Figure 4. Mindfulness process model—open monitoring receptive practice. OM meditation with no object of focus (ambient attention) is illustrated as an example, but the processes are proposed to be the same across receptive styles of practice. Intention is formed along with motivation to practice before an executive “set” is created. The working memory system helps to maintain motivation and practice instructions. Ambient (diffuse) attentional networks are recruited and sustained along with support by component mechanisms including executive monitoring, emotion regulation, and response inhibition. Mental noting and labeling of stimuli arising, passing, and absent in phenomenal conscious awareness is a form of emotion regulation and contributes to extinction and reconsolidation of maladaptive procedural and declarative memories that represent sensory-affective-motor scripts and schemas. Affective responses may arise in response to an object of attention with a positive, negative, or neutral valence and are likely to proliferate unless awareness and de-centering promote response inhibition and disengagement. Over time and continued practice, effortful control is reduced and awareness itself becomes the object of attention as meta-awareness is cultivated as a skill. Clarity and Equanimity increases through practice.

Figure 5. Mindfulness process model. Ethical-enhancement (EE) practice. Intention is formed along with motivation to practice before an executive “set” is created within the working memory system in order to implement and maintain practice instructions. Focused attentional networks are recruited for orienting toward sustained episodic memory recall involving mentalizing and shared experience. OM practice is described to supplement EE practice facilitating awareness of any modality of experience that arises. The processes are proposed to be the same across ethical styles of practice that require imagined suffering or a particular declarative memory [of someone or something]. Mentalizing one’s own or others’ suffering continues recurrently as part of sustained episodic memory recall. With clarity and equanimity for affective reactivity, OM practice allows the practitioner to remain mindfully aware of difficult emotions while the declarative (episodic) memory is positively reappraised. Negative associations are extinguished and reconsolidated into more adaptive or positive memories using prosocial/empathic concern for the object of meditation. Continuous reappraisal along with prosocial concern is thought to enhance exposure, extinction, and reconsolidation processes such that new episodic memories are laid down and inhibit older maladaptive forms of the memory.

Focused Attention (FA) Concentration Practice

FA practice involves sustained attention on a specific mental or sensory object: a repeated sound or mantra, an imagined or physical image, or specific viscerosomatic sensations. The object of focus can be anything, but the method described by the Satipatthāna Sutta identifies a naturally occurring breath focus. In fact, breath is a particularly apt foundation because it integrates conscious awareness with ongoing, dynamic viscerosomatic function. The goal of FA practice is to stabilize the mind from distraction, torpor, and hyperexcitability, all of which are predicted to be negatively correlated with practice. Effort is hypothesized to be inversely correlated with practice duration, providing a neurobiological mechanism for advancement of the practitioner and resulting in decreased allocation of explicit cognitive resources over time. A process model for FA is illustrated in Figure 3.

FA practice is described here to involve an underlying framework of motor learning that functions to strengthen non-conscious, associative memory processes underlying the EES network (and subcortical analogs), and which support the conscious, explicit processes recruited during practice. Intention, a critical component of mindfulness in previous conceptual models (see Shapiro et al., 2006), is described to be critical for motivating the practitioner to begin/sustain practice and activating EES networks that may be the initial steps for extinguishing maladaptive habitual perceptual-motor action tendencies. Intention and motivation may also target attentional tuning and affective control settings which contribute to affect-biased attention (Todd et al., 2012). The instructions for practice form an executive “set” that is created and sustained by working memory processes, while attentional processes operate to focus and sustain concentration on the intended object. Alerting, orienting, engagement, and disengagement all involve discrete networks (Posner and Petersen, 1990) which complement one another in their contributions to the FA practice. An executive monitoring function is proposed to provide feedback when the goal state of concentration on the particular object has shifted and to support the attentional processes related to the return of focus. Meta-awareness is described to arise as a highly developed form of executive monitor that allows the practitioner to have awareness as the object of attention while simultaneously maintaining awareness in its natural state, whichever object of attention it may be resting upon. Sensory clarity is presumed to increase in proportion to the strengthening of meta-awareness. Given the perspective shift one must learn through development of meta-awareness, decentering is described as a psychological process that may be a mechanism to support disengagement processes and sensory clarity.

Equanimity and clarity are both processes that support concentrative practice and are proposed to increase in terms of efficiency and phenomenal quality through time and practice. Similar to the Buddhist description (see Section “Defining Meditation and Mindfulness from the Historical Perspective”), equanimity refers to response inhibition and emotion regulation processes that reduce later attentional and emotional stages of strategic processing potentially involving prolonged sympathetic arousal, cognitive elaboration, or ruminative qualities which may arise in response to an object of distraction. Increased clarity of experience is proposed to operate as one develops an increased phenomenal awareness of the breath and objects of distraction arise and disappear, without necessarily affecting the cognitive access of the contents of conscious experience.

Open Monitoring (OM) Receptive Practice

Figure 4 illustrates a process model for the receptive practice, OM, which is described in detail elsewhere. Novices begin by actively monitoring and labeling external (exteroceptive) or internal (interoceptive) domains of the five senses, including viscerosomatic states and inner speech. The specific sensory modalities provide the scaffolding for meta-awareness to develop in each domain and in relation to self-related processing. There is an explicit distinction between engagement and disengagement with an object early on in the practice. Provided the framework of motor learning, advancement of the practitioner is likely to automatize aspects of attentional processing and improve efficiency of engagement and disengagement processes, thereby reducing bias associated with orienting of attention. For example, when distraction arises, and the practitioner gets “stuck” in thought or emotion, a number of mechanisms supported by mindfulness-based practices may facilitate disengagement. The practitioner is encouraged to continually rest in awareness of intero- and exteroceptive stimuli, and any cognitive or emotional states that may arise. Through training, the practitioner can learn to note and sustain attention on the arising, passing, or absence of each modality of experience—naturally reducing the frequency of cognitive evaluation or rumination. Eventually, through training, the process of mental noting becomes effortless and non-reactive, qualities that describe the development of equanimity. At this point, efficiency of the attentional system has improved, resulting in decreased allotment of attentional resources toward any particular feeling, image, or thought. Cognitive emotion regulation strategies like reappraisal or suppression may be more prevalent in novice practitioners dealing with distraction, but less so in advanced practitioners as attentional systems are strengthened and qualities of equanimity arise. One may hypothesize that the changes from novice to advanced practitioner may be due to the development of psychological processes like non-attachment, de-centering, or from a non-conscious shift in affect-biased attention and the development of meta-awareness. Whichever the mechanism, maladaptive scripts and schemas related to the self are proposed to be extinguished and reconsolidated through a combination of FA and OM practice.

Similar meditation practices instruct the practitioner to maintain a diffuse form of attention or ambient attentional focus such as “Shikantaza” or “just sitting” in Japanese Soto Zen (Austin, 2006); “choiceless awareness” by Krishnamurti (1969), insight meditation from Theravada (Pali: Vipassana) (Lutz et al., 2007), and “absolute rest,” by other western Buddhist teachers (Young, 2010). Vipassana, has also been translated as “removing the veil of ignorance” (Gunaratana, 2002), such that this stage of meditation practice explicitly aims at providing insight into one’s self-related processes. In advanced practices, the practitioner can reside in a non-dual level of awareness, referred to by the Buddhist Abhidharma scholars as a clear understanding of no distinction between knowing subject and perceived or known object (Dunne, 2011). Some Buddhist traditions have referred to this non-dual practice as “Open Presence” (Thangru and Johnson, 2004; Lutz et al., 2007), “open awareness,” “rig.pa” in Tibetan forms of Dzogchen (Rinpoche et al., 2006), or “phyag.chen” in Mahamudra (Thangru and Johnson, 2004). Non-dual practice involves the decreased reliance on engagement and disengagement processes, mental noting or labeling techniques. Such changes in the practice are likely to affect the relationship the practitioner has with his/her own thoughts, so that cognitive structures related to ones’ thought and emotion that represent reality are “de-fused” from what is considered reality itself (Hayes et al., 2003; Dunne, 2011).

Ethical Enhancement (EE) Practices

The diminishing of grasping, aversion, and delusion that arises in insight meditation practice naturally leads the practitioner to greater loving-kindness toward oneself and others (Salzberg, 2011). The cultivation of loving-kindness (Pali; metta), or forgiveness can also have very explicit instructions for visualization and is based on the premise that all beings want to be happy. The following excerpt comes from one of the original teachings of loving-kindness practice.

Even as a mother protects with her life
Her child, her only child,
So with a boundless heart
Should one cherish all living beings

(Karaniya metta sutta, 1.8)

Contemporary instruction for the cultivation of loving-kindness can be found in the works of Sharon Salzberg (Salzberg and Bush, 1999) and by Matthieu Ricard (Ricard, 2003), but such practices are based on the method found in Buddhaghosa’s 5th century text, the Path to Purification (Buddhaghosa, 1991). This traditional approach is characterized by progressive cultivation of loving-kindness toward oneself, a good friend, a neutral person, a difficult person, all four of the above equally, and then gradually the entire universe. The loving-kindness is cultivated in reference to a specific declarative (episodic) memory of each particular individual or event. The practitioner is encouraged to avoid choosing someone to whom they are sexually attracted or who is dead (Ricard, 2003). For a “neutral” person, the practitioner is encouraged to choose someone they may come into contact with every day, but who does not give rise to strong positive nor strong negative emotions. For a “difficult” person, the practitioner is encouraged to choose an enemy, or someone strongly disliked. Furthermore, the practitioner is encouraged to break down the barriers between self and other by practicing loving-kindness repeatedly, achieving impartiality toward the four persons, including him/herself, the close friend, the neutral person, and the hostile person. The EE practice is illustrated in Figure 5.

Compassion practice has also been described as an EE style of practice; however, without the simultaneous cultivation of open presence along with the generation of compassion, the practice has been described to be unsuccessful (see Lutz et al., 2007). Compassion practice is also described as an enactive, emergent process of factors in the attentional and affective domains, the intentional and insight domains, and the embodied and engaged domains of subjective experience (Halifax, 2012). Similar to the metta practice described above is Tonglen, a Tibetan Buddhist term for the meditation technique translated as “giving and taking” or “sending and receiving” in which the practitioner visualizes taking onto oneself the suffering of others on the in-breath, and on the out-breath giving happiness, joy, and kindness (Lama Xiv et al., 2001; Lutz et al., 2007).

Six Component Neurocognitive Mechanisms of Mindfulness

As illustrated, mindfulness is not a unitary construct and the process models are an attempt to illustrate the cognitive and psychological processes that support the meditation practices. These component processes provide the proposed mechanisms by which mindfulness works to increase S-ART—the framework for reducing self-specific biases and sustaining a healthy mind. The S-ART framework includes six component supportive mechanisms underlying the practice and cultivation of mindfulness as a state and trait: (1) intention and motivation; (2) attention regulation; (3) emotion regulation; (4) memory extinction and reconsolidation; (5) prosociality; (6) non-attachment and de-centering. Here we describe the neurobiological substrates of each mechanism and provide support from the extant contemplative science literature.

Intention and Motivation

As originally proposed by Shapiro and colleagues (2006), intention is one of the fundamental building blocks out of which an array of neurocognitive mechanisms may emerge to effectively cultivate mindfulness. Intention and motivation are based on inter-related aspects of affective style and biological disposition, driving goal-directed behavior (Davidson and Irwin, 1999; Ryan and Deci, 2000). External and internal motivation have been distinguished as two unique classes of motivation. Although both classes refer to incentives of reward, external motivation arises from the desire to approach (avoid) an externally imposed reward (punishment); while internal motivation arises from a more automatic, internally driven desire to obtain reward, or satisfy one’s needs (Davidson et al., 2000; Ochsner and Gross, 2005; Rothman et al., 2011). People are clearly motivated to practice by very unique factors with highly varied experiences, incentives, and consequences; however, through advancement of practice, motivation is proposed here to become more internally driven (i.e., suggesting increased control) and less focused on outcome. Intention refers more specifically to a purposive plan of action that is selected and the timing selected for such action (Krieghoff et al., 2011).

Motivation and intention are described as a feedback system for self-schema maintenance. As in any feedback system, sensory input is continually compared to a reference, and discrepancy from the reference feeds back into frontal goal-oriented networks and eventually to pre-motor-action networks modulating functional output. If the reference is some ideal that is unrealistic or if the interpretation of incoming signals related to the self are continually negative in valence, then there is an error signal of which the rate of discrepancy reduction may never progress, perpetuating negative affect (Carver and Scheier, 2011). Negatively biased cognitions and/or distorted perceptions that involve disinterest, disapproval, or rejection have been shown to result in nonsocial, isolating behaviors (including avoidance and withdrawal), impaired self-regulation, and psychopathology; conversely, current data suggest that approach behavior including more engaging perceptions involving interest, approval, or acceptance typically lead to healthier outcomes (Waikar and Craske, 1997; Sin and Lyubomirsky, 2009). Thus, internally driven motivation to engage with experience without craving or aversion (i.e., with equanimity) is suggested to increase self-regulation.

Davidson and Irwin (1999) suggest there are two fundamental resting patterns of neural activity for motivational systems of approach and avoidance. Left anterior PFC activation has been associated with positive affect and increased right-sided anterior activation with negative affect, suggesting that individual differences in asymmetric activation of these brain regions can be associated with dispositional differences in a biological set point for motivation and which may contribute to the overall outcomes of mindfulness training. Davidson and Irwin (1999) further delineates functional differences between ventromedial and dorsolateral sectors of PFC, suggesting that activation of the former represent immediate valenced goal states along with activation of the ventral striatopallidum [incl. nucleus accumbens (nAcc)], while the latter area DLPFC represent valenced goal states in the absence of immediately present incentives. In other words, DLPFC may contribute to a sustained form of motivation that remains in working memory during the meditation practice. Furthermore, animal and human research suggest that activity in the PFC can modulate the ventral striatopallidum and specifically nAcc activity in a top-down manner, postulating a mechanism for tracking internally driven, approach forms of motivation (Heller et al., 2009) as well as regulating aspects of craving associated with more phasic nAcc activity (Robbins and Everitt, 1996). More recent data suggests that decreased ventrolateral PFC (VLPFC) activity may also be evidence for increased approach behavior and expression of positive affect (Light et al., 2011). The contributions of these substrates to habits and automatization are typically described to be hard-wired, but through FA/OM and ethical based practices, it is proposed that extinction and reconsolidation mechanisms (see Section “Extinction and reconsolidation”) can shift biological set-points and modify pathological scripts and schemas into more adaptive trajectories.

There has yet to be extensive investigation of motivation, intention, and reward in meditation and mindfulness research; however there is some PET data demonstrating increased endogenous dopamine release in the ventral striatum during Yoga Nidra meditation (Kjaer et al., 2002), a finding accompanied by a reported “decrease desire for action.” Expression of loving-kindness, a sense of connectedness, and feelings of trust and cooperation have all shown to produce increased activity in the ventral striatum (Cialdini et al., 1997; Bora et al., 2009). Significant increased activations in rostral ACC and OFC have been found in the majority of non-guided meditation studies, suggesting effects related to attentional focus, but also likely related to tonic intentional activity.

Attention Regulation

Volitional shifting of conscious awareness between objects of attention in a serial and/or parallel fashion is suggested to be a critical process for effectively managing or altering one’s responses and impulses. This executive, volitional management of attention can explicitly control where and when attention shifts in top-down fashion. Regulation is also believed to occur from a bottom-up direction, such that cognitive resources are conserved and non-conscious afferent-efferent systems interact with the environment in an efficient, adaptive feedback loop with top-down systems. Concentration forms of practice are proposed to increase the efficiency of the attentional system, including the dissociable, but interrelated sub-components of attentional processing. The subcomponents are identified in the concentrative practice process model (Figure 3) and include alerting and orienting toward an intended object of interest, engaging with the object, sustaining attentional focus, executive monitoring and detecting distraction, disengaging from the source of distraction, and re-engaging on the intended object (Posner and Petersen, 1990; Corbetta and Shulman, 2002; Raz and Buhle, 2006; Posner and Rothbart, 2009).

Efficiency and stability are also measured by sustained attention (i.e., vigilance and alertness), referring to the capacity to detect unpredictable stimuli over prolonged periods of time (Posner and Rothbart, 1998). Deep engagement, vivid absorption, or concentration power, is an embodied state of awareness in which no other sensory or internally generated input can arise beyond a perceptual threshold (Rahula, 1974; Bodhi, 1999). Full absorption in an object with focal awareness is critical for stabilizing the mind during FA meditation; however, it can be maladaptive, such that inhibitory processes can prevent pertinent sensory information from arising to conscious awareness, decreasing available cognitive resources for ongoing task demands, and potentially leading to an overwhelming sensation and maintenance of emotional reactivity related to the object of focus (Mogg and Bradley, 1998). Regulatory efficiency and stability are also measured under various conditions of perceptual load or conflicting contexts (e.g., Stroop phenomena), through continued performance with fewer distractions, through the rapid disengagement from an object of focus to another intended object in rapid succession (e.g., attentional blink paradigm, dotprobe task), and through an ability to shift back and forth between local and global features (e.g., perceptual rivalry task). The neural substrates for these attentional processes are described through S-ART networks. For example, preparatory forms of attention are described through EES circuitry, while the substrates for FA and executive monitoring are described through EPS and the FPCS. It is likely that distinct circuitry may also be identified for arising, passing, and restful states of concentration for each modality of focus.

Although people generally claim to have full conscious awareness of their environment when their focal attention is directed outward or toward their own internal thoughts, attention research shows that this is quite false. In fact, it is very likely that particular information from the environment is filtered out unless reaching some perceptual or semantically meaningful threshold before it can gain access to conscious awareness. At a very basic perceptual level, our focal awareness is very small relative to the rest of the context around that focus, such that we experience a phenomenon called “change blindness” when cues such as motion, that normally lead to a shift of attention, are suppressed and major changes across sensory or semantic modalities in the remainder of the environment are not reported (Simons and Levin, 1997; Posner and Rothbart, 2009). There is ample evidence that early stages of attentional processing and fleeting perceptual traces of sensory information exist before there is conscious awareness (Pessoa, 2005). Receptive forms of meditation (i.e., OM) facilitate more diffuse or ambient attentional mechanisms and consequently, phenomenological awareness of the objects in conscious experience is likely enhanced, without necessarily increasing cognitive access [see (Davis and Thompson, forthcoming)]. Thus, reportability for states of interoception and exteroception may not necessarily be improved, while more subtle, non-conscious processes supported through EES may improve in efficiency. These forms of early, non-conscious processing can be measured in pre-categorical, high capacity sensory memory stores (Sperling, 1960; Chrousos and Gold, 1992), perceptual rivalry, or affect-biased attention (Attar et al., 2010; Todd et al., 2012). Future research in the contemplative sciences will likely reveal more subtle changes in non-conscious states of early stages of perceptual and attentional processing. A recent pilot study used an iconic memory task, in which an array of eight letters was displayed for 30 ms and a target letter was to be identified up to 1000 ms after the trace was removed. The study found dispositional mindfulness, non-attachment, and compassion in advanced meditators was not associated with accuracy, but positively correlated with the subjects’ ability to identify a letter adjacent to the correct target and in the larger array. It is proposed, therefore, that such dispotional factors leads to increased diffuse attention to the periphery without improving iconic perceptual memory traces to the target focus of an 8-item array (Fischer et al., 2012).

Meta-awareness points to the possibility of taking awareness itself as an object of attention and can be disambiguated from sensory-conceptual domains of self-experience especially evident in the NS (Lutz et al., 2007). Meta-awareness contributes the critical mnemonic aspect of mindful awareness responsible for monitoring the meditative state so that one may “see and know” when they have lost the focus of attention on the object, or emotion has become reactive or ruminative. Likewise, the same meta-awareness serves to determine whether there is dullness or too much excitation during the practice and therefore, contributes to the phenomenal intensity or clarity in which the meditation is experienced (Lutz et al., 2007). Without such a highly developed sense of self-awareness, the contents of subjective experience are continually entangled with the patterns of conditioned and consolidated schemas that dictate behavior. Husserl refers to this abstracting from all objects as “epoche,” or bracketing objects of conscious experience in order to reflect on the contents within it (Varela et al., 1991). Meta-awareness may differ from the EPS only in its relation to the world, from which one is able to study the intentional contents of the mind in a transcendent, non-conceptual way.

Recent studies have investigated the role of short- and long-term mindfulness-based meditation practice on the subcomponents of the attentional system (alerting, orienting, engagement, and disengagement), and at the early perceptual and non-conscious stages of processing external stimuli that could reflect changes in attentional bias (Brefczynski-Lewis et al., 2007; Chan and Woollacott, 2007; Slagter et al., 2007; Tang et al., 2007; Cahn and Polich, 2009; Hodgins and Adair, 2010; van den Hurk et al., 2010b; Ganaden and Smith, 2011; Vago and Nakamura, 2011). One study that utilized the Attention Network Test (ANT) found significantly better executive monitoring at baseline in a group of experienced meditators (4-week intensive retreat) than control subjects (Jha et al., 2007). After MBSR, individuals demonstrated improved endogenous orienting ability, while the 4-week retreat group showed improved exogenous alerting-related processes, reflecting the advanced training in receptive forms of practice (Jha et al., 2007; Ganaden and Smith, 2011). van den Hurk et al. (2010a,b) used the same ANT task with long-term (~15 years. practice) meditators and found improved orienting and executive monitoring ability in comparison to age-matched controls (van den Hurk et al., 2010a). Five days of IBMT, a receptive practice also demonstrated improvements in executive attention using the ANT task (Tang et al., 2007). Several recent studies have also reported improvements in sustained attention during the performance of other exogenous cueing tasks, or continued performance tasks. One longitudinal study (3 months, 5 h/day) found that retreatants showed enhanced perceptual discrimination and vigilance, an observation that remained constant 3 months after the retreat ended (MacLean et al., 2010). These results suggest a form of plasticity in attention-related networks that can manifest between 5 days and 3 months of intensive mental training.

Alerting attention toward a stimulus can operate at very early levels of perceptual processing (<300 ms) and is sometimes considered an automatic process, taking the form of attentional bias toward, or away from, an object (Mogg et al., 1995). A few hundred milliseconds before an eye movement, visual attention is directed toward the forthcoming target locations, shifting activations in saccade and attention areas of the brain and enabling planning of actions toward those locations (Deubel, 2008). In a sample of female fibromyalgia patients, a preliminary study of mindfulness meditation on attentional bias revealed decreased avoidance at early stages of attention toward pain-related cues in those patients exposed to 8-weeks of mindfulness training (Vago and Nakamura, 2011). Other studies have also showed the influence of meditation practice on early phases of non-conscious attention (Srinivasan and Baijal, 2007; Cahn and Polich, 2009) and bias (Garland et al., 2010; Paul et al., 2012). For example, Srinivasan and Baijal (2007) demonstrated increased mismatch negativity (MMN) amplitudes in concentrative meditation (Sahaj Samadhi) practitioners immediately before and after practice, and in contrast to novice practitioners. These studies suggest enhanced perceptual sensitivity even before conscious forms of attention are allocated.

As attentional effort decreases over time and practice (see Figures 3 and 4), the efficiency of attentional networks improves and more cognitive resources are made available. Increased efficiency (decreased effort) of attentional networks has been demonstrated elegantly in reduced resource allocation in one study of subjects who underwent three months of intensive mindfulness meditation training and were tested using an attentional-blink paradigm (Slagter et al., 2007). Attentional blink refers to a deficit in perception of a second target (T2) when presented in rapid succession (<500 ms) following an initial target (T1) embedded in a stream of stimuli presented in close temporal proximity (Slagter et al., 2007). This deficit is believed to result from competition between the two targets for limited attentional resources. A smaller attentional blink and reduced brain-resource allocation to T1 was reflected by a smaller T1-elicited P3b, a brain-potential index of resource allocation (Slagter et al., 2007). Interestingly, those individuals that showed the largest decrease in T1-elicited P3b amplitude generally showed the greatest reduction in attentional-blink size. Another study demonstrated that expert meditators with least hours of experience (10,000–24,000 h) showed increased activity (compared to novices) in the left DLPFC during FA practice (vs. rest), while expert meditators with the most hours of practice (37,000–52,000 h) showed decreased activity in left DLPFC than both groups. Other studies have specifically found improvement in efficiency, reduced interference and increased control. Manna and colleagues (2010) found that right DLPFC deactivation was positively correlated with expertise, suggesting decreased effort in advanced practitioners during FA meditation. Relative meditation experience (ranging from 82 to 19,200 h) has been positively correlated with reduced interference on the Stroop task (Chan and Woollacott, 2007), although Stroop interference was not found in a group of meditators trained in an eight-week MBSR course (Anderson et al., 2007). These data suggest that as meditation practice becomes increasingly effortless in advanced practitioners, more efficient allocation of resources is observed.

Emotion Regulation

There is growing evidence that mindfulness training improves emotion or self-regulation skills as evident from a variety of self-report, physiological, and neuroimaging methods (Baer et al., 2009; Carmody, 2009). One explanatory mechanism underlying alterations in health, brain structure and function may rest on the fact that mindfulness-based meditation practices strengthen neural systems important for emotion regulation, specifically evaluative, expressive, and experiential aspects of emotion. Emotion regulation is therefore also very similar, if not partially redundant with the concept of self-regulation, referring to the ability to shift focus of attention at will, and modulate ongoing emotional activity (e.g., decrease elaborative processing of thoughts/feelings) (Gross, 1998; Northoff, 2005; Carver and Scheier, 2011; Koole et al., 2011). In this context, emotion is measured through multiple components including cognitive, viscerosomatic, behavioral, and physiological responses.

In several publications, Davidson (2000, 2004) has referred to emotion regulation in terms of affective style and has described opportunity for regulation at various dynamics of affective chronometry: (1) the threshold to respond; (2) the magnitude of the response; (3) the rise time to the peak of the response; (4) the recovery function of the response; and (5) the duration of the response. For example, one can consider the recovery function of the response to be rapid (steep) and time to the next response as a refractory phase that may be reduced in practitioners of mindfulness-based meditative techniques. Regulation has been described in the literature through automatic and controlled strategies (Parkinson and Totterdell, 1999; McRae et al., 2011). Automatic regulation refers to the modulation of affect-related variables at a non-conscious level. For example, homeostatic mediators associated with the stress response and involved in generating affect can be non-consciously regulated through training. Controlled regulation can include antecedent-focused or response-focused strategies that require some behavioral or cognitive process. Antecedent-focused strategies focus on controlling the selection or modification of the context to avoid the emotion altogether or to modify the emotional impact (e.g., by performing a secondary distracting task, or cognitive reappraisal) (McRae et al., 2011). Response-focused strategies only affect the output of appraisal process and either augment or suppress behavioral manifestations of one’s emotional state, such as smiling, frowning, or experiential avoidance (Gross, 1998; Koole et al., 2011). Many of these controlled strategies may be used in novices when emotional reactivity arises as a form of distraction; however, cognitive strategies are less likely in more advanced practitioners. Non-cognitive strategies are typically encouraged in dealing with the arising and passing of emotions during practice, even when the emotions are particularly intense or stressful. For example, mental noting and labeling the modalities of the emotional experience becomes the first line of defense, a technique shown to reduce amygdala activity and emotional expression (Lieberman et al., 2007). This is more likely an expression of equanimity that naturally arises through continued development of meta-awareness.

The S-ART framework supports the notion that mindfulness-based practice can strengthen controlled emotion regulation processes early on in novice practitioners. Most research in the area of emotion regulation includes a focus on various subdivisions of the PFC that have most consistently been implicated in emotion and cognitive control processes, including decision-making, appraisal, and impulse control. These areas include the OFC, DMPFC, ACC, DLPFC, and VLPFC (Davidson and Irwin, 1999; Ochsner and Gross, 2005; Phillips et al., 2008). Across a number of different MBI studies (Lazar et al., 2005; Holzel et al., 2008; Luders et al., 2009; Grant et al., 2010a; Holzel et al., 2010), differences in GM volume and density (in comparison to non-meditating controls) have been found using MRI in OFC and lateral PFC. Increased activity in prefrontal areas has also been found in experienced meditators (compared to non-meditators) during meditative states (Jevning et al., 1996; Lazar et al., 2000; Holzel et al., 2007, 2011b; Newberg et al., 2010), in some cases specifically in lateral PFC (Baron Short et al., 2010; Farb et al., 2007; Raffone and Srinivasan, 2010). There is some evidence to suggest that abnormal GM concentration and/or functional activity in the left sgACC may be indicative of dysfunctional emotion regulation ability and potential for developing depressive symptomology (Drevets et al., 1997). Other areas found to be critical in their respective interactions with the PFC are the amygdala, hippocampus, striatum (including the nAcc), thalamus, and insula (Davidson and Irwin, 1999; Phillips et al., 2008). The uncinate fasciulus (UF) is a tract of fibers connecting limbic structures to the PFC which is highly plastic over the course of development and critical for emotion regulation and transformation. There is growing evidence that dispositional mindfulness may act as a marker for improved emotion regulation skills (Creswell et al., 2007). For example, increased dispositional mindfulness (measured by the MAAS) has been correlated with increased activation of VLPFC and attenuated activation in the amygdala (Lieberman et al., 2007). Garland and colleagues (2009, 2011) propose that positive reappraisal, a cognitive coping strategy, is a mechanism by which mindfulness functions to regulate emotion and stress. One study supports these claims by demonstrating dispositional mindfulness (assessed by the KIMS) positively correlated with bilateral activation of DMPFC during reappraisal of negative pictures (Modinos et al., 2010). It remains unclear however, to what extent reappraisal is used in mindfulness practice given the emphasis on non-cognitive processing. Other studies have demonstrated evidence for non-appraisal strategies associated with mindfulness (Grant et al., 2010b; Gard et al., 2012). In support of more non-cognitive forms of emotion regulation, reduced activity in executive, evaluative and emotion areas during acute pain (PFC, amygdala, hippocampus) have been found in adept practitioners of Zen, compared to controls (Grant et al., 2010b). Meditators with the most experience showed the largest activation reductions, suggesting a decreased need for effortful control in advanced meditators. Importantly, the lower pain sensitivity and higher thresholds for pain in meditators was strongly predicted by reductions in functional connectivity between executive and pain-related cortices. Results suggest a functional decoupling of the cognitive-evaluative and sensory-discriminative dimensions of pain, possibly allowing practitioners to view painful stimuli more neutrally and with equanimity (Grant et al., 2010b). Grant et al. (2010a,b) also demonstrated increased cortical thickness in dACC that positively correlated with lower pain sensitivity in the Zen meditators, suggesting an ability to monitor and express emotions related to pain without a high negative valence (Grant et al., 2010a).

The S-ART framework also supports the idea that mindfulness-based practice can improve automatic forms of regulation like homeostasis in the face of emotional or physical stress by protecting the internal milieu from the harmful effects of a stressor, which can be referred to as the “raincoat effect,” and by facilitating recovery, which we refer to as the “towel effect.” The raincoat acts as a metaphor for protection, as it protects one from getting wet; the towel acts as a metaphor for recovery, as it facilitates drying off when one already has gotten wet. These metaphors have been used similarly to previous theoretical models for psychoneuroimmunology (Ader et al., 1987; Feder et al., 2009), and are used here to describe the expression of equanimity that is tightly-coupled with the skills of mindful awareness. Equanimity is supported by psychological processes like decentering and non-attachment, de-coupling the sensory and affective components of the stressor. As a result, sympathetic tone is reduced and hypothalamic-pituitary-adrenal (HPA) axis-mediated mobilization is suppressed. A decreased stress response to innocuous cues and rapid return to physiological and emotional baseline in response to real threat should be apparent across all physiological mediators such as the catecholamines (e.g., epinephrine and norepinephrine) from the adrenal medulla, glucocorticoids (e.g., cortisol) from the adrenal cortex, pituitary hormones (e.g., ACTH, prolactin, and growth hormones) and cytokines (e.g., IL-1, IL-6, and TNF-α) from cells of the immune system. Mindfulness training is proposed to reduce the chance for pathophysiological processes or unfavorable psychopathological outcomes by preventing perseverative and chronic responses in the presence or imagined presence of a stress-related physiological challenge and through effective habituation to repeated stress-related challenges (McEwen, 2008). This form of self-regulation prevents “allostatic load,” the cumulative wear and tear on the body and brain due to the overactive or inefficiently managed stress response (McEwen, 1998; Sterling and Eyer, 1988). Long-term adverse effects are avoided, including immunosuppresion, cardiovascular dysfunction and disease, accumulation of abdominal fat, loss of bone minerals, reproductive impairments, decreased neurogenesis, increased neuronal cell death and associated atrophy in the limbic system (Jameison and Dinan, 2001; Sapolsky, 2003; McEwen, 2008).

The connections of the insular cortex with the thalamus, HPA axis and brainstem are thought to provide a mechanism for viscerosomatic and homeostatic feedback to the executive monitor and regulation of sympathetic tone (Critchley, 2005; Craig, 2009). In order to maintain homeostasis, the internal state of the body is critical for the parasympathetic division of the autonomic nervous system (ANS) to respond effectively to stressors. There are two parasympathetic nerve centers in the brainstem medulla: dorsal and ventral. The vagal nuclei promote parasympathetic tone, increased energy conservation through control of cardiovascular and visceral function. The ventral vagal nuclei dampen the sympathetic response specifically by targeting cardiorespiratory organs; whereas, the dorsal nuclei target the gut and associated viscera (Porges, 1995). The S-ART framework predicts mindfulness can facilitate both dorsal and ventral vagal tone through FA on the body in the context of stressors.

There have been a number of studies reporting on the physiological profile of meditators, which appears to be significantly influenced by meditation practice, and may strongly contribute to the neuroimmunological profile as well. For example, studies comparing experienced meditators to controls or short-term meditators have demonstrated a particular physiological profile across styles of practice suggestive of an alert, but hypometabolic state in which there is decreased sympathetic nervous system activity, and increased parasympathetic activity (Young and Taylor, 1998; Benson, 2000; Cahn and Polich, 2006). Across meditation styles, electromyography has revealed muscle relaxation in spite of the upright, unsupported posture (Austin, 2006). Skin conductivity, along with oxygen consumption, heart rate, blood pressure, cortisol, muscle tone, urinary vanillyl mandelic acid (VMA) (a catecholamine metabolite), and rate of breathing all decrease during meditation (Jevning et al., 1992; Benson, 2000; Lazar et al., 2005; Austin, 2006). Critically, this hypometabolic state has been shown to be qualitatively and quantitatively different from simple rest or sleep and more suggestive of a role in prevention of stress-related hypertension or cardiovascular disease (Young and Taylor, 1998; Cahn and Polich, 2006). Other more advanced meditation techniques focus specifically on altering control of breath (e.g., pranayama), and body-temperature (e.g., Tummo), techniques which may be more highly specified toward impacting automatic regulatory processes. During seated meditation, at a middle level of samatha, the breathing rate may drop to only two or three breaths a minute (Lazar et al., 2000; Austin, 2006), while normal adults at sea level breathe about fifteen times per minute. The specific and non-specific effects of mindfulness-based meditation on homeostatic regulation are clearly an aspect of self-regulation that needs further investigation.

Physiological studies have also supported the effects of meditation on non-reactivity related to a rapid change back to baseline after an emotional response. This rapid regulatory mechanism is proposed to be an objective measure of equanimity. For example, experienced transcendental meditators (>2 years experience) showed more rapid decreases in skin conductance following aversive stimuli (Goleman and Schwartz, 1976), while other studies have shown decreased startle amplitude (Delgado et al., 2010; Levenson et al., 2012), and other low-level bottom-up forms of emotion regulation (van den Hurk et al., 2010b). Goldin and Gross (2010) found that social anxiety patients show a more rapid decrease in activation of the amygdala in response to negative self-beliefs after a MBI (Goldin and Gross, 2010). Additionally, Britton et al. (2012) demonstrated more rapid decreases in self-reported state anxiety following a psychosocial stressor (in comparison to pre-MBCT) (Britton et al., 2012). Other studies investigating the effects of meditation on homeostatic regulation have found improved function and effective allocation of inflammatory responses in novice normal meditators and in subjects with major disease diagnoses (e.g., cancer) (Kabat-Zinn et al., 1998; Davidson et al., 2003b; Smith, 2004; Carlson et al., 2007; Pace et al., 2010). For example, Davidson and colleagues (2003b) found that MBSR produced significant increases in antibody titers to influenza vaccine compared with those subjects in a wait-list control group. Interestingly, the magnitude of increase in left-sided anterior PFC activation in EEG predicted the magnitude of antibody titer rise to the vaccine (Davidson et al., 2003b). Pace and colleagues (2010) reported that after 6-weeks of compassion meditation practice, there were reduced innate cytokine (IL-6) and subjective distress responses to a standardized laboratory psychosocial stressor [Trier Social Stress Test (TSST)]. Moreover, individuals with meditation practice times above the median exhibited lower TSST-induced IL-6 and POMS distress scores.

Extinction and Reconsolidation

The elimination of suffering, the end state of mindfulness practice is described often as “stillness of the mind” (Sanksrit: nirvana) (Buddhaghosa, 1991). The word “Nirvana” has literally been translated as “blowing out” or “extinction.” In this context, it refers to the extinction of the afflictions (Sanskrit: klesha), which prevent happiness and flourishing. (Buddhaghosa, 1991; Analayo, 2003; Bodhi, 1999). Maladaptive habits, distorted perceptions, and biases accumulate through the conditioning or reification of the NS, most of which are not accessible to conscious awareness. The narrative one creates about oneself in terms of self-reflection or future projection becomes increasingly more rigid as it is conditioned over time through a causal chain of repetition (see Figure 1). Each trajectory of self-development represents a repeatedly reconstructed, reinforced, and reified NS with reliable patterns of subject-object relations that are relative stable and accessed during self-specifying processes.

The S-ART framework suggests mindfulness acts as the master self-regulatory mechanism for de-coupling and efficiently integrating experiential and NS modes of processing with the potential to transform the reified self from maladaptive trajectories into more positive, adaptive trajectories. This form of transformation is hypothesized to use circuitry associated with extinction and reconsolidation. Biases of attention and memory related to habitual distortions are proposed to be extinguished and reconsolidated; however, the dosage and quality of meditation time required for such change remains unclear. Although change in constitutionally-based temperament or biological disposition is thought to be less likely than those aspects of self that can be modulated by experience and training (Rothbart and Ahadi, 1994; Kagan, 2003), there is evidence that conditioned fear can be extinguished (Phelps et al., 2004). Exposure, extinction, and reconsolidation are critical mechanisms for reducing habitual anxiety and fear, and in facilitating therapeutic change in past clinical models (Wells and Matthews, 1996; Bishop, 2007; Beck, 2008). Extinction does not erase the original association, but is a process of novel learning that occurs when a memory (explicit or implicit) is retrieved and the constellation of conditioned stimuli that were previously conditioned to elicit a particular behavior or set of behavioral responses is temporarily labile and the associations with each other are weakened through active or passive means. Changing the association contingency of contextual cues and what they mean results in a new memory trace. The new memory trace involves re-consolidated associations of particular contextual stimuli and its previous associations with new stimuli and behavior (Nader et al., 2000; Quirk and Mueller, 2008). The behavioral expression of the new memory is then thought to be in competition with the previously conditioned memory, with particular pathological-oriented biological dispositions proving to increase the time or number of trials necessary for extinction and sensitivity to old fear-based associations (Milad et al., 2007). Extinction and reconsolidation can depend on numerous factors, such as level-of-processing (Craik, 2002), emotional salience (Kensinger and Schacter, 2005), the amount of attention paid to a stimulus (Loftus, 1979), the expectations at encoding regarding how memory will be assessed later (Frost, 1972), or the reconsolidation-mediated strengthening of memory trace (McGaugh, 2000). The closer in time the retrieval occurs to the experience, the more likely weaker components of a memory trace can be reconsolidated given the tendency for weak traces to decay over time (i.e., forgetting). It has been reported that the hypometabolic state of the practitioner may act to facilitate the extinction process by creating novel parasympathetic associations with previously anxiety-provoking stimuli (Cahn and Polich, 2006). The noting and labeling of sensory, cognitive, and emotive states while in a hypometabolic system, such as during OM practice, is also likely to influence the nature of conditioned response toward reconsolidation of more positive trajectories. Awareness alone is proposed to change the conditioned response contingency toward one’s patterns of behavior and feelings toward oneself and others, although intentional cognitive processes may also contribute. In association with positive reappraisal strategies, the practitioner may reflect on a particular emotion and decide if it is desirable or undesirable, warranted or unwarranted. If one finds that one would like to change something about the emotional state one is affected by, one can intentionally act to transform the emotions. If, indeed, extinction proves to be a common mechanism for many types of habitual sensory, affective, and motor processing in mindfulness training, one can begin to redefine the solid boundaries and limited interactions between automatic and controlled processing.

In both rodents and humans, the brain areas involved in conditioning and extinguishing fear include the hippocampus, amygdala, rhinal cortices, and VMPFC. Right DLPFC has also been implicated in the extinction of aversive associations, while right VLPFC has been implicated in non-aversive extinction recall (Morgan and Ledoux, 1999). The cerebellum contributes to simple associative learning, while the inferior temporal cortices and dorsal striatum are responsible for conditioning and habitualizing more complex sensory-affective-motor scripts and schemas in visuo-spatial contexts (Buckner and Wheeler, 2001). Structural connectivity between DLPFC and amygdala has also been implicated in learned safety, suggesting effortful control over expression of fear-related behavior (Pollak et al., 2010). The VMPFC appears to modulate the amygdala response and extinguish the expression of fear in this functional circuitry. In fact, cortical thickness of the VMPFC has been shown to positively correlate with extinction recall, and functional connectivity with the hippocampus appears to be related with learned safety (Milad et al., 2007). The DMPFC and dorsal ACC have been shown to be modulatory over the physiological and behavioral expression of fear and thus, may be critical in extinction and reconsolidation as well (Quirk et al., 2010). Dorsal ACC has been shown to be positively correlated with sympathetic activity related to fear expression; whereas, ventral ACC (including pACC and VMPFC) has been more associated with inhibition of expression during extinction, and recall of extinction of fear after extinguishing prior conditioned associations (Etkin et al., 2011).

On a more cellular level, extinction and reconsolidation research has focused on the molecular cascade of events that follows experience and results in reorganization of long-term memory storage. Some lines of investigation have demonstrated that the presence of kinase activity, which persists in order to maintain synaptic potentiation, may be indicative of new memory traces (Wallenstein et al., 2002). Similarly, labeling of epigenetic or transcription factors related to expression of various genes necessary for synaptic plasticity has also been shown to be indicative of extinction and reconsolidation processes (Sweatt, 2009). Several recent studies have shown that extinction learning in rats can be accelerated and strengthened through modulation of these molecular systems as well as through pharmacological noradrenergic and dopamine systems [e.g., D-cycloserine (DCS)] in the mPFC (Quirk et al., 2010).

Although relatively few studies have explored this possibility, there is some evidence that mindfulness-based practices involve exposure, extinction, and reconsolidation processes [see (Holzel et al., 2011a; Treanor, 2011) for review]. Multiple morphometric studies have demonstrated structural changes in the circuitry related to extinction following mindfulness training of as little as 8-weeks (Lazar et al., 2000; Holzel et al., 2008, 2011b; Luders et al., 2009). Cross sectional studies comparing meditators and non-meditators found greater GM concentration in the hippocampus of meditators, suggesting enhanced circuitry for extinction learning and retention (Holzel et al., 2008; Luders et al., 2009). Aside from the changes noted above in memory-related structures, thickness of the mPFC has been found to be directly correlated with extinction retention after fear conditioning, suggesting that its increase in size following training might explain ability in meditators to modulate fear, a mechanism that has yet to be fully explored (Milad et al., 2005; Ott et al., 2010a; Holzel et al., 2011b). Moreover, GM concentration in this region has been correlated with the amount of meditation practice. Holzel and colleagues also recently demonstrated that decreased perceived stress over the 8-weeks of MBSR positively correlated with decreased GM concentration in the right amygdala (Holzel et al., 2011b). Previous research has shown that such morphometric changes are associated with improved emotion regulation and extinction of fear (Milad et al., 2005; Etkin et al., 2011). Functional increases in these regions have also been found during meditation or in the context of emotional probes, suggest a strengthening of the circuitry involved in extinction and a critical role for PFC-hippocampal dynamics in mediating changes in advanced practitioners. We further propose that the posited reductions of negative ruminations about the self reportedly due to mindfulness training (Ramel et al., 2004) may also be mediated by the same functional extinction circuitry.

Prosociality: Improving Social Cognition

Humans are inherently social beings with a gifted ability for social cognition—to understand others’ emotions, intentions, and beliefs. Empathic forms of behavior, including empathy, sympathy, and altruism, have been implicated in conceptual models and theories about social cognition and associated experience sharing and prosocial behavior (Hein and Singer, 2008; Eisenberg et al., 2010; Zaki and Ochsner, 2012). Prosocial behavior is typically defined as voluntary behavior intended to benefit another (Eisenberg et al., 2010). Major facets of prosociality and inter-relationships include forms of mentalizing and perspective taking. For example, imagining the intentions of others has been called “theory of mind” (ToM). Empathy is distinct from ToM and refers to our ability to share the experiences (emotions and sensations) of others. It is believed that mindfulness practice can cultivate a framework for interdependence of self in a social network and include a cognitive framework supporting empathy and mentalizing. In such cases, the conceptual form of meta-awareness (i.e., meta-cognition) allows one to disengage from the contents of awareness and move toward the experiencing of an other’s sensory or affective state (Decety and Chaminade, 2003; Singer and Lamm, 2009).

Although prosociality has a biological basis and dispositional differences in empathic responding and prosocial action exist from early life onwards (Rothbart and Ahadi, 1994), patterns of empathic responding and related prosocial behavior are relatively plastic not only during childhood and adolescence, but also during adulthood. Specifically, studies of adults have revealed enhancements in functional activity of a social-cognitive network supporting facets of mentalizing and empathy (Singer and Lamm, 2009; De Greck et al., 2012; Fan et al., 2011). This suggests that some of the neural circuitry underlying empathy can be enhanced through the practice of forms of mental training designed to increase positive affect and prosocial behavior across the lifespan. It is now commonly believed that empathic ability affects the perception and affective style of others’ behavior toward oneself (Eisenberg et al., 2010), mutually reinforcing the behavior toward each other. Thus, social and moral cognitions and behavior reflect the nature and valence of self and others’ empathic behavior. The S-ART model proposes that plasticity associated with prosociality may be indicative of self-transcendence, dissolving distinctions between self and other and reflecting loving-kindness to both. A greater sense of well-being is likely to depend upon this skill. In fact, altruistic behavior has been associated with greater sense of well-being and acceptance (Lyubomirsky et al., 2005), suggesting all other stated mechanisms can be modulated by the cultivation of prosocial and ethical behavior.

Brain areas that are involved in mentalizing include TPJ, TP, AIC, precuneus, DMPFC, while the ACC, IFG, pSTS, and IPL have been implicated more with prosociality and empathic concern and experience sharing (Singer and Lamm, 2009; Frith and Frith, 2012; Fan et al., 2011; Roy et al., 2012). Imagined self-action relative to imagined experimenter action strongly activates IPL (including TPJ), somatosensory cortex, and precuneus (Ruby and Decety, 2001; Farrer et al., 2003), while Trancranial Magnetic Stimulation (TMS) or lesions of the right TPJ has shown to disrupt the sense of ownership or agency (Tsakiris et al., 2010) and produce dissociative, out-of-body experiences (Blanke and Arzy, 2005). The OMPFC, has been shown to be specifically involved in social modulation of reward value, while the mPFC and TPJ have been shown to be more involved in mentalizing and ToM (Frith and Frith, 2012). The right IPL is more often associated with ToM, whereas the left IPL is more often involved in representing one’s own mental states (Decety and Chaminade, 2003). The pSTS (including the TPJ) has been shown to be involved in perceiving biological motion and orienting toward eye gaze (Frith and Frith, 2012). The ACC has been shown to be more involved in a supervisory role and deciphering conflicting information. Posterior IFG has been shown to be involved in emotional judgment and has been suggested to play a role in emotion recognition, predicting emotion, and mirroring action (Frith and Frith, 2012). Fan and colleagues (2011) found that the dorsal ACC and anterior mid-cingulate cortex (aMCC) were recruited more frequently in the cognitive-evaluative form of empathy, while the right AIC was found to be involved in the affective-perceptual form of empathy only, and the left AIC was found to be active in both forms of empathy. Singer and others propose the AIC serves a dual role, a primary mapping of internal states with respect to the self, and the predictive representations or simulation of how the same emotional stimuli feel to others (Singer and Lamm, 2009).

One suggestion to account for overlap between self-other neural substrates has been that understanding the intentions of another is also likely to recruit processes underlying insight into oneself. Interestingly when perceivers passively watch targets experiencing pain or reward, their own engagement of brain areas associated with those states predicts later prosociality (Zaki and Ochsner, 2012). This suggests that increased prosociality may predict decreased differences in the intensity and localization of neural activity in response to one’s own experience in contrast to some target individual’s experience. This would support a system of experience sharing or prosocial concern; whereas mentalizing may engage separable neural systems. In an experiment by Masten et al. (2010), mentalizing is shown to increase future helping behavior, suggesting a distinct system supporting prosocial behavior through experience sharing and mentalizing. The prosocial system is indicated in Figure 2. Research also indicates that positive social behaviors, specifically maternal nurturing (e.g., licking and grooming), induces positive alterations in the brain and in behavior that promote resilience (Champagne and Curley, 2008). Psychosocial factors, such as decreased levels of denial and avoidant coping behavior, increased levels of social engagement, positive emotion, and dispositional optimism have all been shown to promote resilience (Feder et al., 2009). Collectively these data establish that our brains are continuously shaped both functionally and structurally by experience on which explicit training can capitalize to promote more adaptive brain functioning, especially with relation to prosocial behavior.

In support of the development of prosocial behavior, including mentalizing, empathic concern, and experience sharing, there have been a number of studies to suggest meditators show increased activation of the prosociality circuitry over non-meditators. Lutz and colleagues studied novice and expert meditation practitioners that generated a non-referential loving-kindness-compassion meditation state. To probe affective reactivity, emotional and neutral sounds were presented during the meditation and comparison periods. Concern for others cultivated during this form of meditation enhanced affective processing particularly in response to sounds of distress, and this response to emotional sounds was found to be modulated by the degree of meditation training. The presentation of the emotional sounds was associated with increased pupil diameter and dramatic increased activation of right AIC during meditation (vs. rest) (Lutz et al., 2008a). During meditation, activation in right AIC and mid-insula was greater during presentation of negative sounds than positive or neutral sounds in expert vs. novice meditators. The strength of activation in insula was also associated with self-reported intensity of the meditation for both groups. Insula activity was also strongly coupled to heart-rate, suggesting a relationship between generation of compassion and cardiac function (Lutz et al., 2009). One case study of compassion meditation showed similar activity in areas related to empathic processing (Engstrom and Soderfeldt, 2010). These results and other similar studies that demonstrated neural resonances in shared experience [see (Zaki and Ochsner, 2012)] support the role of limbic circuitry in emotion sharing. The comparison between meditation vs. rest states between experts and novices have shown increased activation in amygdala, right TPJ, and right pSTS in response to all sounds, suggesting greater detection of, and enhanced mentation in response to emotional human vocalizations for experts in comparison to novices during compassion meditation. Interestingly, there was a link between expertise in compassion and the activation in the right pSTS, a finding that is consistent with previous research that indicated pSTS activity predicted self-reported altruism (Tankersley et al., 2007). Increased activity in the amygdala may appear to be a counterintuitive finding to traditional models of controlled emotion regulation, but may be more indicative of regulatory states like equanimity.

Together these data indicate that the mental expertise to cultivate empathic behavior and positive emotion for others’ suffering involves a network of prosociality and social cognition.

Non-Attachment and de-Centering

One of the advanced outcomes or aspects of insight, developed through mindfulness-based meditation practice is the realization of “no-self,” which stems from the Buddhist teachings describing the nature impermanence (see Section “Defining meditation and mindfulness from the historical perspective”). This form of insight demonstrates that there is no truly existing self (i.e., subject) that continues through life without change and provides the practitioner with the critical distinction between the phenomenological experience of oneself and one’s thoughts, emotions, and feelings that appear “thing-like” (Varela et al., 1991). This realization of impermanence of all “thing-like” objects including the self is also described as a “release from mental fixations,” or non-attachment (Sanskrit: virāga) (Sahdra et al., 2010). In this sense, mindfulness is a relational process. Mindfulness not only helps to support awareness of the self, it transcends the self-object duality by supporting the realization of the self to be co-dependent with the relations to objects in experience, resulting in a characterization of self as empty and groundless. There may be clear benefits to developing increased awareness of our selves and environment, but the platform from which sustained transformation and insight arise is one that includes the relational quality the practitioner brings toward noting and labeling the modalities of experience—one that is free of the choosing, evaluating, or projecting that is sometimes described as “grasping, aversion, and delusion” (Salzberg, 2011).

Decentering, described as “reperceiving” in past models of mindfulness (Shapiro et al., 2006), is a therapeutic process that introduces a “space between one’s perception and response” allowing the individual to disengage or “step outside” one’s immediate experience in an observer perspective (i.e., fly on the wall) for insight and analysis of one’s habitual patterns of emotion and behavior. Decentering may not necessarily be unique to mindfulness, as it is described in other therapeutic contexts as the “observing self” (Deikman, 1982; Fletcher and Hayes, 2005), or “self-as-witness” (Damasio, 2010). Decentering is compared to clinical constructs such as defusion or psychological distancing (Fletcher and Hayes, 2005; Ayduk and Kross, 2010). It also refers to a process of deautomatization (Ayduk and Kross, 2010), in which there is an undoing of automatic processes that control perception/interpretation. These processes are described to enhance a trait-like ability to shift focus of attention at will, and inhibit elaborative processing of thoughts/feelings. For example, repetitive thinking “I am worthless” vs. recognizing “I am having the thought that I am worthless.” The fusion of self and negative thoughts along with rumination has been shown to play a critical role in exacerbating negative affect, maintaining or heightening anxiety, and increasing cognitive vulnerability to psychopathology (Smith and Alloy, 2009). The insight achieved through mindfulness-based practices provides awareness that one’s thoughts are subjective and transient in nature (Safran and Segal, 1990), thus facilitating non-attachment and subsequently improving satisfaction with life, well-being and interpersonal functioning (Sahdra et al., 2010).

Laboratory studies of mindfulness-based meditation practices have yet to identify biological correlates for the psychological constructs of decentering and non-attachment. However, the S-ART framework suggests a significant role for them in relation to practice effects and development of meta-awareness. Some of the extant data suggest functional dynamics between the self-networks may account for the decentering mechanism by which mindfulness training promotes flexibility of information processing between autobiographical and experiential awareness. For example, one study found that the DMPFC is deactivated and negatively coupled to PIC during interoceptive awareness (Farb et al., 2012). Additionally, Farb et al. (2012) found that DMPFC-insula connectivity was absent during exteroceptive focus. Another study investigating adept Zen meditators (in comparison to novices) found stronger connectivity between DMPFC and IPL, an area with both narrative and EPS involvement (Taylor et al., 2012). These data suggest a mechanism by which DMPFC may act to integrate narrative and experiential information, sustain positive activation in interoceptive areas or decouple from them without suppression. Other studies have emphasized areas from the integrative fronto-parietal network as critical for the type of perspective taking required for decentering or non-attachment. For example, one study demonstrated a strong role for dACC in mediating the conscious appraisal of emotional experience (Critchley et al., 2004), while another study showed strong activity in dACC and DMPFC together (Holzel et al., 2007). Future studies will have to clarify the neurobiological mechanisms behind these critical processes.

Conclusions, Challenges, and Future Directions

Rather than focusing on reducing mindfulness down to a single unitary dimension of cognition, we have attempted to illustrate the complexity of mechanisms by which mindfulness functions to reduce suffering and create a sustainable healthy mind using a framework of self-processing. Suffering is described using both a traditional Buddhist perspective and a contemporary model for psychopathology. Biased self-processing is demonstrated to be a common target for traditional Buddhist and contemporary conceptual models of mindfulness, such that discrepancies in operationalizing mindfulness can find common mechanisms by which mindfulness-based mental training functions. In our attempts to create a unified framework for understanding the mechanisms by which mindfulness functions, we operationalize the concept in two ways: (1) as a broadly defined method for developing a multidimensional skillset that ultimately leads to a reduction in self-processing biases and creates a sustainable healthy mind; (2) A continuous discriminative attentional capacity that we refer to as “mindful awareness”—a skill amongst a set of other skills developed through specific meditation practices. Our interpretation of mindfulness is thus provided by an empirical framework of S-ART to illustrate a method for becoming aware of [and familiar with] the conditions which cause [and remove] distortions or biases in the individual’s construction of his or her external or internal experience. Through training in FA, OM, and EE styles of meditation, a sustainable healthy mind is proposed to be supported—reducing maladaptive emotions and cognitions common to most ordinary experience, such as lustful desire, greed, anger, hatred, worry, etc., increasing pro-social dispositions (e.g., compassion, empathy, and forgiveness) toward self and other, reducing attachments to thoughts and feelings, and removing biases inherent in habitual forms of cognition.

The S-ART framework outlines specific neural networks of self-specifying and NS processing along with an integrative fronto-parietal network that is proposed to be supported by six neurocognitive processes developed in mindfulness-based meditation practices. These processes are conceptualized as a skillset proposed to facilitate the integration of self-experience from both top-down and bottom-up mechanisms. It will be critical to examine how these networks change longitudinally through meditation training using advanced multivariate statistical network analyses that also consider changes in attentional and cognitive bias, reduction of psychological symptoms, and first-person self-reports in the neuroimaging acquisition protocols. A neurobiological model (Figure 2) is used to illustrate functional distinctions between the networks; however, gradients of non-conscious and conscious processing must be considered, as functional specification is likely to be graded in nature. The skillset identified herein is an attempt to dismantle the practices that cultivate mindfulness into component mechanisms so that the contemplative sciences can more precisely investigate the dispositional differences amongst practitioners of mindfulness-based meditation practices and practice-specific changes can be objectively correlated with first-person experience.

The challenge in this emerging field continues to be disambiguating the concept of mindfulness from more common usage of the term, and eliminating the “black box” or singular approach to studying meditation. Amongst all the enthusiasm for research and clinical application of mindfulness, there is much foundational basic science and vital clinical trials that remain to be implemented. It is likely that future research in the contemplative sciences will provide a developmental trajectory of different S-ART-specific subcomponents and individual differences across duration and frequency of practice. Although research is beginning to demonstrate the state effects related to a number of contemplative practices, the continued practice of FA and OM meditation induces trait changes that have yet to be adequately and objectively measured longitudinally across progress as a practitioner. It is assumed that one begins receiving benefits of the practices immediately upon initial practice; however, explicit durations of practice and dose effects are issues of concern that should not go unnoticed. As Davidson (2010) emphasizes, it remains uncertain whether study participants can reliably report on the quality and/or magnitude of their practice. These issues are similar for measurement of intensity or quality of practice in long-term practitioners as well. It is also critical that the emerging field examine the adverse effects of long-term practice—effects that typically go unnoticed in clinical trials, but are more prevalent in longer retreats and in practitioners that may or may not have a predisposition toward developing psychopathology. Constructing objective standards for determining extent of expertise and proficiency in a particular meditation technique beyond reported hours of formal practice (i.e., time on the cushion) will have to be developed to support self-reported facets of mindfulness.

It is our hope that this framework will help clarify adaptive mind-brain-body interactions and their therapeutic relevance in the general population and across neuropsychiatric disorders. Although past reviews have focused on the process of mindfulness from a psychological perspective, we have attempted to focus on self-processing using a neurocognitive lens, and distinguish between contemporary and historical interpretations of mindfulness. The S-ART framework also has attempted to disambiguate the term, mindfulness, from other Buddhist concepts (e.g., equanimity, clarity) and integrate them into the multidimensional skillset that is strengthened through FA, OM, and EE practices. Future work that builds off of S-ART and the process models provided will help provide a systematic way to deconstruct the heterogeneity of meditation practices into component parts and illustrate a representative map for the underlying processes of self-transformation. By revealing the neural circuitry and further identifying endophenotypes for mindfulness skill development, the field of contemplative science may help to better predict clinical outcomes and potential targets for the development of biologically based diagnostic and therapeutic strategies for those suffering with mental illness and clarify our understanding related to the nature of mind and consciousness.

Conflict of Interest Statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Acknowledgments

The authors wish to thank the reviewers (Jud Brewer and Jeremy Gray) along with Evan Thompson, Robert Roeser, Jake Davis, Cliff Saron, Richie Davidson, John Dunne, Sonia Matwin, Shinzen Young, and Emily Stern for their helpful comments on parts of this manuscript. A special acknowledgement to His Holiness the 14th Dalai Lama for his comments on aspects of the S-ART framework at Mind & Life XXIV.

References