Laboratory Reports from the Center on Antisocial Drug Dependence

Quantifying Adolescent Behavioral Disinhibition for Neuroimaging Studies

Susan E. Young*, PhD, Susan K. Mikulich-Gilbertson, PhD, Joseph T. Sakai, MD, Thomas J. Crowley, MD

*Mail Stop F478 12469 East 17th Place Aurora, CO 80045 United States of America

Risky behaviors are behaviors that may result, unpredictably, in either rewards or punishments. They may be socially approved or contrary to social norms and laws (e.g., a boxing match, vs. a street fight). Adolescents as a group do more risky behaviors than adults [1], and imaging of adolescents' brains suggests that age-related events in brain development may underlie that elevated risk-taking [2]. However, compared with other adolescents, those with conduct disorder (CD) [3] do more --and more dangerous --risky behaviors (including frequent fighting, fights with weapons, robberies, break-ins, etc.) [4]. Moreover, many youths who take those risks also take risks with intoxicating substances, so adolescent CD and substance use disorders (SUD) [3] frequently co-occur [5]. In laboratory tests [6,7], as in real life, youths with CD and SUD do more risky behaviors than other adolescents.

Risk-taking is part of a larger complex of interrelated problem behaviors conceptualized as “behavioral disinhibition” (BD). We define BD as a lifelong behavioral disposition that involves excessive pursuit of exciting, risky, appetitive stimuli coupled with an unusual disregard of aversive consequences for such behavior. BD is a latent construct that is assumed to have an underlying continuum of liability ranging from absent to severe among different persons; many youths with severe BD meet criteria for both CD and SUD [8].

We developed this construct through a study of comorbidity in a community-based sample of adolescent twins [9,10]. For these youth, we operationalized BD using summary data collected with structured, self-report, psychiatric interviews and self-report personality questionnaires.  Age-adjusted, lifetime symptom counts for DSM-IV [3] CD and attention­deficit/hyperactivity disorder (ADHD), derived from the Diagnostic Interview Schedule for Children (DISC 4.1 [11]), as well as a count of substances (both licit and illicit) used repeatedly, derived from the Composite International Diagnostic Interview – Substance Abuse Module (CIDI-SAM [12]) provided three of our four BD measures. Additionally, we included a summary score of self-reported novelty seeking based on Tridimensional Personality Questionnaire (TPQ [13]). Using a structural equation model designed specifically for genetically informative samples, the covariance among these four measures (a ‘common BD factor’) was parsed into genetic and environmental components. Results suggested that the heritability of BD exceeded .80 [9]. A strikingly similar study conducted at the University of Minnesota confirmed these findings in an independent (and somewhat older) twin sample, using symptoms of substance use disorders, CD, antisocial personality symptoms, and a questionnaire measure of constraint [14,15]. Further analysis in our longitudinal twin sample demonstrated a significant genetic correlation between BD and laboratory-based indices of executive function, particularly cognitive inhibition [11].

Quantifying BD for Neuroimaging Studies

Because of the high cost and subject burden of neuorimaging procedures used for research, we generally study small samples (n ≈ 20-30) [16].  In order to ensure that our composite BD scores could be scaled relative to individuals in the general population in the same age range, we constructed our BD composite from four measures that were utilized in both the neuroimaging samples as well as a large community-based sample of adolescents aged 14-18 years [10], who were previously assessed using the same instruments. We operationalized BD as the composite of 4 measures including:

(a) Lifetime symptom count for substance abuse and dependence (summed across 10 drug classes) measured with the Substance Abuse Module of the CIDI [12].

(b) Lifetime CD symptom count (measured with DISC-4.1 [11]).

(c) Child Behavior Checklist [17] Inattention problem score based on parent-endorsed items chosen to mimic DSM-IV ADHD symptoms of inattention (Item #s: 8, poor concentration; 13, confused/”in a fog”; 17, daydreams; 61, poor schoolwork; 80, stares blankly). For one case where CBCL information was missing, comparable items from the youth-reported YSR were substituted.

(d) Child Behavior Checklist [17] Hyperactivity/Impulsivity problem score based on parent-endorsed items chosen to mimic DSM-IV ADHD symptoms of hyperactivity/impulsivity (Item #s: 1, acts too young; 10, restless/can’t sit still; 36, accident prone; 41, acts without thinking; 45, nervous; 46, twitching; 62, poor coordination/clumsy; 93, talks too much; 104, unusually loud). Again, if CBCL information was missing, comparable items from the youth-reported YSR were substituted.

We computed means and standard deviations for each behavioral measure in 372 male and 414 female adolescents.  Using standard principal component analysis (PCA) we extracted the maximum covariance among the 4 behavioral measures in the community sample. This produces a standardized factor loading for each behavioral measure. Results are tabulated separately (below) boys and girls, followed by a combined sample of 677 adolescents. The combined sample consists of only 677 individuals because the109 females who had a male sibling in the boys-only sample were removed in order to maintain a sample of individuals from unique families.

Boys (N=372) Min Max  Mean Std. Dev. Factor Wgt.
Abuse/Depend Symptoms 0 22 1.403 3.360 .365
CD Symptom Count 0 12 1.253 1.695 .342
CBCL Attention Score 0 10 1.255 1.744 .368
CBCL Hyper/Impulse Score 0 13 1.702 2.284 .364
Interview Age in Years 14 18 16.13 1.240 ----

Girls (N=414) Min Max  Mean Std. Dev. Factor Wgt.
Abuse/Depend Symptoms 0 26 1.097 3.353 .313
CD Symptom Count 0 7 0.655 0.976 .346
CBCL Attention Score 0 9 0.853 1.436 .404
CBCL Hyper/Impulse Score 0 15 1.469 2.077 .417
Interview Age in Years 14 18 16.20 1.245 ----

Combined* (N=677) Min Max  Mean Std. Dev. Factor Wgt.
Abuse/Depend Symptoms 0 26 1.291 3.423 .349
CD Symptom Count 0 12 0.959 1.442 .339
CBCL Attention Score 0 10 1.087 1.646 .372
CBCL Hyper/Impulse Score 0 15 1.610 2.197 .378
Interview Age in Years 14 18 16.24 1.200 ----

For each neuroimaging subject we express his/her scores on the four behavioral measures as deviations (SD units) from the relevant (i.e., male, female, or mixed gender) community-sample mean. We then generate a composite BD score for each neuroimaging subject by summing his/her factor-weighted scores on each of the four behavioral measures. We present the imaging subjects’ BD scores as plots in standard deviation units (z-scores) from the mean BD score (standardized to 0) of the relevant community sample.


1. Steinberg L (2007) Risk taking in adolescence: New perspectives from brain and behavioral science. Curr Dir Psychol Sci 16: 55-59.

2. Somerville LH, Hare T, Casey BJ (2011) Frontostriatal maturation predicts cognitive control failure to appetitive cues in adolescents. J Cogn Neuroscience 23: 2123-2134

3. American Psychiatric Association (2000) Diagnostic and Statistical Manual of Mental Disorders 4th Edition, Text Revision. Washington DC: American Psychiatric Press.

4. Crowley TJ, Mikulich SK, Ehlers KM, Whitmore EA, Macdonald MJ (2001) Validity of structured clinical evaluations in adolescents with conduct and substance problems. J Am Acad Child Adolesc Psychiatry 40: 265-273.

5. Disney ER, Elkins IJ, McGue M, Iacono WG (1999) Effects of ADHD, conduct disorder, and gender on substance use and abuse in adolescence. Am J Psychiatry 156: 1515­1521.

6. Crowley TJ, Raymond KE, Mikulich-Gilbertson SK, Thompson LL, Lejuez CW (2006) A risk-taking “set” in a novel task among adolescents with serious conduct and substance problems. J Am Acad Child Adolesc Psychiatry 45: 175-183.

7. Lane SD, Cherek DR (2001) Risk taking by adolescents with maladaptive behavior histories. Exp Clin Psychopharmacol 9: 74-82.

8. Iacono WG, Malone SM, McGue M (2008) Behavioral disinhibition and the development of early-onset addiction: common and specific influences. Annu Rev Clin Psychol 4: 325­348.

9. Young SE, Stallings MC, Corley RP, Krauter KS, Hewitt JK (2000) Genetic and environmental influences on behavioral disinhibition. Am J Med Genet B Neuropsychiatr Genet 96: 684-695.

10. Rhea SA, Gross AA, Haberstick BC, Corley RP (2006) Colorado Twin Registry. Twin Research and Human Genetics 9: 941-949.

11. Shaffer D, Fisher P, Lucas CP, Dulcan MK, Schwab-Stone ME (2000) NIMH Diagnostic Interview Schedule for Children Version IV (NIMH DISC-IV): Description, differences from previous versions, and reliability of some common diagnoses. J Am Acad Child Adolesc Psychiatry 39: 28-38.

12. Compton WM, Cottler LB, Dorsey KB, Spitznagel EL, Mager DE (1996) Comparing assessments of DSM-IV substance dependence disorders using CIDI-SAM and SCAN. Drug Alcohol Depend 41: 179-187.

13. Cloninger, C.R. (1987). The Tridimensional Personality Questionnaire: Version 4. St. Louis: Department of Psychiatry, Washington University School of Medicine.

14. Krueger RF (2002) Etiologic connections among substance dependence, antisocial behavior, and personality: Modeling the externalizing spectrum. J Abnorm Psychol 111: 411-424.

15.  Hicks BM, Krueger RF, Iacono WG, McGue M, Patrick CJ (2004) Family transmission and heritability of externalizing disorders: a twin-family study. Arch Gen Psychiatry 61: 922-928.

16. Crowley TJ, Dalwani MS, Mikulich-Gilbertson SK, Du YP, Lejuez CW et al. (2010) Risky decisions and their consequences: neural processing by boys with antisocial substance disorder. PloS ONE 5:e12835 doi:10.1371/journal.pone.0012835.

17.  Achenbach TM, Howell CT, Quay HC, Conners CK (1991) National survey of problems and competencies among four-to sixteen-year-olds: Parents' reports for normative and clinical samples. Monogr Soc Res Child Dev 56: 1-131.