Published online by Cambridge University Press: 07 July 2008
Children's behavior problems may stem from ineffective cortical mechanisms for regulating negative emotions, and the success of interventions may depend on their impact on such mechanisms. We examined neurophysiological markers associated with emotion regulation in children comorbid for externalizing and internalizing problems before and after treatment. We hypothesized that treatment success would correspond with reduced ventral prefrontal activation, and increased dorsomedial prefrontal activation, at the time point of an event-related potential (ERP) associated with inhibitory control. Twenty-seven 8- to 12-year-old children (with usable data) were tested before and after a 14-week community-based treatment program and assessed as to improvement status. Fifteen 8- to 12-year-olds from the normal population (with usable data) were tested over the same interval. All children completed an emotion-induction go/no-go task while fitted with a 128-channel electrode net at each test session. ERP amplitudes, and estimates of cortical activation in prefrontal regions of interest, were measured at the peak of the “inhibitory” N2 and compared between improvers, nonimprovers, and nonclinical children. ERP amplitudes showed no group differences. However, improvers showed an overall reduction in ventral prefrontal activation from pretreatment to posttreatment, bringing them in line with nonclinical children, whereas ventral activation remained high for nonimprovers. Both improvers and nonimprovers showed high dorsal activation relative to nonclinical children. Supplementary analyses indicated that only ventral prefrontal regions, and only within the N2 time window, showed decreased activity from pre- to posttreatment, suggesting changes in regulatory processes rather than in overall emotional arousal. These cortically mediated changes may permit a reduction in the overengaged, rigid style of emotion regulation characteristic of children with behavior problems.
We gratefully acknowledge the financial support provided by the Canadian Institutes for Health Research (to M.D.L. and I.G.) and the Natural Sciences and Engineering Research Council of Canada (to M.D.L.). We thank Sidney J. Segalowitz for his suggestions for analyzing and interpreting the data and the scientists at Electrical Geodesics Inc. for their continuing support and advice.