This study—Effects of Cognitive Control Training among Adolescent Offenders—took place at two Youth Development Centers in North Carolina, which include male and female adolescents adjudicated for antisocial behavior. It examined how cognitive control relates to social information processing, substance use histories, and serious conduct problems. It also examines the effect of a cognitive enhancement training task on trained and untrained cognitive and social information processing skills. Knowledge from this study will pinpoint novel avenues for prevention studies that might use cognitive enhancement training as a technique to prevent substance use and disruptive behavior disorders.

This study aligns well with the Transdisciplinary Prevention Research Center (C-StARR)’s overriding goal to translate evidence from basic-science research on regulatory processes into novel research projects that prevent substance use and other conduct problems in adolescence. It also promotes a novel translation and integration of a cognitive developmental science theory with the prevention, intervention, and developmental psychopathology literature. Furthermore, this study provides new evidence at the interface of these disciplines, and it will generate new studies that will contribute to the dissemination of promising evidence-based practices that will inform practitioners and prevention efforts.

Prior to C-StARR funding, we secured Institutional Review Board approval from both Duke University and the Department of Juvenile Justice and Delinquency Prevention to conduct the project. The first three months of the project were used to pilot test the protocol, develop a relationship with facility staff members, and train research assistants (RAs). Starting in July 2010, we recruited participants by meeting with all youth in each of three cottages at the facilities. Recruitment sessions occurred three times during the project. At these sessions, participants were told about the details of the study and youth who wanted to learn more about participating the study put their name on a list that was used by research assistants to then decide which parents to meet with to obtain consent. We arranged a consent process with the facilities where RAs meet with parents just before or after monthly treatment planning meetings are conducted at the facilities. The majority of parents have also participated in the parent part of our study in which they answered questions about their child’s behaviors. Once a parent gave his (her) consent for their child to participate in our study, we met individually with the youth to obtain his /her consent. Once the youth consented to participate in the study, he/she began phase 1. Participants:

1. completed numerous computer tasks that measure cognitive control (i.e., attention, working memory, and response inhibition),
2. answered questions about mental health disorders, drug/alcohol use, and offending behaviors, and
3. completed achievement and intelligence tasks.

This part of the study took 4 hours to complete and was done over the course of one week. Once the participant completed Phase 1, he/she began the second phase. Here, participants completed computerized tasks 5 days a week for 5 weeks. Participants were randomly assigned to receive either a cognitive training or an active control task. Task difficulty in the cognitive enhancement training task was individually adjusted based on each participant’s accuracy rate in pre-specified trials. The active control condition does not train any specific cognitive control skill and task difficulty does not adjust based on individual’s performance. The final phase of the study is a post-assessment, which is administered after all training/control sessions are completed to assess whether cognitive training influences their cognitive control and social information processing abilities. We have limited testing times with youth at both facilities, amounting to approximately 2 hours of testing time per day 5 days a week. As a result, testing has gone slightly slower than expected.

At the time of this report, we successfully collected data from 30 youth and 17 parents for a total of 47 participants. Thirteen of the 30 youth participants have completed all of the phases. All of our activities the last 3 months have focused on collecting data from two sites (boys and girls), which has been a full-time endeavor.

We have analyzed data on 7 of those 13 participants who completed all phases of the study. Thus, our results are very preliminary. We have found that our measures of cognitive control relate to meaningful constructs as predicted (e.g., intelligence, working memory). We also have preliminary evidence that our measures of cognitive control related in meaningful ways to social-information processes (i.e., hostile attribution bias). Furthermore, we found that mean pre- and post-test scores did not differ based on training condition; however, this information is based on only 1 participant in the active control condition. We cannot yet make any conclusive remarks about the effect of our training tasks given the very small sample size.

Although C-StARR funding for this project has ended, the Center for Child and Family Policy is providing funding to continue to recruit, consent, and test participants at both testing sites. We anticipate that data collection will end in September, and subsequently we will begin data cleaning and analyses. We will prepare a manuscript on the results to be submitted during the fall of 2011. We anticipate using data from this study as preliminary data in a grant application to NIH or NSF in the winter of 2012 that will more extensively examine the relations between cognitive control, social-information processing, and cognitive training tasks.