Two people can both say, "I notice my body," and mean almost opposite things.
One notices a tight chest and becomes immediately alarmed. Another notices it and can stay curious. One attends closely to bodily sensation but distrusts every signal. Another rarely notices sensation until it becomes intense, yet feels able to use what they do notice. A third is highly attentive but uses attention mainly to monitor for danger.
A single question about "body awareness" cannot represent these differences.
The Multidimensional Assessment of Interoceptive Awareness, usually called the MAIA, was developed to make that problem measurable without pretending that subjective interoceptive experience is one dimension.
What the MAIA is
The MAIA is a self-report questionnaire designed to assess multiple features of how people attend to, interpret and relate to bodily sensations. The original instrument, published by Wolf Mehling and colleagues in 2012, contains 32 items organized into eight scales. A revised version, MAIA-2, published in 2018, contains 37 items while preserving eight-factor organization and strengthening weaker scales.
The eight dimensions in the MAIA family are commonly described as Noticing, Not-Distracting, Not-Worrying, Attention Regulation, Emotional Awareness, Self-Regulation, Body Listening and Trusting.
Those labels matter because they prevent one score from swallowing qualitatively different patterns.
Noticing asks whether bodily sensations are registered. Not-Distracting concerns the tendency not to ignore or distract from uncomfortable sensations. Not-Worrying concerns emotional reactivity to discomfort. Attention Regulation concerns the ability to sustain and control attention to the body. Emotional Awareness links bodily sensation with emotional states. Self-Regulation concerns using attention to bodily sensation in regulation. Body Listening concerns actively attending to the body for insight. Trusting concerns experiencing the body as safe and trustworthy.
The instrument therefore measures a reported relationship to internal experience, not the fidelity of a sensory receptor.
Why a multidimensional scale was needed
Earlier body-awareness questionnaires often treated heightened attention as uniformly beneficial or uniformly problematic.
That was conceptually unstable. In panic and health anxiety, intense bodily monitoring can amplify distress. In contemplative or somatic practices, deliberate nonjudgmental attention may be cultivated as a skill. The quantity of attention alone cannot distinguish these cases.
Mehling and colleagues' earlier construct work emphasized that body awareness had been used to describe multiple processes: perception of bodily states, attentional habits, emotional appraisal, beliefs, regulatory use and mind-body integration.
The MAIA attempted to preserve that heterogeneity rather than erase it.
This is one reason it became widely used in mindfulness, pain, mental-health, movement and interoception research.
What MAIA scores are, and are not
A MAIA score is a self-report score.
It describes how a person answers questions about their typical internal experience and attentional style. It can be valuable for studying beliefs, habits and perceived capacities that are inaccessible to an external observer.
It is not an objective test of whether the person accurately detects heartbeat timing, respiratory resistance or gastric distension.
This boundary is crucial.
A person can report strong trust in bodily signals and still perform poorly on a laboratory detection task. Another can perform well on a cardiac task while reporting that bodily sensations are frightening or difficult to use.
Neither result invalidates the other. They concern different layers.
Development of the original instrument
The 2012 development work combined conceptual review, item development and psychometric testing. The resulting eight-scale structure reflected a deliberate effort to move beyond a single body-awareness score.
The study also acknowledged weaknesses. Some scales had lower internal consistency than ideal, particularly those with few items. The instrument was therefore a starting architecture, not a finished claim that every dimension had been measured perfectly.
This is important when older studies are compared with newer ones. A validation paper can support use while still identifying places where the instrument needs refinement.
MAIA-2
MAIA-2 was created to address psychometric limitations in the first version. The 2018 study added items, producing a 37-item instrument, and tested the revised structure in a large community sample collected at the Science Museum in London.
The revision improved reliability for previously weak scales while retaining the conceptual eight-factor model.
MAIA-2 should therefore not be treated as a different construct family. It is a refinement of the same multidimensional framework.
Researchers still need to state which version they used. Combining MAIA and MAIA-2 scores without attention to item differences can create artificial comparability.
Cross-cultural validation matters
A questionnaire is made of language.
That fact becomes especially important when the items concern subtle internal experience. Words such as "trust," "listen," "notice" and "worry" carry cultural and linguistic assumptions. Translation can alter intensity, connotation and the implied norm for attending to the body.
Validation therefore cannot stop at literal translation.
Norwegian work published in 2023 found support for the eight-factor structure and generally good internal consistency, while also reporting demographic effects and the need to interpret scores in context. Other validation work has found acceptable structure alongside questions about temporal stability or differences across groups.
The appropriate conclusion is neither "the scale is universal" nor "self-report cannot travel."
It is that measurement invariance and local validation are part of the evidence, not administrative details.
Change scores require care
The MAIA is often used before and after mindfulness, movement or clinical interventions. If scores change, the tempting interpretation is that the person's interoception improved.
That phrase is too broad.
An intervention might increase reported noticing while leaving objective sensory performance unchanged. It might reduce worry about sensations. It might increase confidence in using bodily cues. Each would produce a meaningful change, but they are not the same mechanism.
A recent meta-analysis of mindfulness training illustrates the distinction. Across randomized trials, mindfulness was associated with a small average improvement in self-reported interoception, with MAIA measures frequently used. That result should be described as a change in self-reported interoceptive awareness, not proof of enhanced physiological detection accuracy.
The measure determines the claim.
Should researchers use a total score?
The multidimensional design makes indiscriminate total scores conceptually questionable.
A person could score high on Noticing and low on Trusting. Another could score lower on Noticing but higher on Attention Regulation. A global average can make those profiles look similar.
Some studies report totals for convenience or because a research question asks about an overall tendency. When they do, readers should ask whether the aggregation is psychometrically justified for that sample and whether it obscures a theoretically important subscale pattern.
Subscale interpretation is not automatically superior either. Multiple comparisons can generate false positives, and narrow scales can be less reliable.
The answer is not "always use eight scores."
The answer is to decide in advance what dimension the hypothesis requires.
Clinical and personal use are not the same as diagnosis
MAIA has been used in clinical populations, including chronic pain and mental-health research. That does not make it a diagnostic instrument.
There is no scientifically defensible threshold at which a person becomes "disembodied." Low Trusting does not reveal why the body feels unsafe. High Noticing does not tell us whether attention is adaptive. Scores can reflect current symptoms, treatment context, cultural learning, physical illness, meditation experience and response style.
Questionnaires can organize a conversation.
They cannot replace one.
Nordic relevance without automatic transfer
The Norwegian validation literature is useful because it brings the measure closer to Nordic language and population contexts. It does not validate a Swedish version by proxy.
Norwegian and Swedish are related languages, not interchangeable measurement environments. A Swedish implementation still requires attention to translation procedure, psychometrics and licensing or instrument-use conditions.
This is an example of a broader principle: geographic closeness does not remove the need for validation.
Human-capacity programmes need profile logic
If an institution wants to understand embodied capacity, the MAIA can be useful precisely because it refuses a unitary trait.
A programme might care about whether people can sustain attention to internal changes without immediately worrying, whether they can differentiate bodily and emotional signals, or whether internal attention supports regulation under pressure.
But institutional use raises ethical questions. Self-report about the body can contain sensitive information. Employees may answer strategically if they believe scores affect evaluation. Group averages can be mistaken for individual diagnosis.
A workplace should not turn MAIA into a compliance metric for being "regulated."
The instrument was built for research and reflective assessment, not managerial surveillance.
Response style is part of the data problem
Questionnaires are affected by more than the target construct. Some people use extreme response options readily; others avoid them. Social desirability, familiarity with psychological language and expectations about an intervention can influence answers. Repeated administration can also teach participants the vocabulary of the scale.
That does not make the responses meaningless. It means change scores should be interpreted with comparison groups, appropriate measurement models and attention to whether the scale operates similarly across time. An intervention can change both experience and the way a participant understands an item.
The more subtle the item, the more important this becomes. "Listening to the body" can mean deliberate attention, trust, curiosity, obedience to sensation or simply noticing. Translation and training can shift which of those meanings is salient.
Minimal important change is not self-evident
A statistically significant increase on a MAIA subscale is not automatically a practically important change. Group means can move by a small amount that is detectable in a large sample while producing little difference in everyday functioning. Conversely, a modest score change may matter if it corresponds to a meaningful behavioral shift for a particular population.
Interpreting change therefore needs an external anchor when possible: symptoms, task performance, functioning, behavior or qualitative reports that establish what the numerical difference meant.
Without an anchor, the safest language is descriptive. The score changed by a specified amount. What that change buys in real life remains a separate empirical question.
The instrument is not the intervention
Because MAIA is common in mindfulness and somatic research, it can become conceptually fused with the practices that use it. The scale does not endorse any intervention. It can record change after a programme, no change, or movement in different directions across subscales.
This separation protects both science and practice. A useful intervention should not need its outcome measure to be treated as a theory of how the intervention works.
Norms are descriptive, not ideals
Reference distributions can help researchers understand whether a sample differs from previous populations. They should not be turned into a target profile for a healthy person. The MAIA scales do not define a single optimal configuration, and higher is not uniformly better across every dimension and circumstance.
In particular, Trusting can be complicated by illness, pain or histories in which bodily cues have been frightening. Not-Distracting can have different implications when a person needs temporary attentional relief from severe symptoms. A score becomes meaningful only inside a research or clinical question that respects context.
Institute synthesis: interoceptive profile before interoceptive rank
The Institute uses the principle profile before rank when working with multidimensional capacity measures.
This is Institute synthesis, not MAIA terminology.
The principle means that different configurations should be understood before people are ordered along a single better-worse axis. In interoception, more noticing is not automatically better. Less distraction is not always desirable when a sensation is overwhelming. Trust can be adaptive when calibrated and dangerous when it overrides contradictory medical evidence.
The goal is not maximum inward attention.
It is a workable relationship among noticing, interpretation, regulation and external reality.
What the evidence does not show
MAIA and MAIA-2 do not measure objective interoceptive accuracy. They do not establish that a person with a higher score is healthier, wiser or more embodied. They do not diagnose psychiatric or medical conditions. Validation in one language or population does not establish invariance everywhere. A change after training does not identify the mechanism responsible for that change.
The eight-factor structure is well studied, but no questionnaire can eliminate self-report biases, cultural interpretation or context effects.
The evidence supports the MAIA as a sophisticated self-report instrument.
It does not turn subjective experience into a direct sensor readout.
Open questions
How stable are MAIA dimensions across long periods rather than intervention windows? Which subscales show genuine cross-cultural equivalence? How do self-report profiles relate to respiratory, cardiac and gastrointestinal performance tasks? Can intensive longitudinal versions distinguish state change from trait description? Which dimensions predict adaptive use of bodily information rather than simply greater inward attention?
These questions are not defects in the instrument.
They are what becomes visible once body awareness stops being treated as one thing.
Research references
- Mehling WE et al. The Multidimensional Assessment of Interoceptive Awareness (MAIA). PLoS One. 2012;7(11):e48230. DOI: 10.1371/journal.pone.0048230. PMID: 23133619.
- Mehling WE et al. The Multidimensional Assessment of Interoceptive Awareness, Version 2 (MAIA-2). PLoS One. 2018;13(12):e0208034. DOI: 10.1371/journal.pone.0208034.
- Mehling WE et al. Body awareness: Construct and self-report measures. PLoS One. 2009;4(5):e5614. DOI: 10.1371/journal.pone.0005614.
- Norwegian validation study of MAIA/MAIA-2. 2023. PMID: 37430262.
- Treves IN et al. A meta-analysis of the effects of mindfulness meditation training on self-reported interoception. Scientific Reports. 2025;15:38889. DOI: 10.1038/s41598-025-22661-4.
- Interoceptive awareness in a Southeastern US college sample: validation of the Multidimensional Assessment of Interoceptive Awareness, version 2. BMC Research Notes. 2024. https://doi.org/10.1186/s13104-024-06894-6