Self-specific processing in the meditating brain: a MEG neurophenomenology study

Abstract

Self-specific processes (SSPs) specify the self as an embodied subject and agent, implementing a functional self/nonself distinction in perception, cognition, and action. Despite recent interest, it is still undetermined whether SSPs are all-or-nothing or graded phenomena; whether they can be identified in neuroimaging data; and whether they can be altered through attentional training. These issues are approached through a neurophenomenological exploration of the sense-of-boundaries (SB), the fundamental experience of being an 'I' (self) separated from the 'world' (nonself). The SB experience was explored in collaboration with a uniquely qualified meditation practitioner, who volitionally produced, while being scanned by magnetoencephalogram (MEG), three mental states characterized by a graded SB experience. The results were then partly validated in an independent group of 10 long-term meditators. Implicated neural mechanisms include right-lateralized beta oscillations in the temporo-parietal junction, a region known to mediate the experiential unity of self and body; and in the medial parietal cortex, a central node of the self's representational system. The graded nature as well as the trainable flexibility and neural plasticity of SSPs may hold clinical implications for populations with a disturbed SB.

Keywords: : self-specific processes; MEG; beta band; meditation; minimal self; neurophenomenology; parietal cortex.

Figure 1

Experimental setup for Stage 1 (top) and Stage 2 (bottom). SB, sense of boundaries; NS, narrative self-mode; MS, minimal self-mode; SL, selfless mode. All epochs were initiated by an auditory cue.

Figure 2

Sensor-level results. Determining FOI: (a) Frequencies (x-axis) regression t-values (y-axis) plot, averaged overall sensors. Red circles indicate statistically significant t-values (P<0.0005, FDR corrected); Statistical bar plot: (b) Mean power (y axis) and standard error bars averaged over FOI (22-33 Hz) and SOI for SB1, SB2, and SB3; Raw effect: (c) of percent-in-signal-change (psc) between SB1 and SB2 (left) and SB2 and SB3 (right). Color bar indicates psc from 0.2 (dark red) to –0.2 (dark blue); Determining SOI: (d) 2D regression t-map averaged over the FOI (22–33 Hz). Dots on the map represent sensors; stars signify significant sensors (P<0.0005, Monte Carlo permutation corrected). Color bar scale indicates t-values from 0.6 (dark red) to –0.6 (dark blue).* P<0.0335; ** P<1.07 x 10⁻⁷ (both Bonferroni corrected).

Figure 3

Beamforming beta band source estimation statistical images for subject S. Lateral (A1 and B1) and medial (A2 and B2), left (A1 and A2) and right (B1 and B2), views of S’s source estimates overlaid on SUMA 3D cortical surface model. Color bar indicates t-value degree from 6 (dark red) indicating a positive linear pattern to − 6 (dark blue) indicating a negative linear pattern. Images significant at P<0.0005 (Monte Carlo Permutation corrected). Lateral views (top) highlight the TPJ regions in both hemispheres; while the medial views (bottom) highlights the Prc and M/PCC gyrus bilaterally, and the SMA in the right hemisphere. For more detailed anatomical information, refer to Table 2.

Figure 4

Beamforming beta band source estimation statistical images for meditators group (n = 10). Lateral (A1 and B1) and medial (A2 and B2), left (A1 and A2) and right (B1 and B2), views of the meditators group (n=10) source estimates overlaid on SUMA 3D cortical surface model. Color bar indicates t-value degree from 6 (dark red) indicating a positive linear pattern to − 6 (dark blue) indicating a negative linear pattern. Images significant at P<0.0005 (Monte Carlo permutation corrected). No results were found in the left hemisphere (A1 and A2). Right lateral view (B1) highlights the TPJ region; while the right medial view (B2) highlights the M/PCC and the Prc. For more detailed anatomical information, refer to Table 3.