Interoceptive brain network mechanisms of mindfulness-based training in healthy adolescents

Abstract

Introduction: This study evaluated changes in the white matter of the brain and psychological health variables, resulting from a neuroscience-based mindfulness intervention, the Training for Awareness, Resilience, and Action (TARA), in a population of healthy adolescents.

Methods: A total of 100 healthy adolescents (57 female, age ranges 14-18 years) were randomized into the 12-week TARA intervention or a waitlist-control group. All participants were imaged with diffusion MRI to quantify white matter connectivity between brain regions. Imaging occurred at baseline/randomization and after 12 weeks of baseline (pre- and post-intervention in the TARA group). We hypothesized that structural connectivity in the striatum and interoceptive networks would increase following the TARA intervention, and that, this increased connectivity would relate to psychological health metrics from the Strengths and Difficulties Questionnaire (SDQ) and the Insomnia Severity Index (ISI). The TARA intervention and all assessments, except for the MRIs, were fully remotely delivered using secure telehealth platforms and online electronic data capture systems.

Results: The TARA intervention showed high consistency, tolerability, safety, recruitment, fidelity, adherence, and retention. After 12 weeks, the TARA group, but not controls, also demonstrated significantly improved sleep quality (p = 0.02), and changes in the right putamen node strength were related to this improved sleep quality (r = -0.42, p = 0.006). Similarly, the TARA group, but not controls, had significantly increased right insula node strength related to improved emotional well-being (r = -0.31, p = 0.04). Finally, we used the network-based statistics to identify a white matter interoception network that strengthened following TARA (p = 0.009).

Discussion: These results suggest that the TARA mindfulness-based intervention in healthy adolescents is feasible and safe, and it may act to increase structural connectivity strength in interoceptive brain regions. Furthermore, these white matter changes are associated with improved adolescent sleep quality and emotional well-being. Our results suggest that TARA could be a promising fully remotely delivered intervention for improving psychological well-being in adolescents. As our findings suggest that TARA affects brain regions in healthy adolescents, which are also known to be altered during depression in adolescents, future studies will examine the effects of TARA on depressed adolescents.

Clinical trial registration: https://clinicaltrials.gov/study/NCT04254796.

Keywords: MRI; adolescence; brain connectivity; interoception; mindfulness; mindfulness practices such as meditation.

Figure 1

CONSORT diagram.

Figure 2

Changes in ISI and SDQ. We show the changes in (A) the Insomnia Severity Index (ISI) and (B) the Strengths and Difficulties Questionnaire (SDQ--emotional problems subscale). The ISI shows a significant reduction in the TARA group, indicating that the TARA mindfulness intervention significantly improved sleep quality in adolescents. The change in SDQ was not significant.

Figure 3

Changes in structural connectivity node strength. We show violin plots of all evaluated regions, which included the (A) left caudate, (B) right caudate, (C) left putamen, (D) right putamen, (E) left insula, (F) right insula, (G) left anterior cingulate cortex (ACC), and (H) right ACC. Only the right caudate (B) showed a significant increase in node strength following TARA.

Figure 4

TARA effect of regions on sleep and emotional well-being. We show the significant group-level effects between the change in SDQ and (A) the right insula and (B) the left ACC. We also show the significant group-level effect between the changes in (C) the ISI and right putamen.

Figure 5

Interoception network strengthening from TARA. (A) We show the significant subnetwork identified with the network-based statistic (NBS). Each row/column is labeled with the according ROI. (B) The total change in the subnetwork (sum of all edge changes) reflecting the significant difference identified by NBS. (C) The post-hoc analysis finding a significant correlation between the interoception subnetwork change across all study subjects and improved sleep quality (i.e., decreased ISI). (D) A visualization of the interoception brain network using Brian Net Viewer (Xia et al., 2013). Edge thickness is weighted by the test statistic.