The Neuroscience of Mindfulness: Effects on the Brain as a Whole written by Richik Neogi

            In the previous article, we put forth a model of consciousness in the context of mindfulness in which there are three networks in the brain that communicate with one another heavily. These do not constitute all the large-scale brain networks within the brain, but these three networks are the most relevant for our discussion on mindfulness. They are the salience network (SN), the default mode network (DMN), and the task-positive network (TPN), the latter of which consists of a few subnetworks (among them are the frontoparietal control network or FPCN and the dorsal attention network or DAN). In the previous article, we put forth the hypothesis that mindfulness strengthens the ability of the salience network to control our ability to effortlessly switch between activity in the DMN and the TPN via the SN. In this article, we will explore this hypothesis further, providing evidence for it from the literature. In addition to the effects of mindfulness training on the function of the three networks, we will also go over evidence suggesting that mindfulness training can alter the structure of the brain on a macroscopic level. Before we do our deep dive into how mindfulness affects large-scale brain networks, it is important to highlight how researchers “measure” mindfulness, as this informs researchers regarding pre-existing mindful tendencies as well as potential changes in these tendencies because of mindfulness training.

          Trait mindfulness refers to dispositional mindful qualities. In other words, it refers to one’s tendency to be mindful as an attribute of personality. It can be contrasted with state mindfulness, which refers to context-dependent mindful qualities (i.e. during mindfulness meditation). Trait mindfulness is usually measured using psychometric tests or questionnaires – as a result, measures of trait mindfulness are self-reported. There are several psychometric tests that researchers can use to evaluate trait mindfulness. Two major questionnaires that are used to evaluate trait mindfulness are the Five Facet Mindfulness Questionnaire (FFMQ)1 and the Mindful Attention and Awareness Scale (MAAS)2. The former provides a more holistic measure of one’s dispositional mindfulness (the score can be divided into five sub-scores, each corresponding to a facet of mindfulness) while the latter measures the attentional dimension of mindfulness, giving only a single score. These are not the only psychometric tests that evaluate trait mindfulness, but they are the most encountered in the literature. Some studies measure trait mindfulness in individuals who do not receive any mindfulness training (individuals naïve to mindfulness practices in other words) and draw correlations between scores in the psychometric measures and functional connectivity as measured by resting-state functional connectivity fMRI (Parkinson et al., 2019; Bilevicius et al., 2018 – these articles are reviewed in Sezer et al., 2022). In brief, both studies in mindfulness-naïve individuals noted decreased connectivity within the DMN with increasing mindfulness scores. Another finding from Bilevicius et al., 2018 was that both the FPCN and the DMN displayed increased connectivity with a node in the ventral attentional network (the ventral counterpart of the dorsal attention network, hereon abbreviated VAN which is also called the mid-cingulo insular network), the right middle frontal gyrus, in individuals who scored highly on the MAAS. We have not considered the VAN in our past discussions regarding large-scale brain networks, but suffice to say, it is an important component of the TPN responsible for orienting attention towards unexpected external stimuli. On the other hand, the dorsal attention network (the DAN) is responsible for control of attention towards internal stimuli. This same study (Bilevicius et al., 2018) also found decreased functional connectivity between the DMN and the SN as well as increased intra-SN connectivity with increasing MAAS score. These studies individually noted various correlates between increasing score in measures of mindfulness (the MAAS in Bilevicius et al., 2018 and the FFMQ in Parkinson et al., 2019). The detailed findings of these studies are beyond the scope of this article, as we will primarily discuss changes as a result of mindfulness acquired with practice, but I encourage you to refer to the literature – Table 1 in the review by Sezer et al., 2022 summarizes the findings from these two studies.

         It is possible to increase one’s trait mindfulness, and it has been shown that mindfulness training can increase scores in measures of trait mindfulness (Quaglia et al., 2016). One would expect that the changes in functional connectivity between individuals before and after mindfulness training would be like the differences observed between those who scored lower in psychometric measures of mindfulness and those who scored higher in psychometric measures of mindfulness, respectively. However, according to Sezer and coworkers, there are some key differences (Sezer et al., 2022), such as decreased connectivity between the middle frontal gyrus and the DMN in experienced meditators compared to increased connectivity in individuals scoring high on the MAAS among other differences. Now that we have briefly discussed trait mindfulness, we can move on to discussion regarding the effects of mindfulness training on large-scale brain networks. 

         There is robust evidence that mindfulness training increases within-network connectivity in the SN and alters connectivity between the SN and the DMN. In the review by Sezer and coworkers, evidence is presented that in groups that underwent mindfulness meditation training, decreased connectivity was observed between a node in the DMN (the cuneus) and the SN (particularly the insular node of the SN; Sezer et al., 2022). However, the review also cites evidence that mindfulness training increases connectivity between the dorsal anterior cingulate cortex (dACC) node of the SN and the posterior cingulate cortex (PCC) node of the DMN. These seemingly contradictory findings highlight the importance of individual nodes within large-scale brain networks.

       A meta-analysis (meta-analyses are systematic investigations of many different publications; the systematic method of analysis differentiates a meta-analysis from a simple review paper) by Rahrig and coworkers concluded that mindfulness meditation training increased connectivity between the SN and the DMN, though it should be noted that the authors specifically found increased connectivity between the dACC and the PCC node of the DMN (Rahrig et al., 2022). This may appear to contradict the results from a study in which decreased connectivity was observed between the DMN (seeds centered in the mPFC and the PCC) and the ACC node of the SN with increasing meditation experience (Hasenkamp and Barsalou, 2012). However, this latter study split participants into high experience and low experience groups and further split meditation into four phases each with distinct seed regions selected for connectivity analysis (the phases were: mind-wandering, focus, shifting, and awareness). Recall that while mindfulness practices do not attempt to eliminate mind-wandering, a goal of mindfulness practice is increased control over mind-wandering. Thus, decreased connectivity between the ACC and the PCC/mPFC-centered seeds during mind-wandering in experienced practitioners makes sense: they can consciously decrease the engagement between the SN (mediating the object of their attention) and the DMN. In a recent study by Bremer and coworkers (Bremer et al., 2022), independent components analysis (of dynamic functional connectivity) found increased connectivity between the superior posterior subnetwork of the DMN (which includes the PCC) and two subnetworks of the SN (the insular SN and the dorsal SN) after mindfulness meditation training. Seed-based analysis found increased connectivity between the dorsal SN (with the seed based in the ACC) and nodes of the DMN (including the PCC) in individuals after mindfulness meditation training. Despite the modest sample size (46 participants total, 20 in the mindfulness training group and 26 healthy controls), the study by Bremer and colleagues presents high-quality evidence for increased connectivity between the DMN and the SN by mindfulness training. First, the authors utilized analysis of dynamic functional connectivity. Static functional connectivity assesses connectivity as a single snapshot which averages connectivity during the entire scan duration. Dynamic functional connectivity analysis, which uses a sliding window analysis, is sensitive to subtle changes in connectivity during the scan. The authors did employ analysis of static functional connectivity, but they were unable to detect any changes induced by mindfulness training. Second, Bremer and coworkers utilized two different approaches for analysis of the data obtained via fMRI. They utilized a seed-based approach in which seed regions were selected (these seed regions were placed in key nodes in networks hypothesized to be modulated by mindfulness) and they utilized a statistical/computational approach called independent components analysis which teases out correlations in activity between regions. 

          Which nodes in the SN specifically appear to downregulate activity of the DMN? A study by Brewer and coworkers (Brewer et al., 2011) found that across three different variants of mindfulness meditation (choiceless awareness, loving-kindness, and concentration) in experienced practitioners, there was decreased activation in two nodes of the DMN, the PCC and the mPFC. This decreased activation was retained even at baseline (i.e. while individuals were not actively meditating). Functional connectivity analysis using the PCC as a seed region revealed that experienced meditators (relative to healthy controls) possessed increased connectivity between the dACC and the PCC. This may appear to contradict the finding by Hasenkamp and Barsalou, which showed decreased connectivity between the PCC seed region and the ACC, but remember that Hasenkamp and coworkers split meditation into four phases (Hasenkamp and Barsalou, 2012). In this study, during the mind-wandering phase, the seed regions selected were the ventromedial prefrontal cortex and the posterior cingulate cortex. The study by Brewer and colleagues (Brewer et al., 2011) did not split meditation practice into phases and thus represents a more static image of functional connectivity. Additionally, Brewer and coworkers observed increased connectivity with the posterior insula when the mPFC node of the DMN was used as a seed region in experienced meditators regardless of the specific type of meditation – while the posterior insula is generally not canonically considered as part of the SN (the anterior insula is more relevant to the SN), it forms connections with the insular node of the SN, and some authors have included it as part of the SN (Mooneyham et al., 2016). It is typically associated with interoception and awareness of visceral sensation. Thus, there is extensive evidence from the literature that mindfulness training can modulate the connectivity between the SN and the DMN. Next, we will discuss modulation of connectivity between the SN and elements of the TPN by mindfulness practices.

        Recall that the TPN consists of several subnetworks including the DAN and the FPCN. It is not as cohesive a network as the SN or even the DMN. Indeed, different subnetworks within the TPN serve different functions. There is evidence from a longitudinal study that individuals who underwent mindfulness-based stress reduction (MBSR) training demonstrated increased activation of the dlPFC node within the FPCN (see Goldin et al., 2010 and Allen et al., 2012; reviewed in Tang et al., 2015). Regarding the FPCN, before proceeding, it is important to mention that the FPCN is actually a broader network with various functions. Some authors state that a more narrowly defined executive network has increased relevance to mindfulness compared to the broader FPCN, which itself is a subnetwork within the TPN (Mooneyham et al., 2016). Nevertheless, since we already established the general function of the FPCN in the previous article, we will stick to utilizing the FPCN to refer to the executive network when speaking about mindfulness.

       Mindfulness meditation training increased connectivity between the insula (which includes a node within the SN) and the dlPFC (reviewed in Tang et al., 2015) while decreasing connectivity between the insula and the mPFC. The authors of this study (referenced by Tang and coworkers) concluded that this change in connectivity represented a shift away from self-referential processing towards more objective interoceptive/exteroceptive processing. The authors of another review stated that increased functional connectivity between the insula and the dlPFC can be observed after mindfulness training (Mooneyham et al., 2016). Similar to this finding of increased insula-dlPFC connectivity, during the focus (sustained attention) phase of FA mindfulness meditation assessed by Hasenkamp and coworkers (Hasenkamp and Barsalou, 2012), increased connectivity between the dlPFC and the insula was observed in individuals with more mindfulness experience. In line with this finding, Bremer and coworkers (Bremer et al., 2022) found that in the mindfulness meditation group, increased connectivity between the insular node of the SN and two nodes that are parts of the TPN (the frontal eye fields and the dlPFC, nodes in the DAN and the FPCN, respectively) was observed. The Hasenkamp study (Hasenkamp et al., 2012) also found increased intra-FPCN connectivity during the shift (re-orientation of attention) phase of meditation, perhaps suggesting that connectivity with the SN is more important for maintaining prolonged focused attention than for shifting attention back towards the “anchor”. On the other hand, the anterior insula has been implicated in dynamic switching involved in activation of the FPCN and deactivation of the DMN. Activation of the anterior insular node of the SN tends to precede activation within the FPCN (Mooneyham et al., 2016), highlighting the importance of this node in regulating activity within the FPCN. With regards to FPCN-SN connectivity, Mooneyham and coworkers concluded that one of the most consistent functional connectivity finding mediated by mindfulness practices across all reviewed studies was increased functional connectivity between the insula and the dlPFC. The meta-analysis by Rahrig and coworkers did not find evidence for changes in connectivity between the FPCN and the SN, but it should be noted that only two of the analyzed studies in that meta-analysis mentioned the FPCN (Rahrig et al., 2022). Thus, we have elaborated on modulation of TPN-SN connectivity by mindfulness practices. Next, we will consider the possibility that the SN may not necessarily mediate all the changes in activity within the TPN and the DMN mediated by mindfulness (in other words, there may be evidence of changes to direct connectivity between elements of the TPN and the DMN by mindfulness training). We will examine how this may not necessarily contradict the triple-network model hypothesis put forth in the previous article.

        In the previous article, we put forth the idea that mindfulness practice increases the degree of control that the SN has over the TPN and the DMN. According to my review of the literature, this appears to be the case. However, it would not be accurate to state that there is no direct connectivity between the DMN and elements of the TPN. Not all the regulation of activity in the DMN and the TPN goes through the SN. This is neither completely true of general consciousness nor of states that are experienced during mindfulness. Indeed, as we mentioned in the previous article, there is evidence that the FPCN can mediate a decrease in DMN activity and an increase in DAN activity (Gao and Lin, 2012). Just because activity (observed using fMRI in human subjects) in the TPN and the DMN is generally anticorrelated does not necessarily mean that there is not direct connectivity between the two networks. In the review by Sezer and coworkers (Sezer et al., 2022), the authors cited evidence that mindfulness training increased coupling between the dlPFC node of the FPCN and the PCC node of the DMN. They suggested that the PCC is a heterogenous structure with many different functions and that fMRI at the time may not have had the spatial resolution to distinguish different parts of the PCC that are connected to different networks. Moreover, they also suggested that the FPCN can be split into two subnetworks: FPCNA and FPCNB. The former subnetwork consists of nodes such as the dlPFC and the posterior medial cortex. It communicates more directly with the DMN, especially the PCC node. This FPCNA subnetwork is involved during tasks that demand internally directed attention. The second subnetwork (FPCNB) contains the inferior parietal sulcus and is more well connected with the DAN. The FPCNB subnetwork is involved during tasks that demand externally-directed attention. Indeed, the review by Sezer and coworkers cites at least three publications that found changes in connectivity between the PCC node of the DMN and the dlPFC node of the FPCN (FPCNA specifically). Notably, in the study by Brewer and coworkers (Brewer et al., 2011), increased connectivity between the dlPFC node of the FPCN and the PCC node of the DMN was observed across all practitioners of mindfulness meditation, although this finding did not reach the threshold for statistical significance due to the lower strength of anticorrelations in controls (we use the term significance here not to refer to “important” findings, but rather to refer to the notion of statistical significance). When analyzing large-scale brain networks, it is important to realize that an individual network may communicate with many other networks and nodes within a given network may communicate with many different nodes, some in other networks and some in the same network. It is not possible to accurately take a reductionistic view of network function in the brain and analysis of changes in activity induced by mindfulness practice is no different. Now that we have gone over changes in activity and connectivity within and between large-scale brain networks mediated by mindfulness, we will switch gears to a slightly different, but related topic: anatomic changes produced by mindfulness practice.

          Thus far, we have focused primarily on changes in activity within large-scale brain networks mediated by mindfulness training as measured by fMRI. These changes in activity and connectivity are, at a finer level, likely occur through changes in the structure of the brain. In other words, you can literally change the composition and organization of the brain through mindfulness practice (without needing to resort to chemical means of altering brain activity). While there is a growing body of literature characterizing these changes through mindfulness training, we will focus on one anatomical change in particular: structural changes in the anterior cingulate cortex 

         A typical neuron consists of three main parts: dendrites, the soma, and the axon. The dendrites receive input. For most neurons, this input is in the form of neurotransmitters. Neurotransmitters bind to protein receptors on the surface of the neuron. When a neurotransmitter binds to a receptor, a common outcome is electrical activity. This change in electrical activity is transmitted to the body of the neuron, the soma. However, along the way, the electrical activity decays (thus signals from dendrites that are farther from the soma, called distal dendrites, are typically weaker than signals from dendrites that are that are closer to the soma). At the soma, this electrical activity is integrated and summed. If the electrical activity is above a certain threshold, the neuron will fire an action potential – in this sense, action potentials are all-or-nothing phenomena (either the neuron fires or it doesn’t). The action potential is transmitted down the axon which is lined with special proteins that regenerate the electrical activity (voltage-gated ion channels, most importantly ion channels for sodium ions). However, this process takes time and energy, valuable resources when attempting to convey information. To minimize this expenditure of time and energy, some of the axons within our neurons are lined by sheaths made up of a material called myelin. You can think of this myelin as analogous to the insulation around an electrical wire, except in this case the charged particles are ions rather than electrons. Myelination prevents ions from leaking out of the axon, causing the signal to decay. On myelinated neurons, myelination is interrupted by gaps (“naked” regions) called nodes of Ranvier. Why? Without these gaps, there would also be a decay in electrical activity. In this manner, an impulse along a myelinated axon hops from one node of Ranvier to the other. These nodes of Ranvier contain voltage-gated ion channels important for regenerating the action potential. Not all neurons are myelinated, but the ones that are myelinated conduct action potentials much more rapidly and at a lower energetic cost than those that are unmyelinated. Neurons themselves do not synthesize myelin. Rather, specialized cells form these myelin sheaths (oligodendrocytes in the spinal cord and the brain). White matter tracts consist of bundles of myelinated axons.

         Mindfulness practice has been demonstrated to change the integrity of white matter in the anterior cingulate cortex, a key node in the salience network. Importantly, this change occurs quite rapidly. In the review by Tang and coworkers (Tang et al., 2015), the authors cite evidence that mindfulness practice produces changes in white matter located in the anterior cingulate cortex. Just eleven hours of mindfulness practice (through a technique called integrative body-mind training or IBMT) produced increases in fractional anisotropy, a measure of the efficiency and integrity of white matter (Tang et al., 2010). Remember that expert practitioners of mindfulness may accumulate hundreds or even thousands of hours of mindfulness practice over their lifetime. The 2010 study by Tang and coworkers demonstrated increased fractional anisotropy in multiple white matter tracts connected with the ACC including the superior corona radiata, the anterior corona radiata, the corpus callosum, and the longitudinal fasciculus after eleven hours of IBMT divided into thirty-minute sessions over a period of a month. The most significant change was observed in the anterior corona radiata. This change was not observed after three or six hours of IBMT, indicating the importance of consistent practice. Although it is not possible to directly observe the molecular and cellular mechanism of this change in living human subjects, it was speculated by Posner and coworkers that mindfulness practice increases frontal theta rhythm, a product of coordinated electrical activity within populations of neurons (Posner et al., 2014). This increased electrical activity was proposed to activate an enzyme called calpain which alters the structure of scaffolding within axons. Although Tang and coworkers did not observe changes in ACC gray matter (gray matter, in contrast to white matter, consists of the cell bodies of neurons), this does not exclude the possibility that mindfulness training can increase gray matter density within the ACC. Indeed, in experienced Zen meditators (Zen meditation is a form of meditation in which mindfulness is a key tenet), increased thickness as a result of increased gray matter density in the ACC and the insula was observed (Grant et al., 2016). Thus, there is evidence that mindfulness practice can directly alter not only the function of brain regions, but it can also directly alter the structure of these regions.

      There are many other examples of how mindfulness can directly alter the structure of the brain (see Tang et al., 2015 for review). I have not reviewed all the relevant literature on this topic exhaustively, but seeing as how mindfulness practice is associated with decreased emotional reactivity, I would speculate that there might also be changes in the structure of the amygdala, particularly in individuals who suffer from depression or anxiety disorders. Chronic stress (associated with depression) can induce hypertrophy (excessive growth in terms of size) of the amygdala. At the same time, individuals with various anxiety disorders also tend to have increased activity of the amygdala. Considering the general benefits of mindfulness in terms of emotional reactivity, it would thus make sense that mindfulness practice modulates the structure of the amygdala. A quick glance at the evidence suggests that mindfulness does indeed decrease activity of the amygdala and can also decrease gray matter density in the amygdala (Tang et al., 2015).

       Thus, in this article, we have reviewed how researchers quantify trait mindfulness using psychometric assessments (the MAAS and the FFMQ, for example). We discussed how trait mindfulness can be correlated with differences in the connectivity of large-scale brain networks. For the remainder of the article, we discussed how mindfulness practices influence the connectivity of these brain networks in the triple-network model. Finally, we discussed how mindfulness can directly change the structure of the brain. A key takeaway here is that although some people are naturally more mindful than others, consistent practice of mindfulness can result in an increase in trait mindfulness. This training can directly affect the structure and function of the brain, which can give rise to many benefits for one’s mental health as well as one’s physical health.

References

Allen, M., Dietz, M., Blair, K. S., van Beek, M., Rees, G., Vestergaard-Poulsen, P., Lutz, A., & Roepstorff, A. (2012). Cognitive-affective neural plasticity following active-controlled mindfulness intervention. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 32(44), 15601–15610. https://doi.org/10.1523/JNEUROSCI.2957-12.2012

Bilevicius, E., Smith, S. D., & Kornelsen, J. (2018). Resting-State Network Functional Connectivity Patterns Associated with the Mindful Attention Awareness Scale. Brain Connectivity, 8(1), 40–48. https://doi.org/10.1089/brain.2017.0520

Bremer, B., Wu, Q., Mora Álvarez, M. G., Hölzel, B. K., Wilhelm, M., Hell, E., Tavacioglu, E. E., Torske, A., & Koch, K. (2022). Mindfulness meditation increases default mode, salience, and central executive network connectivity. Scientific Reports, 12(1), 13219. https://doi.org/10.1038/s41598-022-17325-6

Gao, W., & Lin, W. (2011). Frontal parietal control network regulates the anti‐correlated default and dorsal attention networks. Human Brain Mapping, 33(1), 192–202. https://doi.org/10.1002/hbm.21204

Goldin, P. R., & Gross, J. J. (2010). Effects of Mindfulness-Based Stress Reduction (MBSR) on Emotion Regulation in Social Anxiety Disorder. Emotion (Washington, D.C.), 10(1), 83–91. https://doi.org/10.1037/a0018441

Grant, J. A., Courtemanche, J., Duerden, E. G., Duncan, G. H., & Rainville, P. (2010). Cortical thickness and pain sensitivity in zen meditators. Emotion (Washington, D.C.), 10(1), 43–53. https://doi.org/10.1037/a0018334

Hasenkamp, W., & Barsalou, L. W. (2012). Effects of meditation experience on functional connectivity of distributed brain networks. Frontiers in Human Neuroscience, 6, 38. https://doi.org/10.3389/fnhum.2012.00038

Mooneyham, B. W., Mrazek, M. D., Mrazek, A. J., & Schooler, J. W. (2016). Signal or noise: Brain network interactions underlying the experience and training of mindfulness. Annals of the New York Academy of Sciences, 1369(1), 240–256. https://doi.org/10.1111/nyas.13044

Parkinson, T. D., Kornelsen, J., & Smith, S. D. (2019). Trait Mindfulness and Functional Connectivity in Cognitive and Attentional Resting State Networks. Frontiers in Human Neuroscience, 13, 112. https://doi.org/10.3389/fnhum.2019.00112

Posner, M. I., Tang, Y.-Y., & Lynch, G. (2014). Mechanisms of white matter change induced by meditation training. Frontiers in Psychology, 5, 1220. https://doi.org/10.3389/fpsyg.2014.01220

Quaglia, J. T., Braun, S. E., Freeman, S. P., McDaniel, M. A., & Brown, K. W. (2016). Meta-analytic evidence for effects of mindfulness training on dimensions of self-reported dispositional mindfulness. Psychological Assessment, 28(7), 803–818. https://doi.org/10.1037/pas0000268

Rahrig, H., Vago, D. R., Passarelli, M. A., Auten, A., Lynn, N. A., & Brown, K. W. (2022). Meta-analytic evidence that mindfulness training alters resting state default mode network connectivity. Scientific Reports, 12(1), 12260. https://doi.org/10.1038/s41598-022-15195-6

Sezer, I., Pizzagalli, D. A., & Sacchet, M. D. (2022). Resting-state fMRI functional connectivity and mindfulness in clinical and non-clinical contexts: A review and synthesis. Neuroscience & Biobehavioral Reviews, 135, 104583. https://doi.org/10.1016/j.neubiorev.2022.104583

Tang, Y.-Y., Hölzel, B. K., & Posner, M. I. (2015). The neuroscience of mindfulness meditation. Nature Reviews Neuroscience, 16(4), 213–225. https://doi.org/10.1038/nrn3916

Tang, Y.-Y., Lu, Q., Geng, X., Stein, E. A., Yang, Y., & Posner, M. I. (2010). Short-term meditation induces white matter changes in the anterior cingulate. Proceedings of the National Academy of Sciences, 107(35), 15649–15652. https://doi.org/10.1073/pnas.1011043107

1. A web version of the FFMQ questionnaire can be found here: http://www.awakemind.org/quiz.php

2. A web version of the MAAS questionnaire can be found here: https://psytests.org/cbt/maasen.html