A Network Investigation of Core Mindfulness Processes

Abstract

Objective: Mindfulness has been associated with improved psychological well-being and health, although outcomes from mindfulness-based interventions are mixed. One challenge is a limited understanding about which specific processes are core to mindfulness. Network analysis offers a method to characterize the core processes of mindfulness.

Methods: This study used network analysis to identify which processes are central to mindfulness (have the strongest connectivity with other mindfulness processes) as measured by the Five Facets Mindfulness Questionnaire- Short Form, analyzed at the item-level. Data were obtained from baseline of a randomized clinical trial of smartphone app-based mindfulness training for smoking cessation.

Results: The most central processes in the mindfulness network included, "I think some of my emotions are bad or inappropriate and I shouldn't feel them," an aspect of Nonjudgment/acceptance; as well as "I can easily put my beliefs, opinions, and expectations into words," and "It's hard for me to find the words to describe what I'm thinking," aspects of Describing.

Conclusions: Findings help to clarify which processes are to mindfulness, contributing to a better understanding of the definition of mindfulness, and suggest factors that may be promising to target in mindfulness-based interventions. Future research should examine if mindfulness-based interventions may be improved by targeting these core mindfulness processes.

Keywords: acceptance; awareness; mindfulness; network analysis; network theory.

Figure 1.

Mindfulness network. Thicker lines represent stronger relationships

Figure 2.

Centrality plots. Strength (left) and expected influence (right) values of each node are presented. Centrality indices are shown as standardized z-scores, ordered by highest value.

Figure 3.

Network Stability. a) Bootstrapped confidence intervals of estimated edge-weights. The red line indicates the raw values, the black line the bootstrapped values, and the gray area the bootstrapped CIs. Network edges are ordered based on edge-weights, from highest to lowest. The y-axis labels have been removed for clarity. b) Sequential average correlations between centrality indices of random subsets and the original sample. c) Bootstrapped difference tests (α = 0.05) between edge weights that were non-zero in the estimated network

Figure 3.

Network Stability. a) Bootstrapped confidence intervals of estimated edge-weights. The red line indicates the raw values, the black line the bootstrapped values, and the gray area the bootstrapped CIs. Network edges are ordered based on edge-weights, from highest to lowest. The y-axis labels have been removed for clarity. b) Sequential average correlations between centrality indices of random subsets and the original sample. c) Bootstrapped difference tests (α = 0.05) between edge weights that were non-zero in the estimated network

Figure 3.

Network Stability. a) Bootstrapped confidence intervals of estimated edge-weights. The red line indicates the raw values, the black line the bootstrapped values, and the gray area the bootstrapped CIs. Network edges are ordered based on edge-weights, from highest to lowest. The y-axis labels have been removed for clarity. b) Sequential average correlations between centrality indices of random subsets and the original sample. c) Bootstrapped difference tests (α = 0.05) between edge weights that were non-zero in the estimated network

Figure 4.

Bootstrapped strength (left panel) and expected influence (right panel) difference results. Black boxes indicate nodes that differ significantly from one another, and grey boxes represent nodes that do not differ significantly from one another, and white boxes show node strength.