RESUMO
BACKGROUND: There is growing recognition that violence and other forms of conduct problems increase during adolescence. The exact relationship between biological, psychological, and social variables has not been defined yet. OBJECTIVES: To analyze whether Intelligence Quotients (IQS), neurological history, child behavioral problems, executive functions, and soft neurological signs (SNS) can differentiate between undisciplined and unreliable adolescents (Behavioral Dysregulation Disorder subjects, BDD) and normal controls. METHOD: Twenty-five 13 to 16-year-olds, adolescents with BDD and 25 matched controls were used in this study. WISC-R, executive function assessment, neurological history, child behavioral problems, and SNS scores were analyzed using a Multivariate Analysis of Variance (MANOVA). A Multiple Regression Stepwise with Criteria Probability of F Analysis was used for predicting criteria variable variance. RESULTS: WISC-R Verbal IQ (VIQ), Information, Similarities, and Vocabulary subtests presented statistically significant differences between BDD and controls (p < .001). No Performance IQ (PIQ) variables established significant differences between both groups. Executive function scores did not detect significant differences between groups either. Prenatal, neonatal, and neurological history scores were similar between both groups. Two child behavioral problem variables were significantly different, with higher scores in BDD group: use of weapons and drug-use (p < .05). A Multiple Regression Stepwise (Criteria Probability of F < .05) model, entering the predictive variables in each domain (intelligence, executive function, neurological antecedents, child behavioral problems, and SNS), and using the score on the criteria variable as dependent variable, found two predictive models: (1) WISC-R Information (Ad-R-SQ = 0.172 F-Ch. = 11.176, p < .01); and (2) WISC-R Information and drug-use (R-SQ: 0.26; F-Ch = 9.605 p < .001). CONCLUSIONS: A verbal factor and drug-use predicted fairly 30% of the variance of the criteria variable used for classifying adolescents with BDD. These results would mean that a language underlying factor and an environmental drug-use factor would be related to the BDD in adolescents.