Does noticing your thoughts more clearly mean noticing more of what happens around you? During a breathing practice, attention may wander to a memory and return to a bodily sensation. That shift helps frame the question of which kind of attention is being trained and how to test its effects.
Noticing a thought and detecting a signal are different tasks
While attending to the breath, we may notice a thought arising and return our focus. Studying that process requires clarifying what 'attention' means. Noticing a thought, holding a phone number in mind and detecting a faint light on a screen are different tasks. Yuval Hadash, Amit Bernstein and colleagues start from a puzzle: attention is central to theories of mindfulness, yet experiments testing whether training improves it have produced mixed findings. Their Mindfulness Internal Attention, or MIA, framework tries to make the question precise enough to test.
Mindfulness training can take different forms and is studied for different outcomes, including well-being. This review does not ask whether every program helps everyone. It asks which attentional processes a practice may exercise and which tests are capable of detecting that change. If a study uses a task far removed from what was practiced, an unchanged score may say little about the proposed mechanism. Conversely, a better score on one task cannot justify a claim that all attention improved.
The authors' first move is to distinguish information coming through the senses from information generated or held within the mind. Both can require effort and control, and the border is not absolute. Still, the distinction is useful. A meditation practice often asks a person to notice sensations, thoughts and emotions as they arise and to choose where attention returns. The key activity may be the selection and monitoring of internal experience, not the earliest processing of external signals.
Looking inward takes several steps
Suppose a person realizes, halfway through an exercise, that they have been planning tomorrow's work. First they must notice that attention moved; then identify what captured it; then decide whether to return to the breath or continue thinking. The content is internal, and so is much of the control required. The review calls this family of operations internal attention. It includes attention to memories, mental images, bodily sensations and the flow of thought. It is not identical to being calm or having an empty mind. The practice may involve recognizing a distraction without automatically following it.
The authors review studies of mind wandering and meta-awareness. Mind wandering is the shift of thought away from a current task. Meta-awareness means noticing that this has happened. A person can spend minutes planning dinner without realizing it; another may notice the shift almost at once. Mindfulness training could affect the second process even if thoughts continue to arise. Counting every thought as a failure would miss the skill being tested. A better measure asks how often or how quickly a person detects the change and what they do next.
Another line of work concerns interoception: sensing conditions inside the body, such as heartbeat or breathing. Paying attention to the body does not guarantee accurate perception of every signal. People can report greater awareness while a performance test changes little, partly because the tests measure different things. The framework asks researchers to specify whether they mean noticing, sustaining attention, judging accurately or regulating a response. These distinctions matter before a result can be called evidence for or against the theory.
Some outward tasks share the same mental tools
Internal and external attention are not separate sealed systems. Reading a page requires holding earlier words in mind while new ones arrive. Listening to an instruction may require inhibiting an irrelevant response and updating what the task demands. The review argues that executive functions and working memory can support both internal attention and later stages of attention to external material. If mindfulness training exercises these shared processes, some outward tasks may improve even though the practice mainly focuses inward.
Working memory is the limited ability to hold and use information for a short period. Executive control helps set goals, switch between them and resist competing responses. A person returning to the breath repeatedly may practice noticing a shift and reestablishing the chosen goal. That operation resembles some laboratory tasks more than others. The framework predicts transfer where processes overlap, not wherever a task happens to be labeled 'attention.' Similar names can hide very different demands.
The evidence reviewed is mixed rather than a simple endorsement. Some studies find changes in executive functions or working memory; others do not. Training length, comparison groups, baseline ability and the precise tasks used can matter. Improvement on a practiced test may reflect familiarity with that test rather than a broad new capacity. A strong study should compare mindfulness with a credible alternative activity and examine whether any change extends to untrained tasks that share the proposed mechanism.
Why spotting an external signal may stay the same
Consider a task in which a small mark flashes briefly at the edge of a screen. Detecting its first appearance depends on early stages of visual attention. The authors propose that those stages do not draw on the same resources as attending to thoughts or steering working memory. Mindfulness training might therefore leave performance on such a task unchanged. That would not automatically mean the training 'failed.' It would be consistent with a theory that says the training targets something else.
The distinction is subtle because an external task may include both early detection and later decisions. A participant first senses the mark, then identifies it, remembers a rule and presses a key. If a score improves, which step changed? A task described broadly as visual attention may in fact rely heavily on executive control after the signal appears. The review urges researchers to analyze the task instead of sorting tests by their everyday labels. This is how a framework can make mixed findings informative.
It is also a testable risk for the theory. If careful experiments repeatedly show reliable change in early external processing, the framework would need to explain why. If they show no change there but improvements in internal monitoring and shared executive operations, the proposed distinction gains support. A good theory is not protected by explaining every possible result after it arrives. It states in advance which results would be surprising and which measures could distinguish competing accounts.
The hardest part is measuring an inner event
A flashing light is easy to time; the instant someone notices a thought is harder to observe from outside. Researchers can interrupt participants and ask what they were thinking, invite real-time reports or examine performance changes linked to mind wandering. Each method adds a problem. An interruption alters the experience; a report depends on memory and words; a physiological signal may have more than one interpretation. Internal attention needs better measures, but 'better' means more than adding a sensor. The measure must match the process claimed.
The authors suggest combining methods and designing tests that distinguish their predictions from alternatives. A study could measure how quickly people notice their mind wandering, how they return to a chosen focus and whether a similar executive operation changes on an external task. It should specify the training, the comparison condition and the timing of outcomes. A larger sample does not solve a poorly defined construct. If every questionnaire and button press is treated as the same form of attention, more data simply enlarge the confusion.
Evidence about health and well-being raises another step. Even if training changes internal attention, that change might not be what causes an improvement in well-being. Sleep, expectations, social support or the time reserved for practice could contribute. To argue for a mechanism, studies must show that the proposed attentional change precedes and helps explain the outcome, while testing other plausible routes. The review treats this as an open agenda, not a settled chain from meditation to health.
A more honest promise
The MIA framework offers a useful change of wording. Instead of claiming that mindfulness strengthens a single, general 'attention muscle,' ask whether a particular practice helps someone notice and guide inner experience, and whether any benefit transfers to an external task with overlapping demands. This narrower question is more interesting because it can be wrong in identifiable ways. It also respects the difference between feeling more aware and performing better on a laboratory measure.
For someone deciding whether to practice, the review does not provide a personal prescription. People differ in goals, experience and response. A person might value taking a moment to recognize an emotion even if a screen-based detection task never changes. Another might need help with a condition that calls for professional care rather than an app promising better focus. The scientific claims should stay tied to the measured practice and outcome; the personal value of a practice is a separate question.
Return to the person who noticed a thought and returned to breathing. The meaningful change may lie in that moment of noticing and choosing, not in seeing every light more quickly. Hadash and colleagues use this distinction to turn apparently inconsistent findings into sharper research questions. What kind of attention was trained? What process did the test require? Where did any improvement travel? Those questions make the evidence easier to interpret without making the practice sound magical.
What to keep
- Attention is not one process.
- Mindfulness may target internal attention.
- Mixed findings can refine theory.
The nuance
What this story does not let us claim
The framework organizes initial evidence and still requires direct, discriminating tests.
Go deeper: concepts and optional reading practice
Reversible cards
Test the concept
Topic assessment
Check whether you can explain the idea, not merely recognize it.
Every answer is grounded in the explanation above. Some questions introduce new cases to test whether you can apply and analyze the ideas, not merely recall them.
Source & scope
The Mindfulness Internal Attention (MIA) Framework: Uncovering the Attentional Mechanisms of Mindfulness Training. Yuval Hadash, Omer Dar, Iftach Amir, Todd S. Braver & Amit Bernstein. Annual Review of Psychology, vol. 77 (2026).
This editorial guide is an independent explanation. Consult the original publication for the studies, references and full context.
Open the original review ↗