Mindfulness is not a single thing. Different practices — and different states of mind — occupy genuinely different locations in a multidimensional space. This means that when a researcher says "mindfulness improved," or a teacher says "be more mindful," or a practitioner switches from breath focus to open awareness, they may each be pointing to different cognitive configurations. The matrix proposed here gives all of these states common coordinates — making it possible to compare, study, and communicate about them with far more precision.
↓ Explore the Spectrum of Experience — interactive map belowIn the scientific literature, the term mindfulness carries at least three distinct meanings: a stable personality trait measurable by questionnaire; a soteriological path or way of life rooted in Buddhist traditions; and a set of cognitive processes that can be deliberately trained. Each of these is useful in some context, but each also creates problems when used as the sole framework for research. Trait measures cannot detect how practice changes the quality of experience from session to session. The broad soteriological conception is too general to generate mechanistic hypotheses. And single-process cognitive definitions cannot account for the fact that Focused Attention meditation and Open Monitoring meditation are doing genuinely different things, even though both are commonly called "mindfulness."
The authors of this paper propose a different approach: treat mindfulness not as a single thing to be defined, but as a family of practices whose members can be described using a shared set of phenomenological dimensions. Instead of asking "was this person being mindful?" the framework asks: "where in a multidimensional space does this person's current mental state sit?" That shift from category to coordinates is the central move of the paper.
"We interpret it as a continuum of practices involving states and processes that can be mapped into a multidimensional phenomenological space which itself can be expressed in a neurocognitive framework." — Lutz et al. (2015)
The framework identifies seven dimensions of experience that mindfulness practices target — three primary and four secondary. The three primary dimensions are orthogonal, meaning they can vary independently of one another. The four secondary qualities describe the texture within any given configuration. Together, they form a coordinate system in which any style of practice, at any level of expertise, can be given a location.
The paper calls these dimensions "primary" because they are the main targets across all styles of mindfulness training, and because they are what most significantly distinguish different styles of practice from one another.
How strongly is experience oriented toward a specific object? FA meditation involves deliberate, sustained orientation toward one selected object (breath, sound, sensation). OM involves progressively releasing that grip, moving toward objectless awareness.
To what degree are thoughts, feelings, and perceptions experienced as mental events rather than as accurate depictions of reality? At low dereification, a thought feels like a fact. At high dereification, it is perceived as just a thought — a transient process arising in awareness.
How aware is awareness of itself? Meta-awareness is not introspection (a deliberate inward turn) but a background capacity to notice — without redirecting attention — that one has drifted, that the affective tone has shifted, that the quality of attention has changed.
This dimension concerns the phenomenological sense that an experience is oriented toward some object or class of objects. "Object" here means any content that one is aware of — a sensory sensation, a visualized image, a conceptual focus, or even a vague sense of presence. What matters is not whether an object exists in the world, but whether the practitioner's state is phenomenologically bearing on something.
In Focused Attention practice, a practitioner typically selects breath sensations at the nostrils or abdomen and sustains voluntary attention on them moment to moment. Off-target content (sounds, thoughts, bodily sensations) arises, but does not phenomenologically dominate — the state still bears on the selected object even when distractors compete. In advanced OM practice, by contrast, the explicit maintenance of a selected object is gradually released. Practitioners shift their orientation from the object toward the field of awareness itself. At the extreme, some Buddhist traditions describe states of "choiceless awareness" in which object orientation is suspended entirely — attention is wide open, and no single content is singled out. This is a different kind of attention, not a failure of attention.
During rumination, a script such as "I am a failure" may arise and feel not like a thought but like a fact — an accurate description of who one is. The same phenomenon occurs with addictive craving: the craving presents itself as a true need, not as a mental event. This is high reification — the mental contents claim to be reality.
Dereification is the capacity to perceive these contents as mental processes rather than as accurate depictions of reality — to notice that the thought is a thought, the craving is a craving, the image is an image. The paper notes that dereification is related to what is called "decentering," "cognitive defusion," or "phenomenological reduction" in the clinical and philosophical literature. It develops through mindfulness practice in two ways: in FA, dereifying distractors helps the practitioner disengage from them and return to the object without full elaboration. In advanced OM, dereification becomes so complete that even sensory experience loses some of its representational weight — thoughts are just sensory-like events, arising and passing in an open field.
Importantly, the paper emphasizes that meta-awareness and dereification are distinct and can be dissociated. Some people with anxiety disorders show heightened meta-awareness of their mental states but very low dereification — they know they are anxious, but the anxious thoughts still feel completely true. This is what makes dereification its own independent dimension rather than a byproduct of attention to the mind.
Meta-awareness as defined in this paper is not the same as introspection. When you stop to think "how am I feeling right now?", you are introspecting — you have turned attention inward deliberately. Meta-awareness is something different: a background, continuous capacity to notice features of one's experience simultaneously with whatever the primary focus is on. It is the awareness that detects, while still looking at a sunset, that there has been a subtle shift in affective tone; the awareness that notices, while still focused on the breath, that the quality of attention has just changed.
In novice FA practitioners, meta-awareness is just beginning to develop. Practitioners can detect distraction after the fact — they notice that they have wandered — but are not yet able to notice subtle perturbations before they become full-blown distractions. Expert FA practitioners can detect these finer-grained shifts early, and respond with less elaboration and less effort. In OM practice, meta-awareness becomes the primary target of cultivation rather than a supporting skill. The practitioner is no longer primarily attending to an object; they are attending to the process of attending itself. This form of sustained, background awareness is what the paper refers to as the distinguishing feature of advanced OM practice.
What it is: How wide or narrow the field of attention is — whether awareness is concentrated on one point or open to everything at once.
Small circle = narrow · Large circle = wide open
What it is: The vividness with which experience presents itself — how sharply and distinctly objects or states are perceived.
Dark fill = dim · Light fill = vivid
What it is: How persistently a state holds over time. A stable state resists disruption; an unstable one drifts or fluctuates readily.
Thin light ring = unstable · Thick dark ring = very persistent
What it is: Deliberate mental exertion needed to sustain the state. High effort means active control; low effort means the state maintains itself naturally.
Dashed stalk = effortless · Thick Y-fork = high effort
FA and OM are broad categories that encompass many styles of practice. The matrix helps show what distinguishes them — not as better or worse, but as occupying different regions of the phenomenological space and training different cognitive capacities, even when practiced by the same person in the same tradition.
FA is commonly practiced before OM: building the capacity to sustain and regulate attention on an object tends to facilitate the meta-awareness faculty that OM then relies on. However, the paper notes that this is not a mandated sequence — both can be cultivated independently, and both train meta-awareness and dereification, though by different routes. The same practitioner with the same instruction set may produce different phenomenological plots depending on individual differences in how they interpret and implement those instructions.
One of the more striking features of the matrix is that it can also plot mental states that are not meditation practices at all. Mind-wandering, rumination, and addictive craving are included in Figure 1 of the paper — not as contrast conditions to be avoided, but as states that occupy specific coordinates in the same space. Mapping them alongside FA and OM makes visible what mindfulness practice is actually moving away from — and shows that these problematic states are not simply the "opposite" of mindfulness but occupy distinct and instructive positions.
Mind-wandering is characterized by low meta-awareness (one is absorbed in the wandering content, not aware that the mind has wandered), very low dereification (the train of thought presents itself as continuous reality, not as a stream of mental events), and low object orientation in the deliberately selected sense. Crucially, it requires no effort — it is represented in the figure with a dashed stalk, indicating that it arises without deliberate intention. It is the baseline state that most styles of mindfulness training move away from, but not because wandering is inherently bad: its coordinates simply reflect a configuration in which no deliberate monitoring or dereification is active.
Rumination is represented as a state with some meta-awareness — the person knows they are ruminating — but very low dereification. The negative thoughts feel real and true and important, not like mental events that could be noticed and released. The object orientation is also high: the mind is stuck on a specific content (a failure, a threat, a regret). This combination — awareness that you are doing it, paired with the felt truth of the content — is what makes rumination so persistent and painful. It is phenomenologically distinct from mind-wandering, and the matrix shows why: they share low dereification, but rumination differs in its high object orientation and some meta-awareness.
Addictive craving is depicted at very high object orientation — the mind is strongly and repeatedly drawn toward the craved object — with low meta-awareness and very low dereification. The craving does not present itself as a passing mental event; it presents itself as a true state of the world ("I need this"). The high stability of this state in the figure reflects the persistent, intrusive, hard-to-disengage character of craving. The paper uses this state to illustrate that dereification is clinically relevant: one mechanism by which mindfulness may reduce craving is precisely by training the capacity to recognize craving as a mental event rather than as a truth about what the world requires.
The paper proposes three large-scale brain networks as particularly useful lenses for understanding the phenomenological dimensions, while explicitly cautioning against any one-to-one mapping between a dimension and a single brain region. Brain regions are multifunctional, and any phenomenological feature will likely involve interactions across multiple networks. What follows is a selective overview based on the paper's review of the neuroscience literature.
Active in goal-directed behavior, working memory, and top-down attentional selection. The paper proposes that the CEN is relevant to the object orientation dimension — sustaining voluntary attention on a selected object likely involves CEN recruitment, and functional changes in CEN nodes are reported in studies examining the attentional effects of mindfulness training.
Typically active during mind-wandering, self-referential thought, and internal mentation. Studies suggest that FA practice tends to reduce activity in this network, consistent with the selection of a sensory object and reduced engagement with narrative self-processing. OM practice does not straightforwardly suppress the DMN, as it monitors the arising and passing of discursive thought rather than actively suppressing it. The specific role of the DMN in OM remains largely an open research question, according to the paper.
Detects and prioritizes salient stimuli, including internal states. The paper proposes that the SN, particularly through the anterior insula and dorsal anterior cingulate cortex, plays a role in implementing key aspects of meta-awareness — the moment-to-moment monitoring capacity that allows practitioners to notice subtle shifts in the quality of experience, detect distraction, and monitor affective tone. The interplay between SN and CEN in this monitoring capacity requires further investigation.
The paper notes that while these three networks provide useful starting points for generating testable hypotheses, the relationship between any phenomenological dimension and neural activity is necessarily complex and many-to-many. The value of the framework is not to reduce dimensions to networks but to generate more specific hypotheses about which cognitive functions are being trained and through what mechanisms.
The authors are explicit that this framework is a heuristic tool, not a complete account of mindfulness. It is designed to facilitate research — to help investigators specify what they are measuring, communicate across traditions and disciplines, and formulate hypotheses about mechanisms. It deliberately avoids producing a single definition, because the richness and diversity of mindfulness practices across Buddhist traditions, clinical programs, and secular contexts cannot be reduced to one.
The points plotted in Figure 1 represent hypothetical positions based on phenomenological descriptions in traditional accounts and contemporary practice instructions. They are not derived from empirical first-person data — that data remains to be collected. The paper identifies this as one of the most important directions for future research: gathering structured first-person phenomenological reports from practitioners at various levels of expertise, so that the hypothetical plots can be tested and refined.
The framework is also explicitly provisional with respect to the body, culture, and motivation. A more comprehensive model would account for physical posture, proprioception, the ethical and motivational context of practice, and the specific teacher-student relationship in which the practice is embedded. These "contextual features" are shared across mindfulness practices but are not explicitly mapped by the seven dimensions of the matrix. The authors flag this as a limitation, not a dismissal of their importance.
What the matrix offers is something precise and practical: a common vocabulary for a field that has been generating conflicting results partly because different researchers, in different traditions, using different populations, have been measuring different things under the same name. A coordinate system does not settle those debates, but it makes the disagreements locatable — and that is the necessary first step toward resolving them.