The plain phenomenology of attention

Part I: The field before the spotlight

You are standing on a railway platform, reading a difficult message while waiting for a change of platform announcement. The words are central. The station hum, the pressure of your bag strap and the movement at the edge of vision are still present. You have already decided to listen for your train number. Then a metal bottle hits the floor. Your reading stops, your head turns and the sharp sound becomes central. A second later you return to the message.

Nothing mysterious happened, yet five familiar words now compete to describe it. You were conscious of the platform. You paid attention to the message. You intended to catch the announcement. You concentrated until the crash. You remained aware of more than the line you were reading. In ordinary speech these claims blur because the episode was successful: what you intended, prioritised and stabilised mostly coincided.

The hidden assumption is that focus is one mental quantity. Turn it up and consciousness becomes clear, attention becomes strong, intention becomes firm, concentration deepens and awareness increases. That picture feels plausible because the variables often rise together during deliberate work. It fails as soon as one of them moves alone.

Mechanism

The useful unit is not a word but an intervention. Keep the scene broadly constant, change one causal property, then ask what changed in the lived organisation of the episode. A sudden sound changes priority without waiting for intention. A decision can persist while priority is captured elsewhere. A narrow focus can become more stable while the wider field becomes less disclosed.

A movable centre within a wider field. The bottle need not create consciousness from nothing. It can reorganise what is central within an already present scene. The decision-relevant question is therefore not simply “Was I aware?”, but “What occupied the centre, what remained peripheral and what controlled the movement?” Illustrative conceptual model, informed by James’s selective account of attention and Watzl’s priority-structure account.

A vocabulary for observation, not a final ontology

William James described attention through selection among simultaneous possibilities, while later work distinguishes alerting, orienting and executive control. Contemporary philosophy also treats attention as an organisation of relative priority rather than a single beam or brain location.^{1, 2, 3} These sources do not settle one definition. They do support a practical lesson: a report that someone was “attentive” leaves many causal questions unanswered.

This paper uses five terms in deliberately restricted roles. It does not claim that every scientific laboratory, philosophical school or contemplative tradition should adopt them. “Awareness” in particular is used both as a synonym for consciousness and as a more reflective or encompassing relation. Here it names field disclosure: how much of the ongoing scene, including its centre-periphery organisation, is available within the episode. When the fact of attending is itself noticed, the narrower term meta-awareness is used.

Consciousness: Experiential presence: that an episode or content is present for the subject at all. This does not explain why experience exists.
Attention: Relative priority: which content, location, feature or task is weighted for processing, response or experiential centrality.
Intention: Directional set: a commitment or control state aimed at selecting, maintaining or changing what happens next.
Concentration: Temporal stability: the persistence of a selected priority despite interference, fatigue or competing content.
Awareness: Field disclosure: the breadth and immediacy with which centre, periphery and ongoing change are noticed within experience.

Why ordinary introspection compresses the variables

Most successful action aligns the coordinates. You decide to read, the words gain priority, the ranking remains stable, the sentence is clearly present and little else demands notice. When asked afterwards whether you were attentive, one “yes” efficiently summarises the whole success. Language evolves for coordination, not for preserving experimental variables. The compression becomes visible only when action fails or when a practice deliberately changes one relation.

The variables also influence one another. Priority can increase perceptual clarity, memory and response readiness. A strong intention can repeatedly restore a target. Stable concentration can make the target seem more continuous. Broad field disclosure can reveal more distractors and therefore create the impression of weaker concentration. Correlation is therefore expected. The error is to turn expected coupling into identity.

Report timing adds another compression. By the time you say “I noticed the bottle”, the episode already contains capture, orienting, recognition and perhaps a judgement that you were distracted. A single retrospective noun replaces a sequence. The first discipline is temporal: recover what happened before deciding what it was. This is why the later protocol asks for transitions before ratings and confidence before interpretation.

Thought experiment 1 · The waiting ear

Keep reading while silently deciding to catch the words “Platform four”. For ten seconds no announcement occurs. Your intention is active, yet auditory attention may remain weak. Now imagine the exact phrase sounding quietly behind you. Does it enter the centre before you deliberately turn towards it? Finally, replace the phrase with an equally loud irrelevant sentence. The contrast changes behavioural relevance while holding volume roughly constant. If intention, attention and consciousness were one quantity, these three moments should not separate so cleanly.

Evidence status · Practitioner synthesis. The five-coordinate vocabulary is an operational proposal. Published work supports dissociations among top-down attention, stimulus-driven capture, conscious perception, mind-wandering intentionality and meta-awareness. No cited study validates this exact five-part scheme as a natural kind.

The argument in one pass

Chart · flow

flowchart TD
  A["Part I, The field before the<br/>spotlight"] --> B["Part II, Pull one lever at a time"]
  B --> C["Part III, What the<br/>dissociations reveal"]
  C --> D["Part IV, Turn observation into<br/>operating evidence"]
  D -. "re-test" .-> A
  classDef first fill:#0a1424,stroke:#5B5CF6,stroke-width:2px,color:#e6f3ff
  classDef step fill:#0a1424,stroke:#3f7da0,stroke-width:1px,color:#e6f3ff
  class A first
  class B,C,D step

A structural route through The plain phenomenology of attention: from Part I, The field before the spotlight to Intervene on the coordinate that…. Reading use: follow the relation, then return to the surrounding argument for evidence and limits.

Part II: Pull one lever at a time

Definitions become useful only when they predict a difference. The simplest test is to hold four properties roughly steady and perturb the fifth. Lived experience is not a clean laboratory, but ordinary episodes contain natural interventions: a flash captures priority, a goal persists through distraction, noise tests stability and a deliberate widening changes field disclosure.

Five levers, five primary signatures. A clean intervention should move its target coordinate more than the others. Real episodes contain cross-effects, shown by dotted paths. The framework succeeds only if those cross-effects can be recorded instead of being renamed as a single improvement in “focus”. Illustrative causal framework. The primary and cross-link assignments are hypotheses, not established neural pathways.

Attention is priority, not necessarily choice

A sudden sound can become central before a deliberate decision. Goal-directed and stimulus-driven control recruit partly distinct attention systems, although their interaction is extensive.⁴ Phenomenologically, the useful signature is a changed ranking: the crash now matters more than the sentence. The movement can be spatial, as when the eyes turn, or internal, as when a remembered worry becomes central without any external change.

Priority is relational. Saying that the crash received “more attention” means it outranked other available contents for some function or within experience. It does not mean a measurable fluid flowed into it. This avoids a common mistake: treating attention as a substance that a person possesses in a fixed amount.

Intention is a directional set that can fail

An intention points towards a future selection or action. You can intend to read while attention remains captured by a notification. You can also intentionally let the mind wander during an easy task. Research on mind-wandering has found that intentional and unintentional episodes behave differently across task demands, so “off task” does not by itself reveal whether control was lost.^{5, 6}

An intention need not be a sentence in the head. It may be felt as readiness, restraint, anticipation or a recurring pull. Its presence is inferred from the episode’s temporal organisation and from counterfactuals: what would you return to after interruption, and what counts as a deviation?

Concentration is stability under interference

Concentration is often treated as intense attention. The better test is persistence. Two moments can give the text equal priority, yet one collapses after a footstep while the other survives a loud conversation. Concentration therefore concerns the trajectory, not a single frame. It includes maintenance, detection of drift and recovery, but those phases should not be conflated.

Focused-attention meditation provides a tractable example. Hasenkamp and colleagues modelled a cycle containing mind-wandering, awareness of mind-wandering, shifting and sustained attention.⁷ The cycle is useful because it separates the lapse from noticing the lapse and from restoring the target. A low rate of distraction and a short recovery latency are different achievements.

Awareness is how much of the organisation is disclosed

Imagine reading the same sentence in two ways. In the first, the words fill the experiential centre and the body disappears from notice. In the second, the words remain central while the page, breath, posture and station noise are also clearly present. Priority can remain stable while the field becomes wider. Lutz and colleagues use a related contrast between focused-attention and open-monitoring families of practice, while stressing that meditation is a family of methods rather than one procedure.^{8, 9}

This article’s “awareness” coordinate includes breadth, sensitivity to change and the possibility of noticing the current attentional organisation. It does not mean a metaphysical witness, pure consciousness or an entity separate from experience. Those stronger claims require different arguments.

Consciousness is presence, and first-person absence is a limit

Consciousness here answers the most basic question: is there something experientially present? Yet a first-person log cannot sample complete absence, because there is no present observer report from within that absence. It can compare the clarity, determinacy and reportability of contents within experience. It can also be paired with behavioural or physiological paradigms that manipulate visibility.

Psychophysical work argues that top-down attention and conscious perception can dissociate, including cases of attention to stimuli not consciously seen and conscious scene gist with little directed attention.^{10, 11} The interpretation remains contested because measures of awareness and attention can contaminate each other. The modest conclusion is enough here: conscious presence and prioritisation should not be defined into identity.

The centre is functional, not necessarily spatial

The railway case uses sights and sounds, but the same grammar applies internally. A remembered conversation can become central while the visual scene remains stable. A faint pain can sit at the periphery, then gain priority when you wonder whether it is worsening. An intention can target an image, a calculation or the absence of verbal thought. Concentration can stabilise any of these without fixing the eyes or body.

This matters because “internal attention” is sometimes treated as a different faculty. The object changes, but the questions remain. What is present? What outranks what? Which directional set is active? How stable is the ranking? How much of the surrounding field and the act of attending is disclosed? The answers may recruit different mechanisms, yet the observational relations are comparable.

Centre and periphery are therefore not locations inside a mental screen. They mark organisation. A sound directly in front can be peripheral while an imagined sentence is central. Likewise, broad awareness does not mean equal processing of everything. It means that more of the field remains disclosed without every item receiving the same priority.

Worked example 1 · A five-coordinate episode

Use an ordinal scale from 0 to 4, where numbers mark report categories rather than physical quantities. While reading, an episode might be recorded as (4, 4, 4, 3, 1): strong experiential presence, high text priority, strong intention to read, fairly stable concentration and narrow field disclosure. After the bottle crash it might become (4, 1, 4, 0, 3). Presence and intention remain high, while text priority and stability fall and the wider field opens.

E_read = (p, q, d, s, b) = (4, 4, 4, 3, 1)
E_crash = (4, 1, 4, 0, 3)
ΔE = (0, −3, 0, −3, +2)

Here p is phenomenal presence, q relative priority, d directional set, s stability and b breadth. The subtraction only highlights which coordinates changed. It does not justify averaging them or producing an “attention score”. The assumption is that the categories are ordered enough for within-person comparison. Remove that assumption and the safer representation is a sequence of labelled states rather than numbers.

The episode is a trajectory, not a snapshot. The reading intention can remain constant while priority and stability collapse. Field breadth can widen during capture, and meta-awareness can arrive after the shift. Training “attention” without identifying the targeted trajectory risks optimising the wrong property. Illustrative trajectory. Shapes are synthetic and encode the worked scenario, not measured data.

Worked example 2 · Three tones and a kettle

Play a repeating sequence of three soft tones while a kettle warms in the next room. At first, count the middle tone. The sequence is conscious, the middle tone has priority, the counting intention is active and concentration is tested by each pause. Next, stop counting but keep the sequence present as a whole. The same sounds remain conscious while priority becomes less sharply ranked. Finally, continue hearing the sequence while including the kettle, posture and room space. Field disclosure widens without requiring the tones to disappear.

Now let the kettle click off unexpectedly. The click may seize priority, although the intention to hear the tones remains. If you notice the capture only after missing the next middle tone, meta-awareness lagged behind the attentional shift. This tiny example can be followed without equipment, yet it produces five separately reportable questions. Was sound present? Which sound was central? What were you trying to do? How long did the ranking hold? How much of the surrounding field was disclosed?

Thought experiment 2 · The silent arrow

Fix your gaze on the centre of a screen. An arrow instructs you to attend left, establishing a directional set. Before the target appears, a bright flash occurs on the right. You can still intend left while priority is briefly captured right. Now reduce the flash until it no longer captures you. The intention is unchanged, but attentional movement differs. The experiment separates a control state from its success.

Thought experiment 3 · The same breath, two apertures

Keep the breath at the centre for thirty seconds. In the first half, exclude other sensations as strongly as possible. In the second, keep the breath central but allow sounds, posture and thought fragments to remain plainly present. If the centre stays equally stable, concentration is roughly controlled while field disclosure changes. Narrowness and stability are therefore not the same achievement.

Part III: What the dissociations reveal

A distinction earns its place when it changes prediction. The five-coordinate scheme predicts not one hierarchy but a family of dissociations. Some are familiar in laboratories, others in ordinary work and contemplative practice. None requires assuming that introspection is infallible.

A dissociation matrix, not a dictionary. Each cell names a case that should be observable or experimentally inducible. A distinction is strengthened when multiple methods reproduce the same selective change and weakened when every manipulation moves all coordinates together. Illustrative evidence instrument. Examples combine published paradigms with practitioner synthesis.

Consciousness and attention: related, not safely identical

Attention often amplifies, stabilises or selects conscious content. That close coupling explains the everyday conflation. It does not prove identity. Koch and Tsuchiya’s influential review argued for attention without consciousness and consciousness with little top-down attention; van Boxtel and colleagues developed related necessity and sufficiency arguments.^{10, 11} Other accounts place attention closer to conscious access, and debates about “overflow” show that the measurement bridge remains unsettled.^{12, 13}

The plain phenomenological contribution is smaller. Within a conscious scene, centrality changes without requiring the whole scene to blink out. Conversely, something can be experienced vaguely or peripherally without becoming the target of deliberate selection. Attention usually edits the organisation of consciousness; it is not thereby identical to experiential presence.

Intention and attention: command is not execution

The railway example shows failed control. A more revealing case reverses it: attention can be organised without a conscious intention to organise it. A salient face, your own name or a sudden movement may take priority. Corbetta and Shulman’s distinction between goal-directed and stimulus-driven systems provides a neural and behavioural framework for this contrast, without implying a simple one-network-per-experience mapping.⁴

This matters for self-assessment. “I could not pay attention” may mean the target never acquired priority, the intention was weak, interference was high, monitoring noticed lapses more often or recovery was slow. Treating all five as willpower produces poor diagnosis and punitive training.

Concentration and awareness: a trade-off, not a law

Focused-attention and open-monitoring instructions make scope visible. The former typically selects an object and trains return after distraction. The latter reduces object selection and monitors moment-to-moment content more broadly.⁸ These are families of practice, and individual techniques mix them. A stable focus can coexist with broad field awareness, but early training often reveals a trade-off because widening introduces more potential distractors.

Recent review work emphasises that more than two decades of mindfulness research has produced mixed findings on attention, and proposes that internal attention may be a more specific target than early external selection.¹⁴ That boundary is important. A precise phenomenological distinction can improve experimental design without implying that any meditation label reliably changes a broad cognitive faculty.

One scene, three legitimate design objectives

Consider a proof-reader, a night driver and a therapist listening to a difficult disclosure. Each needs attention, yet the useful profile differs. The proof-reader benefits from narrow priority, high stability and a strong error-detection intention. Excessive field breadth may slow the task. The driver also needs stability, but a narrow field is dangerous because hazards appear outside the current centre. The therapist must keep the speaker central while remaining aware of tone, bodily response, silence and the impulse to formulate the next question.

A single “focus” score could reward the proof-reader’s profile and penalise the driver’s safer one. It could also mistake the therapist’s broad sensitivity for distraction. The five coordinates expose the design choice. Selection, stability and breadth can be optimised differently because the environment imposes different costs of omission and capture.

This comparison also reveals a boundary. No profile is intrinsically superior. Narrowing can be skilful or reckless. Broadening can be sensitive or diffuse. Stability can support expertise or perseveration. The target profile must be justified by the task’s failure costs, not by a generic ideal of maximal concentration.

Worked example 3 · A configured dissociation study

Recruit participants for four twelve-second blocks. The visual target remains a stream of letters. In block A they read freely. In B they intend to detect the letter sequence RAIL. In C an irrelevant tone is made highly salient. In D they keep RAIL central while also monitoring body pressure and room sound. Collect immediate coordinate ratings, target accuracy and response time. The key prediction is selective: the tone should reduce target priority and stability more than intention; the broad-monitoring instruction should increase field disclosure without necessarily reducing target accuracy.

A positive result would show that the instructions and perturbations produce separable report patterns with behavioural consequences. It would not prove that each coordinate has a unique neural mechanism. A negative result could mean the concepts collapse, the intervention was weak, the rating vocabulary was confusing or report demand reshaped the episode. The design therefore needs manipulation checks and repeated within-person trials.

Rival explanations and break conditions

Suppose the configured study produces the predicted profiles. A common arousal change could still explain several coordinates at once. The salient tone might increase bodily activation, speed responses and widen reported awareness without selectively changing priority. Eye movements could create the apparent visual shift. Participants could infer the hypothesis from the rating labels and comply. Task difficulty could reduce every rating through frustration rather than a specific attentional mechanism.

Each rival suggests a control. Match arousal with a non-task-relevant sound. Track gaze or require fixation. Separate the person giving instructions from the hypothesis. Counterbalance the coordinate vocabulary, and include free description before ratings. Match difficulty across blocks and measure subjective effort separately. Add a no-report block to test whether repeated introspection changes performance.

The framework should be abandoned or revised if no intervention shows coordinate selectivity across these controls, if the coordinate structure disappears outside the training vocabulary, or if independent measures never track the reported contrasts. It should not be rescued by redefining every null result as “subtle awareness”. A useful phenomenology needs break conditions as much as evocative descriptions.

Correct output is not a mechanism report. Matching behaviour can arise from deliberate focus, automatic response or chance. Remove the target instruction, add a secondary task or vary confidence to separate the paths. This logic applies equally to human attention studies and claims about attention in artificial systems. Illustrative counterfactual generator model.

Thought experiment 4 · The perfect score

Two participants identify every target. One reports a stable central stream and notices each distractor without losing the task. The other thinks about dinner and responds automatically whenever a familiar pattern appears. Their accuracy is identical. Now change the font, delay the response or insert a novel target. Performance diverges. The same output did not imply the same attention, intention or awareness.

Illustrative profiles that cross. Focused reading and open monitoring can both be conscious and intentional while differing in priority, stability and breadth. Capture can leave the original intention intact while target priority collapses. Crossing profiles are the signature that a single “focus” scale would destroy information. Illustrative chart using synthetic ordinal values.

Worked example 4 · The vocabulary trap

A participant completes a target task accurately and later says, “My concentration was excellent.” A rushed analysis treats this as evidence of stable attention. A sequence interview reveals something else. The first half was deliberate and narrow. During the second half, the response became automatic, thoughts moved to an upcoming meeting and only the target beep occasionally returned to the centre. Accuracy stayed high because the mapping was overlearned.

The global label concealed a mechanism change. “Excellent” referred to outcome, not continuous stability. The repair is to ask for transitions before adjectives: when did the target cease to be central, what remained of the intention, what triggered the response and when was the drift noticed? If the participant cannot recover those distinctions, the correct result is not a forced coordinate profile. It is a low-confidence report with missingness preserved. Uncertainty is more informative than a polished retrospective fiction.

The strongest boundary: reports are interventions too

Asking “What were you aware of?” can broaden the very field being measured. A retrospective answer can replace a brief event with a theory about the event. Familiar vocabulary also compresses distinctions: a participant may call vividness, effort, accuracy or obedience “concentration”. Micro-phenomenological methods respond by returning to one concrete episode, evoking its sequence and asking how rather than why.^{15, 16}

These methods are not immune to reconstruction. A study of micro-phenomenological self-inquiry with initially untrained participants did not replicate an earlier proposed reliability effect, suggesting that training, task design and interview structure matter.¹⁷ The right conclusion is not that first-person evidence is useless. It is that its provenance, timing, prompts and inferential limits must be recorded like any other measurement.

Boundary condition

A first-person report can describe a conscious episode; it cannot by itself establish the mechanism that caused it, the absence of unreported content or the metaphysical nature of awareness. Behaviour, task manipulation, physiology and trained elicitation can constrain one another, but they are not interchangeable ledgers.

How the claim changes under pressure

Chart · state

stateDiagram-v2
  [*] --> Claim
  state "A vocabulary for observation,<br/>not a final ontology" as Claim
  state "Why ordinary introspection<br/>compresses the variables" as Test
  state "Attention is priority, not<br/>necessarily choice" as Revision
  Claim --> Test: expose mechanism
  Test --> Decision: evidence survives
  Test --> Revision: rival survives
  Revision --> Claim: narrow or rebuild
  Decision --> [*]

The revision loop begins with A vocabulary for observation, not a… and keeps counterexamples inside the reasoning path. Reading use: follow the relation, then return to the surrounding argument for evidence and limits.

Part IV: Turn observation into operating evidence

The practical contribution is a small protocol for recording brief episodes without collapsing them into a global judgement. It is designed for personal inquiry, pilot studies and experimental design. It is not a diagnostic tool and not a consciousness detector.

The report-production cutaway. Immediate logging, concrete evocation and descriptive prompts reduce several failure paths. They cannot erase memory construction or demand effects. A trustworthy report therefore preserves how it was elicited and where uncertainty entered. Practitioner synthesis informed by micro-phenomenological interview and analysis methods.

The protocol

Choose a short, repeatable episode. Ten to twenty seconds is enough. Define one target and one intervention before beginning. Examples include reading with or without a notification, listening for a soft tone, sustaining the breath while widening the field, or performing a simple task before and after a distractor.

Immediately afterwards, reconstruct the sequence before rating it. Ask: What was central first? What changed? Was the movement invited, captured or unnoticed? Did the original intention persist? What marked the lapse? How did return occur? Only then assign the five coordinates. Add confidence and a note about the prompt, because uncertainty is data.

Repeat the same intervention several times. Compare within the same person before comparing people. The terms may be used differently across participants, while a person’s own contrasts can still be stable. Pair reports with a task measure when the claim concerns performance, and with an external perturbation when it concerns causation.

Analyse transitions, not averages

After several entries, resist the urge to average the five numbers into one score. Instead, compare transition patterns. How long did the target remain central before capture? Did the intention persist during the lapse? How quickly was the lapse noticed? Was return abrupt or gradual? Did widening the field reduce stability, or did that trade-off disappear after repeated trials?

A simple session analysis can use four quantities without pretending to measure consciousness. Dwell is the reported time a target remains central. Capture latency is the interval between a perturbation and a priority shift. Noticing latency is the interval between the shift and meta-awareness. Recovery latency is the interval from noticing to restored priority. These quantities can be estimated from button presses, event markers or immediate sequence reports, with uncertainty attached.

Consider two sessions with identical target accuracy. In the first, capture happens often but is noticed and repaired quickly. In the second, capture is rare, yet each lapse persists. The same average accuracy conceals different control problems. The first may benefit from reducing salience or interruption frequency. The second may need better monitoring or a clearer return cue. Coordinate profiles explain the state; transition measures explain its dynamics.

Cross-person comparison should come later. Begin by checking whether the same participant uses the scale consistently under repeated, counterbalanced interventions. Then compare qualitative sequence structures before comparing numbers. If one person uses “4” for effortless clarity and another for forceful effort, numerical alignment is spurious. A shared coordinate name does not guarantee a shared threshold.

Decision rule: claim a useful distinction only when an intervention changes its predicted coordinate more than plausible rivals, the effect repeats, and the report procedure cannot trivially manufacture the difference.

Executable lab: the five-coordinate episode log

Record one moment immediately after it occurs. The tool keeps data only in this page session and can export a JSON file. Use repeated entries to inspect selective changes, not to calculate a single “quality of attention” score.

Episode label Intervention

Presence 3

Priority 3

Directional set 3

Stability 3

Field disclosure 2

Sequence note Confidence in this report

Current profile

Presence

Priority

Direction

Stability

Disclosure

Moment	P	Q	D	S	B
No moments recorded yet.

The smallest useful experimental loop. Predeclaring the intervention prevents the explanation from being chosen after the report. Repetition and a rival intervention test selectivity. The red centre is an inference boundary: the method maps reported organisation, not the ultimate nature of consciousness. Original experimental protocol.

What the artefact tests

The log tests the selective-lever hypothesis: changes in visibility, salience, goal cue, interference or monitoring aperture should produce different coordinate signatures. It encodes three assumptions. Participants can make useful within-person ordinal comparisons. Immediate sequence reconstruction contains more information than a single retrospective label. Repeated interventions can reveal patterns despite noisy reports.

A positive result permits a narrower claim: this person, under this task and report method, distinguishes at least some coordinates reliably enough to predict an intervention or behavioural change. A negative result means the current protocol did not separate them. It does not show that the underlying distinctions are impossible. The intervention, language or reporting delay may have failed.

The artefact cannot establish unconscious processing, phenomenal absence, neural mechanisms, spiritual attainment or clinical status. It should not be used to rank people. The ordinal values are handles for comparison, not a measurement of consciousness.

Implementation schema and invariants

The exported JSON follows this conceptual TypeScript schema. The central invariant is that no aggregate score is stored. Each coordinate, intervention and uncertainty field remains separately inspectable.

TypeScript · episode schema

type Ordinal0to4 = 0 | 1 | 2 | 3 | 4;
type Intervention =
  | "none"
  | "salience"
  | "goal"
  | "interference"
  | "aperture"
  | "visibility";

interface AttentionEpisode {
  id: string;
  label: string;
  intervention: Intervention;
  coordinates: {
    presence: Ordinal0to4;
    priority: Ordinal0to4;
    directionalSet: Ordinal0to4;
    stability: Ordinal0to4;
    fieldDisclosure: Ordinal0to4;
  };
  sequenceNote: string;
  confidence: "low" | "medium" | "high";
  capturedAt: string;
}

function invariant(e: AttentionEpisode): boolean {
  return e.label.trim().length > 0
    && Object.values(e.coordinates)
      .every(v => Number.isInteger(v) && v >= 0 && v <= 4);
}

// Prohibited: sum(coordinates) as a consciousness or attention score.

Open research hypothesis · Selective levers

The proposal is strengthened if predeclared interventions yield coordinate-specific, repeatable changes that predict independent task measures. It is weakened if all coordinates move as one factor after vocabulary, confidence and demand effects are controlled. A rival explanation is that the rating interface teaches participants to manufacture distinctions. The smallest useful test therefore compares trained and neutral vocabulary, includes a no-rating control block and checks whether behavioural effects survive.

The evidence-to-decision surface. A report alone supports description. An intervention alone supports a task effect. A prediction alone may reflect a correlated proxy. When all three converge, a bounded causal decision becomes reasonable. Even then, the metaphysical conclusion remains outside the permitted region. Original decision instrument.

The evidence conversation

Chart · sequence

sequenceDiagram
  participant Q as Question
  participant M as Mechanism
  participant E as Evidence
  participant D as Decision
  Q->>M: Part I, The field before the spotlight
  M->>E: A vocabulary for observation, not a final ontology
  E-->>M: compare rival explanations
  E->>D: Intervene on the coordinate that actually changed
  D-->>Q: revise scope or act

Question, mechanism, evidence and decision exchange the burdens developed in The plain phenomenology of attention. Reading use: follow the relation, then return to the surrounding argument for evidence and limits.

Intervene on the coordinate that actually changed

Stop asking whether a person, practice or system “has more attention” as though one quantity had risen. Ask which relation changed. Did content become experientially present? Did a target gain priority? Did a directional set form? Did priority remain stable through interference? Did the wider field and the fact of attending become disclosed?

For practice, choose the property you need. A proof-reader may need stable narrow priority. A driver needs stable priority with a wide safety field. A researcher studying conscious perception must separate visibility from selection and report. A designer evaluating an artificial attention mechanism should not infer consciousness from weighted routing. The same word does not licence the same conclusion across these cases.

The practical decision is to replace global praise or failure with a coordinate-specific intervention. When reading repeatedly collapses, strengthening intention may be pointless if salience and interference dominate. When concentration becomes rigid, widening field disclosure may matter more than adding effort. When awareness reports improve, ask whether sensitivity changed or merely the reporting vocabulary.

The plain phenomenology of attention is therefore not a catalogue of inner objects. It is a discipline of controlled contrast. Change one property, watch the field reorganise, preserve the sequence and stop the inference where the evidence stops. That method does not solve consciousness. It does something more immediately useful: it prevents five different problems from receiving the same name.

Glossary

Experiential presence: The occurrence of a content or episode as part of what it is like for the subject.
Relative priority: The ranking that makes some content more central, usable or action-relevant than alternatives.
Directional set: A goal-like orientation towards selecting, maintaining or changing a target.
Temporal stability: The persistence of a selected priority despite interference and drift.
Field disclosure: The breadth and immediacy with which centre, periphery and transitions are noticed.
Meta-awareness: Awareness of the current mental or attentional state while it is occurring.
Micro-phenomenology: A family of elicitation and analysis methods for detailed descriptions of singular lived episodes.
Rival generator: A different process that could produce the same report or behavioural output.

References

James, W. (1890). “Attention.” In The Principles of Psychology, Vol. 1, Chapter XI. Classics in the History of Psychology.
Petersen, S. E., & Posner, M. I. (2012). The attention system of the human brain: 20 years after. Annual Review of Neuroscience, 35, 73–89. https://doi.org/10.1146/annurev-neuro-062111-150525.
Watzl, S. (2023). What attention is: The priority structure account. WIREs Cognitive Science, 14(1), e1632. https://doi.org/10.1002/wcs.1632.
Corbetta, M., & Shulman, G. L. (2002). Control of goal-directed and stimulus-driven attention in the brain. Nature Reviews Neuroscience, 3, 201–215. https://doi.org/10.1038/nrn755.
Seli, P., Risko, E. F., & Smilek, D. (2016). On the necessity of distinguishing between unintentional and intentional mind wandering. Psychological Science, 27(5), 685–691. https://doi.org/10.1177/0956797616634068.
Seli, P., Risko, E. F., Smilek, D., & Schacter, D. L. (2016). Mind-wandering with and without intention. Trends in Cognitive Sciences, 20(8), 605–617. https://doi.org/10.1016/j.tics.2016.05.010.
Hasenkamp, W., Wilson-Mendenhall, C. D., Duncan, E., & Barsalou, L. W. (2012). Mind wandering and attention during focused meditation: A fine-grained temporal analysis of fluctuating cognitive states. NeuroImage, 59(1), 750–760. https://doi.org/10.1016/j.neuroimage.2011.07.008.
Lutz, A., Slagter, H. A., Dunne, J. D., & Davidson, R. J. (2008). Attention regulation and monitoring in meditation. Trends in Cognitive Sciences, 12(4), 163–169. https://doi.org/10.1016/j.tics.2008.01.005.
Dahl, C. J., Lutz, A., & Davidson, R. J. (2015). Reconstructing and deconstructing the self: Cognitive mechanisms in meditation practice. Trends in Cognitive Sciences, 19(9), 515–523. https://doi.org/10.1016/j.tics.2015.07.001.
Koch, C., & Tsuchiya, N. (2007). Attention and consciousness: Two distinct brain processes. Trends in Cognitive Sciences, 11(1), 16–22. https://doi.org/10.1016/j.tics.2006.10.012.
van Boxtel, J. J. A., Tsuchiya, N., & Koch, C. (2010). Consciousness and attention: On sufficiency and necessity. Frontiers in Psychology, 1, 217. https://doi.org/10.3389/fpsyg.2010.00217.
Block, N. (1995). On a confusion about a function of consciousness. Behavioral and Brain Sciences, 18(2), 227–287. https://doi.org/10.1017/S0140525X00038188.
Dehaene, S., Changeux, J.-P., Naccache, L., Sackur, J., & Sergent, C. (2006). Conscious, preconscious, and subliminal processing: A testable taxonomy. Trends in Cognitive Sciences, 10(5), 204–211. https://doi.org/10.1016/j.tics.2006.03.007.
Hadash, Y., Dar, O., Amir, I., Braver, T. S., & Bernstein, A. (2026). The Mindfulness Internal Attention framework: Uncovering the attentional mechanisms of mindfulness training. Annual Review of Psychology, 77, 255–283. https://doi.org/10.1146/annurev-psych-012925-030843.
Petitmengin, C. (2006). Describing one’s subjective experience in the second person: An interview method for the science of consciousness. Phenomenology and the Cognitive Sciences, 5, 229–269. https://doi.org/10.1007/s11097-006-9022-2.
Petitmengin, C., Remillieux, A., & Valenzuela-Moguillansky, C. (2019). Discovering the structures of lived experience: Towards a micro-phenomenological analysis method. Phenomenology and the Cognitive Sciences, 18, 691–730. https://doi.org/10.1007/s11097-018-9597-4.
Sparby, T., Lumma, A.-L., Edelhäuser, F., Glaser, R., Schnitzler, L., & Weger, U. W. (2021). First-person access to decision-making using micro-phenomenological self-inquiry. Scandinavian Journal of Psychology, 62(6), 798–805. https://doi.org/10.1111/sjop.12766.