Teen Brain Training (TBT)

This phased innovation application in response garnered support for an initial (R61) 2-year phase for milestone-driven testing of neural targets of intervention by a novel neurofeedback treatment. Using neurofeedback (NF) we targeted the neurocircuitry of affect regulation and self-processing in adolescents and young adults (ages 11-18) with current significant suicide ideation and a history of a recent suicide attempt. Attaining our the milestones would trigger support for three additional years (R33 phase) to confirm target engagement in a larger sample with random assignment to active NF intervention vs. Placebo NF, to assess the relationships between target engagement and changes in functional outcomes. Affect dysregulation and abnormal self-processing are known to contribute to poor long-term outcomes and predict repeated suicide attempts in at risk populations. They are insufficiently addressed by medications or most existing behavioral treatments. NF training has resulted in enduring improvements in those dimensions in prior Placebo controlled NF trials. To further develop this intervention, we first identified the neural target best engaged by NF during a critical developmental period for affect regulation and self-processing. Our pilot data and theoretical considerations supported the overarching hypothesis that increased amygdala or dACC activity and their functional connectivity (FC) with the middle prefrontal cortex (mPFC) represented treatable targets via NF training. Accordingly, in the initial R61 phase measured intervention-related increases in dACC and right amygdala activity and their FC with mPFC during self-processing and affect regulation tasks with high-quality imaging and clinical data obtained during, before and after NF training. The goal of the R61 phase was to determine which loci is best up-regulated by the targeted population and was best associated with amelioration in the target functional outcomes. If we were to meet our proposed milestones that the increase in dACC or right amygdala and their FC with the mPFC exceeded a specific effect size and account for an appreciable proportion of the variance in behavioral measures of affect regulation and self-processing, we would proceed to the R33 phase. In the R33 phase, we would expand the study to test adolescents randomized to the active NF intervention (dACC or Amygdala) or to a placebo NF who would provide complete high quality pre- and post-intervention and clinical data. We obtained state-of-the-art magnetic resonance imaging (MRI) data with a primary focus on functional MRI. Our specific aims in the R33 phase would be to confirm target engagement in the groups randomized to active vs. the Placebo NF group; to examine the relationship between changes in the neural target and clinical improvements in affect regulation, self-processing and suicide ideation; and to identify mediators of improved functional outcomes. Evidence supporting the validity of dACC or amygdala circuits as a modifiable neural target would then support future studies to further enhance the effectiveness of neurofeedback in adolescents. Importantly, even negative results would be informative regarding the location and direction of neurofeedback (top=dACC vs. down=amygdala) that best attains substantial effects. Inconclusive results from brain imaging would still inform Bayesian priors for future studies of the neural substrates of affect regulation and self-processing impairments and their amelioration.