Alterations in task-induced activity and resting-state fluctuations in visual and DMN areas revealed in long-term meditators

Abstract

Recently we proposed that the information contained in spontaneously emerging (resting-state) fluctuations may reflect individually unique neuro-cognitive traits. One prediction of this conjecture, termed the “spontaneous trait reactivation” (STR) hypothesis, is that resting-state activity patterns could be diagnostic of unique personalities, talents and life-styles of individuals. Long-term meditators could provide a unique experimental group to test this hypothesis. Using fMRI we found that, during resting-state, the amplitude of spontaneous fluctuations in long-term mindfulness meditation (MM) practitioners was enhanced in the visual cortex and significantly reduced in the DMN compared to naïve controls. Importantly, during a visual recognition memory task, the MM group showed heightened visual cortex responsivity, concomitant with weaker negative responses in Default Mode Network (DMN) areas. This effect was also reflected in the behavioral performance, where MM practitioners performed significantly faster than the control group. Thus, our results uncover opposite changes in the visual and default mode systems in long-term meditators which are revealed during both rest and task. The results support the STR hypothesis and extend it to the domain of local changes in the magnitude of the spontaneous fluctuations.

Introduction

In recent years there is an increasing interest in the phenomenon of slow spontaneous fluctuations, emerging in the absence of sensory stimulation or an explicit task (also termed “resting-state”), which appears in all networks of the cerebral cortex (Biswal et al., 1995, Golland et al., 2007, Nir et al., 2008). A number of studies have demonstrated that these fluctuations show correlation structures (also termed “functional connectivity”) that reflect the co-activation of regions comprising fundamental networks that are activated or suppressed during task performance (Cordes et al., 2000, Fox et al., 2006, Smith et al., 2009).

The study of spontaneous “resting state” fluctuations has dramatically expanded in recent years, leading to detailed characterizations of these fluctuations across thousands of individuals. However, the source of the resting state fluctuations, and particularly – what can be learnt from their patterning – remains unknown (e.g. Moutard, Dehaene, & Malach, 2015). The present study was based on a recently proposed hypothesis named “spontaneous trait reactivation” (STR) (Harmelech & Malach, 2013), arguing that the patterning of spontaneous fluctuations represents not only large-scale, task related networks, but also subtle individual differences that reflect individually-unique habitual network activations. More specifically, the STR hypothesis suggests that spontaneous fluctuations can be actively remodeled in a long-term manner by focused and intense cortical training. While such potential links have been amply discussed in the clinical literature, there is limited evidence for it in healthy individuals. If this hypothesis has some validity, one would expect that individuals whose life-style entails unique habitual brain activations should show correspondingly unique changes in their resting state patterns. For example, the STR hypothesis predicts that professional musicians, jugglers or professional translators, should have subtle and unique differences in their resting state patterns compared to more typical individuals. The long-term meditators provide a unique testing ground along two main dimensions: first, the number of practice hours is especially long and habitual. Second, the type of attentional control is unique in its distinction from a direct engagement with sensory or physical activity, typical of other kinds of expertise.

A number of studies appear to be compatible with the STR hypothesis in showing the correlation of spontaneous fluctuations during resting-state with unique clinical populations (Greicius et al., 2009, Hahamy et al., 2014, Hahamy et al., 2015a, Hahamy et al., 2015b). More recently this correlation was extended even to the domain of gene expression (Richiardi et al., 2015).

However, in contrast to the rapidly expanding research relating resting-state changes to clinical populations, data about such associations in healthy individuals with unique habits which can be considered ‘cortical training’ is more limited (albeit see, for example, Adelstein et al., 2011, Guidotti et al., 2015, Johnen et al., 2015, Rohr et al., 2015). In that respect, individuals that practice various forms of meditation provide a unique testing ground for the STR hypothesis. These individuals engage in a unique life style, typically involving 0.5–1 h daily of a specific and highly controlled mental task – meditation – that is not shared by the general population. Thus, a straight forward prediction of our STR hypothesis was that these individuals should manifest unique changes compared to controls in their resting state patterns of spontaneous activations. Such tasks have been amply demonstrated to significantly alter brain function and network connectivity (Fox et al., 2014, Tang et al., 2012). Specifically, previous research has demonstrated reduced blood oxygenated level dependent (BOLD) signal level in the Default Mode Network (DMN) (Raichle et al., 2001) — a system associated with intrinsically-oriented and self-related functions (Buckner et al., 2008, Golland et al., 2007, Preminger et al., 2011) during various kinds of meditation. (e.g Brewer et al., 2011, Pagnoni et al., 2008, Pagnoni, 2012, but see also opposite evidence, Hölzel et al., 2007, Xu et al., 2014).

In parallel, changes in functional-connectivity during spontaneous resting-state fluctuations have been reported to occur in long-term meditators. Reports of such resting-state changes in the DMN include both reduced (Taylor et al., 2013) and increased (Jang et al., 2011, Taylor et al., 2013) functional connectivity among various DMN nodes, as well as altered functional connectivity between DMN nodes and regions outside of the DMN including the cognitive control Network (Brewer et al., 2011), sensory regions (Kilpatrick et al., 2011), and the orbitofrontal cortex (Hasenkamp and Barsalou, 2012, Jang et al., 2011).

However, the STR prediction has not been addressed by these previous studies. Thus, it is not clear whether the observed changes in the resting-state fluctuations reflect similar changes in task-induced activations in mindfulness meditation (MM).

It is important to emphasize that our study was not aimed at establishing a causal relationship between the practices of meditation and changes in resting-state fluctuations, but merely to explore the possible correlation between these two phenomena. This is reflected in the cross-sectional, instead of a longitudinal, design. Thus, we keep open the possibilities that the resting-state fluctuations may either reflect a personality trait that, on the one hand, leads individuals to more readily engage in meditation, and on the other hand, is reflected in concurrent changes of task-activations. Alternatively, the changes in resting state fluctuations may be due to causal changes produced by meditative training effects. Of course these possibilities are not mutually exclusive, and the correlation between rest and task may reflect both training and disposition effects.

To examine these questions, we compared brain activations during a visual recognition-memory task (VRM) and during resting-state in long-term meditators and controls. The VRM task had been used extensively in our laboratory, proving as an effective manner of evoking robust visual brain activations in a number of previous studies conducted on a typical population of participants (e.g. Yellin, Berkovich-Ohana, & Malach, 2015). We report here results obtained, specifically, in long-term Mindfulness meditators (MM). Mindfulness meditation is a technique aimed at cultivating an intentional focus of attention on momentary experiences, practiced by continuously returning one's focus from mind-wandering to the object of meditation such as the breath, while cultivating a receptive attitude towards all arising experiences (Gunaratana & Gunaratana, 2011). Accumulating evidence suggests that MM practice increases attention and emotion-regulation, as well as alters self-awareness (Chiesa and Serretti, 2010, Hölzel et al., 2011, Tang et al., 2015).