Characterizing the Dynamical Complexity Underlying Meditation

Abstract

Over the past 2,500 years, contemplative traditions have explored the nature of the mind using meditation. More recently, neuroimaging research on meditation has revealed differences in brain function and structure in meditators. Nevertheless, the underlying neural mechanisms are still unclear. In order to understand how meditation shapes global activity through the brain, we investigated the spatiotemporal dynamics across the whole-brain functional network using the Intrinsic Ignition Framework. Recent neuroimaging studies have demonstrated that different states of consciousness differ in their underlying dynamical complexity, i.e., how the broadness of communication is elicited and distributed through the brain over time and space. In this work, controls and experienced meditators were scanned using functional magnetic resonance imaging (fMRI) during resting-state and meditation (focused attention on breathing). Our results evidenced that the dynamical complexity underlying meditation shows less complexity than during resting-state in the meditator group but not in the control group. Furthermore, we report that during resting-state, the brain activity of experienced meditators showed higher metastability (i.e., a wider dynamical regime over time) than the one observed in the control group. Overall, these results indicate that the meditation state operates in a different dynamical regime compared to the resting-state.

Keywords: dynamical complexity; fMRI; ignition; integration; meditation; resting-state; whole-brain.

Figure 1

Measuring intrinsic ignition. (A) Events were captured applying a threshold method Tagliazucchi et al. (2012) (see green area). For each event elicited (gray area), the activity in the rest of the network was measured in the time-window of 4TR (see red area). (B) A binarized matrix was obtained representing the synchronized events in each time window (i.e., when two brain areas have triggered an event) (C) Applying the global integration measure Deco et al. (2015), we obtained the largest subcomponent. By repeating the process for each driving event, we calculated the mean and the variability of the Intrinsic-Driven Integration for each brain area across the whole-brain network. Adapted from Deco and Kringelbach (2017).

Figure 2

(A) Mean of the Intrinsic-Driven Integration (IDMI) for each group during resting-state and meditation state. The IDMI was higher in meditators than in controls during resting-state and lower in meditators during meditation. No significant differences were observed in controls between conditions. Furthermore, we show the box plot from the surrogate IDMI data (on the bottom in green). The randomized data were significantly smaller than the original time-series, showing the robust statistical comparisons. (B) Both controls and meditators showed higher local metastability across the whole-brain during resting-state compared to meditation. However, the effect was significantly larger for meditators. Furthermore, the metastability in resting-state was significantly higher for experienced meditators than for controls. P-values are based on Monte-Carlo simulation after Bonferroni correction, *p ≤ 0.025, ***p ≤ 0.0005 and n.s represents not significant. (C) IDMI across events for each group during resting-state and meditation for 268 brain regions.