adolescence provide an example. Expected reductions in the time spent with peers as individuals enter early adulthood means that there are just fewer opportunities to be involved in group offending (a marker of adolescent offending; Zimring 1998). In addition, though, the influence of these peer associations can be expected to be less powerful over adolescence, as adolescents increase their identity and have less reliance on the opinions of their peers. What should happen developmentally is that adolescents spend less time with peers and are less likely to be swayed by them when they do. In this interactive way, social context and internal psychological change come together to make this influence on criminal behavior much less powerful. Each type of developmental change could have an effect on its own, but it is likely that the strongest effects on reduced offending come from these two processes fueling each other.

If we take these observations seriously, the task before us is to map out these interactions over different ages or developmental periods. We have limited knowledge about how individual risk factors for criminal involvement interact with different contexts, and even less about how these may differ over the course of adolescent development. While contexts and individual capacities change over time, they are always working together to make criminal involvement more or less likely for an adolescent.

This challenge is much like the one faced by contemporary geneticists. Now that we know that genes exist and are associated with particular diseases, the next step is to figure out how they are expressed and altered by particular types of environments. An individual’s capacities, either genetic or criminal propensity based, do not stay immutable over time and experience, and the developmental research task ahead is to map out the influences between capacities and the supporting environment at different periods of adolescence.

The considerable challenge of developing a more refined picture of how capacities and context interact at different development periods requires more integrated and focused theory about adolescent development and involvement in delinquency. Delinquency theories have almost all had an implicit “psychology” behind them about why adolescents engage in antisocial behavior. Many times, however, these psychological explanations have been largely post hoc, derived from macro-level observations about associations of variables. For example, theoretical formulations about the primacy of self-control as a psychological factor driving involvement in delinquency across different ages (Gottfredson and Hirschi 1990) arose initially from analyses of the consistency of the “age–crime” curve and the inability of singular explanations for delinquency to eliminate this effect. Consistency of patterns over time in large samples was taken as an indicator of a consistent set of psychological influences that change over development for all individuals. It is questionable, however, if it is ultimately most useful to look for the most “telling” psychological explanations for consistency of behavior patterns over time.

It might instead be useful to look for expected effects on offending during different developmental phases or across different developmental phases. Returning to our example of peer influences mentioned earlier, it would be useful to know when peer associations are most strongly related to criminal involvement and whether this process is driven strongly by self-selection at some points in development. Monahan, Steinberg, and Cauffman (2009) have taken such an approach, finding that self-selection only operates in early adolescence, that peer influence is more consistently related to subsequent offending throughout adolescence, and that an adolescent’s resistance to peer pressure plays a significant role throughout adolescence as well. Such an approach could be taken with a number of other types of effects on antisocial behavior. For example, one could examine how deviations from normative development at one age (e.g., close romantic relationships or moving out of one’s parents’ home) might be related to later patterns of behavior that are not normative for a subsequent developmental period (e.g., involvement in domestic violence). In general, the study of delinquency has been concerned with identifying individuals who show individual differences at one point and continue to show this pattern at some later point in development. This individual identification model is inherently “clinical” in nature, in that it looks for people who are consistently different across developmental periods. It is not developmental in nature, in the sense that it tells us little (except in retrospect) about the processes that might promote or reduce particular types of antisocial behavior at certain times in a person’s life.

Taking the idea of development seriously in delinquency research in this way would mean starting with what is known about adolescent development (about things like the dwindling effects of peer influence mentioned previously or the formulation of “possible selves”; see Oyserman and Markus 1990) and integrating this information into our research design, sampling, measurement, and analyses of delinquent behavior. It is a different task to use information about the regularities of adolescent development as guiding inquiry rather than explaining overall patterns of behavior.

If information about adolescent development is considered at the beginning rather than the end of inquiries into delinquency, it might point the way toward more focused, less global explanations of how adolescents begin or maintain their involvement in crime. For example, developmental stage (in terms of something like ability to project into the future) might provide a powerful mediating or moderating variable for inquiries into how perceptions of the costs and benefits of crime affect subsequent offending. It is likely that this type of predictable developmental change could have a powerful effect on how well specific deterrence works with adolescent offenders (cf. Anwar and Loughran 2011). More focused inquiries that take developmental features into account could refine the type of information that we generate regarding patterns of adolescent offending.

Conclusion

We are in an era where it is generally accepted that consideration of adolescent development holds promise for enriching theory and practice regarding juvenile offenders. What it means to integrate knowledge of adolescent development and delinquency research is not immediately clear, however. This article proposes some ideas about how this could be done more effectively. There are two general points here.

First, it is necessary to move beyond description of patterns of juvenile offending. A large amount of research effort has gone into documenting patterns of adolescent offending over time, but findings about the dynamics of the observed patterns is less well developed. Testing the effect of risk markers at one developmental period with outcomes at a subsequent period of development can only take us so far, and we may have gotten about as far as we can in this regard.

Knowing how changes in one series of behaviors affects changes in the series of offending behaviors is an approach with considerably more theoretical potential. For example, conducting dual trajectory analyses of behaviors that “walk together” over development in the aggregate could be particularly informative for identifying groups that follow similar or disparate paths in both behaviors over time. Looking at the similarities of developmental paths in different types of offending, substance and offending, or victimization and offending could be particularly informative (see Brame, Mulvey, and Piquero 2001; Sullivan and Hamilton 2007). The patterns of relationships among different behaviors may or may not look the same over different developmental periods. For example, the links between “early” heavy drinking may be very different than the impact of victimization at a young age, and these differences might be related to very different unfolding patterns of peer associations; victimization may reduce antisocial peer associations and early drinking might increase them. The exploration of models like these will only become possible when we begin to explore the simultaneous development of delinquency and other risk behaviors. A large number of possible mechanisms related to offending remain to be explored in this way (e.g., family functioning, institutional stays).

Explorations of how changes in social roles affect patterns of antisocial behavior could also be particularly useful. For example, attainment of a certain status, such as parenthood, could have either negative or positive effects on criminal involvement. Being a parent could mean increased obligations to bring in resources; for a person with little legitimate means of support, this might mean more or a different type of crime. Alternatively, being a parent could mean pressure to provide a positive role model, precipitating a more disciplined lifestyle. How development “state” changes affect involvement in different types of crime is a basic, but underexplored, developmental issue.

Second, integrating what is known about adolescent development as possible mediators and moderators in analyses can focus our knowledge of adolescent offending. There are numerous possible interactions between aspects of intraindividual and social context changes during adolescence yet to explore. Information from investigations that start with what we know about adolescent development and then explore these interactions could greatly enrich our understanding of how development and criminal involvement fit together. For example, level of psychosocial maturity could serve as a moderator when examining the impact of different interventions; changes in reports of family functioning or independence from one’s family could be examined as mediators for positive outcomes from interventions. Consideration of possible moderators or mediators (or even moderated mediation) could provide a much richer picture of the interdependency among developmental processes and offending as well as the differential effects of interventions.

The time for integrating theories of adolescent development and delinquency is here. It certainly seems that the potential for work in this area is great. Directed inquiries that take the next steps are not that involved, but could yield valuable results for juvenile justice theory, practice, and policy if we give developmental science its due.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Abbott

1997. “On the Concept of the Turning Point.” Comparative Social Research 16:85–105.

Anwar

Loughran

. 2011. “Testing a Bayesian Learning Theory of Deterrence among Serious Juvenile Offenders.” Criminology 49:667–98.

Apel

Bushway

Brame

Haviland

Nagin

Paternoster

. 2007. “Unpacking the Relationship between Adolescent Employment and Antisocial Behavior: A Matched Samples Comparison.” Criminology 45:67–97.

Bonnie

Johnson

Chemers

Shuck

. 2013. Reforming Juvenile Justice: A Developmental Approach. Washington, DC: National Research Council.

Brame

Mulvey

Piquero

. 2001. “On the Development of Different Kinds of Criminal Activity.” Sociological Methods and Research 29:319–41.

Broidy

Tremblay

Brame

Fergusson

Horwood

Laird

Moffitt

Nagin

Bates

Dodge

Loeber

Lynam

Pettit

Vitaro

. 2003. “Developmental Trajectories of Childhood Disruptive Behaviors and Adolescent Delinquency: A Six-site, Cross-national Study.” Developmental Psychology 39:222–45.

Chassin

Hussong

Beltran

. 2009. “Adolescent Substance Use.” Pp. 723–65 in Handbook of Adolescent Psychology, 3rd ed., edited by Lerner

Steinberg

. Hoboken, NJ: John Wiley.

Cicchetti

Cohen

, eds. 2006. Developmental Psychopathology. Vol. 3. 2nd ed. Hoboken, NJ: John Wiley.

Gottfredson

Hirschi

. 1990. A General Theory of Crime. Stanford, CA: Stanford University Press.

10.

Graham v. Florida, 130 S. Ct. 2011 (2010).

11.

Haviland

A. M.

Nagin

Rosenbaum

Tremblay

. 2008. “Combining Group-based Trajectory Modeling and Propensity Score Matching for Causal Inferences in Nonexperimental Longitudinal Data.” Developmental Psychology 44:422–36.

12.

Kandel

Yamaguchi

. 1999. “Developmental States of Involvement in Substance Use.” Pp. 50–74 in Sourcebook on Substance Abuse: Etiology, Epidemiology, Assessment, and Treatment, edited by Ott

Tarter

Ammerman

. Boston, MA: Allyn & Bacon.

13.

Loeber

Burke

Pardini

. 2009. “Development and Etiology of Disruptive and Delinquent Behavior.” Annual Review of Clinical Psychology 5:291–310.

14.

Miller v. Alabama, 132 S. Ct. 2455 (2012).

15.

Moffitt

1993. “Adolescence-limited and Life-course Persistent Antisocial Behavior: A Developmental Taxonomy.” Psychological Review 100:674–701.

16.

Monahan

Steinberg

Cauffman

. 2009. “Affiliation with Antisocial Peers, Susceptibility to Peer Influence, and Antisocial Behavior during the Transition to Adulthood.” Developmental Psychology 45:1520–30.

17.

Mulvey

Steinberg

Fagan

Cauffman

Piquero

Chassin

Knight

Brame

Schubert

Hecker

Losoya

. 2004. “Theory and Research on Desistance from Antisocial Activity among Serious Adolescent Offenders.” Youth Violence and Juvenile Justice 2:213–36.

18.

Muthen

. 2000. “Integrating Person-centered and Variable-centered Analyses: Growth Mixture Modeling with Latent Trajectory Classes.” Alcoholism: Clinical and Experimental Research 24:882–91.

19.

Nagin

2005. Group-based Modeling of Development. Cambridge, MA: Harvard University Press..

20.

Oyserman

Markus

. 1990. “Possible Selves and Delinquency.” Journal of Personality and Social Psychology 59:112–25.

21.

Piquero

A. R.

2008. “Taking Stock of Developmental Trajectories of Criminal Activity over the Life Course.” Pp. 22–78 in The Long View of Crime: A Synthesis of Longitudinal Research, edited by Liberman

. New York: Springer.

22.

Roper v. Simmons, 543 U.S. 551 (2005).

23.

Rothman

1980. Conscience and Convenience: The Asylum and Its Alternatives in Progressive America. Boston, MA: Little, Brown.

24.

Sampson

Laub

. 2005. “Seductions of Method: Rejoinder to Nagin and Tremblay’s “Developmental Trajectory Groups: Fact or Fiction?”” Criminology 43:905–13.

25.

Scott

Steinberg

. 2010. Rethinking Juvenile Justice. Cambridge, MA: Harvard University Press.

26.

Steinberg

2010. “A Dual-systems Model of Adolescent Risk Taking.” Developmental Psychobiology 52:216–24.

27.

Sullivan

Hamilton

. 2007. “Exploring Careers in Deviance: A Joint Trajectory Analysis of Criminal Behavior and Substance Use in an Offender Population.” Deviant Behavior 28:497–523.

28.

Thornberry

Krohn

, eds. 2003. Taking Stock of Delinquency: An Overview of Findings from Contemporary Longitudinal Studies. New York: Plenum.

29.

Woolard

J. L

. 2012. “Adolescent Development, Delinquency, and Juvenile Justice.” Pp. 107–122 in The Handbook of Juvenile Delinquency and Juvenile Justice, edited by Feld

Bishop

. New York: Oxford University Press.

30.

Zimring

1998. American Youth Violence. New York: Oxford University Press.