The Research Program

The Minimal Postulates

A theory this broad can easily become impossible to test because every claim supports every other claim. The first discipline is compression. What is the smallest set of commitments from which DOT’s distinctive architecture follows?

Book One compresses DOT into seven candidate postulates.

P0 — Occurrence. Experience occurs. This is the minimum datum, not a uniquely DOT claim.

P1 — Big C. Consciousness is ontologically fundamental rather than wholly produced by nonconscious physical processes, and it is capable of ordered differentiation, retained consequence, and self-preserving development. Big C names this joint hypothesis; T and E name capacities attributed to it, not conditions prior to it.

P2 — Differentiation. Fundamental self-awareness can operate at a local scope, called Little c. Big C and Little c share that fundamental nature; a local instance’s distinctive history and developed identity depend on the state carried by its Canvas.

P3 — Rule-bound mediation. Localized experience occurs within structured Reality Frames that constrain state transition, embodiment, action, and consequence.

P4 — Interface coupling. In RF₀, Little c is coupled to but not identical with the biological interface. Intent contributes causally through that coupling.

P5 — Persistent learning. The Canvas carries and updates state. Consequential experience changes its accumulated organization, the Painting, shaping later interpretation and action.

P6 — Development through widening agency. Changes in Canvas state can widen what is available to local awareness and expand meaningful decision-space under stable constraints. Governing Fear tends to contract that space; Love names the condition and practice in which Fear no longer governs it.

These postulates carry different statuses.

P0 is a starting observation. P1–P4 are ontological commitments. P5 contains an empirically accessible psychological claim and an unverified claim that persistence is not exhausted by biological implementation. P6 contains measurable psychological and social hypotheses alongside a larger cosmological interpretation.

The compression reveals dependencies.

If P1 fails, the practical Canvas and agency models may remain useful, but DOT’s ontology fails. If P2 fails, Big C may remain as a form of cosmopsychism while Little c requires another account. If P4 fails, Decoupling and Intent as a non-neural causal contribution must be abandoned or radically revised. If the psychological component of P5 fails, the Experience Loop loses its organizing value. If P6 fails, Fear and Love cannot carry the theory’s developmental claims in their present form.

A mature theory should become simpler as evidence removes unnecessary structure.

A Ledger of Claims

Every major DOT statement should be entered into a living claims ledger with six fields:

18. Claim: one sentence that can be evaluated without reading the author’s mind.

19. Level: observation, external model, DOT derivation, or speculative extension.

20. Rivals: the strongest alternative explanations that predict some of the same evidence.

21. Operationalization: the observations or measures connected to each term.

22. Discriminator: the result expected to differ across models.

23. Revision rule: what result lowers confidence, narrows scope, or removes the claim.

Consider the statement “higher Canvas entropy reduces Intent coherence.” The terms are not ready merely because the direction sounds plausible. Canvas entropy might be operationalized through cross-context conflict among reported values, action variability, intrusive attentional capture, and physiological recovery. Intent coherence might include stability of a chosen aim, correspondence between commitment and action, and recovery after interruption. The study would need to distinguish fragmentation from ordinary exploration, ambivalence, cognitive load, fatigue, depression, anxiety, and limited external options.

The discriminator would not be a simple correlation. Distress, threat, and low control could cause both higher measured fragmentation and lower action stability. A stronger design would manipulate or longitudinally alter a candidate source of fragmentation, measure the mediating process, and test whether changes generalize across tasks and contexts.

The revision rule might state that if preregistered multi-method indicators do not converge, the proposed Canvas-entropy construct should be divided, renamed, or abandoned.

That is what it means to make the model answerable.

Rival Families

DOT should not define its rivals as people who ignore experience. Many physicalist, functionalist, biological, enactive, panpsychist, idealist, illusionist, and dual-aspect approaches take consciousness seriously. The differences concern explanation and ontology.

At minimum, DOT must be compared with five families.

Neural production. Conscious experience is produced or constituted by organized neural activity. Reports, agency, memory, and selfhood can in principle be explained through biological processes and their interaction with environment and culture.

Functional or computational organization. Consciousness depends on an appropriate causal or informational organization, potentially realizable in more than one physical substrate. The relevant boundary is functional, not necessarily biological.

Biological naturalism or emergentism. Consciousness is a real biological feature arising from living nervous systems, even if it is not reducible in a simple way to lower-level descriptions.

Panpsychist or dual-aspect models. Experience or proto-experiential properties are fundamental features of reality, but there may be no intentional Big C, generated Reality Frames, or non-neural authorship as DOT describes them.

Idealist or cosmopsychist models. A universal consciousness is fundamental, and local minds depend on or differentiate within it. These accounts are closest to P1 and P2. DOT must distinguish itself through its specific derivation of Reality Frames, interface coupling, Canvas development, and Love/Fear dynamics.

Agreement with DOT’s foundation does not confirm DOT’s architecture.

The theory also needs null models for its practical constructs. Habit learning, predictive processing, interoception, threat regulation, metacognition, psychological flexibility, self-determination, and social determinants may explain the measured results without Little c or a Canvas that persists independently of biology. If DOT merely renames their joint operation, the integrative vocabulary may still be useful, but the evidence does not rise to the ontology.

Six Research Programs

Program One: A Science of Disciplined Subjective Data

The first program requires no acceptance of Big C.

Its question is methodological: can trained first-person reporting improve the prediction and explanation of measured behavior and physiology beyond ordinary report and third-person measures alone?

Participants would learn to distinguish sensation, valence, arousal, action tendency, appraisal, attention, and confidence. Tasks could involve perceptual ambiguity, developing intention, bodily arousal, or emotionally meaningful decisions. Reports would be time-locked and compared with behavioral, autonomic, neural, and contextual measures.

The critical comparison is not “subjectivity versus science.” It is among methods:

• untrained retrospective report;

• trained structured report;

• behavioral and physiological measures without report;

• integrated first- and third-person modeling.

The primary outcomes should be preregistered: predictive accuracy, reliability across sessions, convergence among independent raters, sensitivity to experimental manipulation, and incremental explanatory value.

Possible success would show that disciplined first-person data reveal stable structure omitted by standard methods. That would support DOT’s methodological foundation, not P1. Possible failure would show that training adds demand characteristics, verbal confidence, or complexity without reliable predictive value.

The program should include participants whose experience and communication differ from the usual university sample. Culture, language, disability, neurodivergence, contemplative training, and clinical condition may alter both the accessible distinctions and the vocabulary used to report them.

Rigor requires methods that permit unexpected structure, not training everyone in the author’s preferred descriptions.

Program Two: Intention, Commitment, and Decoupling

The second program separates a tractable timing science from DOT’s strongest causal claim.

The tractable questions include:

• How does an urge develop into commitment?

• When does a participant gain information sufficient to predict, modify, or veto their own action?

• How do timing estimates change across retrospective clocks, real-time probes, continuous reports, and behavior?

• Which neural events track general preparation, specific action content, experienced authorship, commitment, execution, and later report?

Experiments should move beyond purposeless finger movements. Deliberate decisions involving reasons, values, delayed consequence, social meaning, and genuine alternatives are slower and harder to control, but better represent the agency DOT discusses. Arbitrary and deliberate decisions should be compared rather than treated as interchangeable.

A first phase can model intention as a trajectory. Participants provide confidence about an emerging choice at repeated probes. Neural and physiological data are analyzed with out-of-sample prediction. The design estimates the point at which action content becomes decodable, the point at which the participant can use that information, and the point after which reversal becomes unlikely.

None of this by itself tests an ontologically distinct Little c.

The Decoupling claim requires a stronger discriminator: a lawful contribution from committed Intent that is not exhausted by prior physical state. A proposed test would need:

• high-coverage measurement of relevant neural and bodily state before commitment;

• a task in which the commitment carries information not supplied by an external random cue;

• prospective predictions from both neural-only and Decoupling models;

• strict separation of exploratory model building from confirmatory evaluation;

• independent replication across laboratories;

• a principled account of why any residual is not omitted biology, stochastic dynamics, measurement error, or overfitting.

That last requirement is formidable. Practical unpredictability is not evidence for causation beyond the measured biological system.

Classifier failure alone should not count as victory for DOT.

If neural models progressively predict choice, timing, confidence, reversal, and report with no stable residual structure unique to committed Intent, confidence in P4 should decrease. If an information boundary persists under increasingly strong measurement and adversarial analysis, the interaction hypothesis earns further attention but is not yet proven. A bridge law would still be required.

Program Three: Canvas, Memory, and Character

The third program asks whether the Canvas provides a valid integrative construct for persistent learning.

It begins with a narrower claim:

Patterns distributed across explicit memory, implicit learning, bodily regulation, attention, valuation, and social expectation jointly predict interpretation and action better than any one layer considered alone.

Longitudinal studies are essential. A Canvas is defined by persistence and update; a one-time questionnaire cannot establish either.

Participants could identify a recurring conflict between values and behavior that is safe to study. Over weeks, researchers would measure daily context, bodily state, predicted threat, available action drafts, committed Intent, action, consequence, and later update. Laboratory tasks would measure attentional flexibility, inhibition, generalization, and physiological recovery. Qualitative reports would be coded without forcing every experience into DOT language.

Several models would compete:

• a stable trait model;

• a context-and-state model;

• a habit-learning model;

• a multi-level Canvas model;

• a model allowing the measurement structure to change through learning.

The Canvas earns value if it supplies preregistered incremental prediction, identifies meaningful transition points, and supports interventions that generalize beyond the trained task. It loses value if indicators do not converge, if every result can be explained after the fact, or if ordinary constructs predict equally well with less complexity.

Repainting studies should avoid dramatic claims of erasure. Outcomes should include the availability of additional responses, reduced dominance of the old policy, context generalization, durability, and relapse under stress. Negative and mixed outcomes matter because a new policy can coexist with an old one.

The ontological Canvas claim requires another boundary entirely: evidence that learned organization or verifiable information persists without available biological storage. That question belongs to Program Five.

Program Four: Fear, Love, and Decision-Space

The fourth program operationalizes the theory’s practical center.

DOT predicts that governing Fear narrows effective decision-space. The prediction should be compared with more specific models of acute threat, anxiety, cognitive load, avoidance, trauma, scarcity, and social control.

One experimental family could manipulate credible social-evaluative threat while measuring:

• the number and diversity of action alternatives generated;

• the ability to identify long-term consequences;

• tolerance for disconfirming information;

• physiological activation and recovery;

• action rigidity under changing context;

• experienced autonomy and confidence.

The key phrase is credible threat. Deception and stress induction raise ethical limits. Studies should minimize harm, provide meaningful consent and debriefing, and avoid recruiting vulnerability merely because it increases an effect.

A longitudinal family could examine naturally changing conditions of safety, resources, social support, and institutional voice. Does effective decision-space expand when objective constraints improve? Do some learned threat policies persist after conditions change? Which relational or reflective practices help new safety become usable?

Love cannot be manipulated as a cosmic force. Its human operationalization can be tested through a profile: accurate contact, values-consistent flexibility, proportional boundaries, concern for others’ agency, evidence updating, and repair after harm. The profile must be discriminated from agreeableness, conflict avoidance, positive mood, social desirability, and ideological conformity.

The theory predicts that lower governance by Fear should not simply produce more permissiveness. It should improve context-sensitive action. Under real danger, an accurate boundary may become faster and clearer; under difference without danger, curiosity and tolerance should remain more available. This interaction is more distinctive than a general claim that calm people make better decisions.

At the institutional level, research can ask whether schedule control, voice, security, fair procedure, and reliable repair increase autonomy, trust, error reporting, learning, and collective problem-solving. Again, these results can support the practical model without proving Big C.

If Fear measures merely reproduce established anxiety or threat constructs, DOT should narrow or integrate the term. If Love cannot be measured without circularly labeling preferred outcomes as loving, it should remain an ethical ideal rather than a scientific variable.

Program Five: Continuity Beyond the Interface

The fifth program addresses DOT’s most consequential and most easily corrupted claims.

Does any verifiable information or organized personal continuity persist when ordinary biological storage and communication are unavailable?

The domain includes claimed near-death perception, terminal lucidity, anomalous information, mediumship, and past-life memory. The existence of reports is not in dispute. Their interpretation is.

Retrospective stories are not an adequate foundation. Strong designs must be prospective, blinded where possible, registered before data collection, independently monitored, and built to prevent ordinary information transfer. Target information should be securely generated and audited; inclusion and exclusion rules fixed; all attempts reported; fraud controls and data provenance explicit; and analysts kept blind until pipelines are frozen.

Medical uncertainty requires particular care. A patient described informally as having “no brain activity” may not meet that condition. Periods of impaired circulation, anesthesia, or unresponsiveness do not establish that all biological processes capable of later memory were absent. Claims must match documented physiology.

The primary hypothesis should specify the information expected, the channel ordinary models exclude, the scoring rule, and the effect size that would justify continued study. A vague match judged after the story is known is not enough.

Evidence from this program would need replication by skeptical and sympathetic teams together. A single dramatic case may motivate study but cannot carry the ontology.

Persistent null results under increasingly strong designs would not logically disprove every form of Little c. They would weaken claims of information-preserving continuity accessible through the tested channels. DOT would need to narrow P5 or admit that its survival implications are not empirically supported.

Program Six: Constraints on Reality Frames

The sixth program is theoretical before it is experimental.

DOT currently describes why a stable, consequence-bearing environment would support learning. It does not derive the laws of RF₀. The first task is to convert developmental language into constraints.

A Frame supporting cumulative embodied agency may require:

• distinguishable and sufficiently persistent states;

• lawful transition with bounded uncertainty;

• local information limits, so perspective remains genuinely situated;

• stable memory-supporting structures;

• energy gradients and nonequilibrium processes capable of self-maintenance;

• interfaces through which actions have delayed and discoverable consequences;

• multiple agents whose streams overlap through a shared world;

• enough plasticity for learning and enough stability for learning to matter.

Researchers could ask what classes of dynamical systems satisfy these constraints and whether they yield nontrivial restrictions on possible physics. The result may remain compatible with many worlds. Progress would still come from showing which parts of Reality Frame language follow from development and which were borrowed from software intuition.

The model must also examine alternatives. A world could support experience without biological evolution as we know it. A deterministic Frame might support meaningful agency under a compatibilist interpretation. A highly stochastic Frame might prevent stable learning. A world with unrestricted information access might prevent individuation. These are theoretical experiments on the architecture.

The multiple-Frame hypothesis should remain dormant until a second rule-domain has evidence beyond altered content. Scientific restraint is part of the program.

Methods That Protect the Question

Consciousness theories are vulnerable to flexible interpretation because the target is difficult, the measurements are indirect, and the theories often operate at different levels. Psychology and related fields have also confronted substantial reproducibility problems. In a large collaborative replication project, replication effect sizes were on average smaller than original effects, with success depending on the criterion used.⁴⁸

The lesson is not that psychology cannot know anything.

The lesson is that method must anticipate the human desire to find what a theory predicts.

DOT research should therefore use:

Preregistration. State hypotheses, exclusions, primary outcomes, and analysis plans before seeing the relevant results.

Registered Reports. Seek peer review of the question and method before data collection, with publication not dependent on a positive result.

Adversarial collaboration. Include investigators committed to rival interpretations in designing discriminating tests. The 2025 adversarial comparison of global neuronal workspace and integrated information theories demonstrates both the value and difficulty of forcing theories into prospective confrontation.²

Formal models. Verbal theories can accommodate too much. Formalization should produce quantitative or ordinal predictions, not decorative notation. Psychology’s “theory crisis” has likewise motivated calls for models that make clear predictions and distinguish confirmation from post hoc fit.⁴⁹

Out-of-sample prediction. Build models on one set of participants or conditions and evaluate them on another.

Multi-method measurement. Combine first-person report, behavior, physiology, neural measures, context, and longitudinal consequence where the claim crosses those levels.

Independent replication. A laboratory that developed the theory should not remain the sole judge of its effects.

Open materials and null results. Share protocols, code, de-identified data where ethical, and failed predictions. A hidden negative result silently protects a weak Painting.

Measurement invariance. Ask whether a construct means the same thing across cultures, languages, bodies, and contexts before comparing scores.

Model reduction. Prefer the smallest model that accounts for the result. Do not assign every unexplained variance component to Little c.

These methods do not remove subjectivity from science; they make more of its influence visible and correctable.

Ethics of a Consciousness Research Program

The ontology has moral implications, but uncertainty has moral implications too.

If consciousness is present where behavior cannot express it, failure to detect it can lead to profound harm. If researchers overinterpret ambiguous signs, families can be given false hope. If a theory of Canvas and Fear is applied as diagnosis without validation, it can blame patients, suppress dissent, or authorize unqualified intervention.

DOT research should therefore adopt several constraints.

• Do not infer another person’s Fear, Intent, or developmental level solely from disagreement.

• Do not use Little c to minimize brain injury, medical treatment, disability, or material need.

• Do not recruit people in crisis into high-risk practices designed to prove Decoupling.

• Do not treat altered states as automatically truthful or demand them as a path to development.

• Do not promise survival after death as a scientific conclusion.

• Do not rank human worth by a proposed measure of Canvas coherence.

• Do not build an authority structure in which criticism of DOT is interpreted as pathology.

Participants should know which claim level a study addresses. Community input should shape research involving disability, trauma, end-of-life care, contemplative practice, and culturally specific experience. Data about interior life deserve unusual privacy protection.

A theory that claims to protect decision-space must protect the decision-space of the people used to test it.

A Staged Roadmap

The program should proceed in stages rather than beginning with the most dramatic claim.

Stage 1 — Definition. Freeze a versioned glossary, postulates, dependency graph, claims ledger, and rival matrix. Stress-test the adopted Big C/T/E relation, and develop the individuation rule enough for discriminating comparison.

Stage 2 — Measurement. Develop and validate narrow measures for structured subjective report, action-draft diversity, Intent commitment, Canvas conflict, and Fear-governed narrowing. Publish failures.

Stage 3 — Practical predictions. Test whether integrated Experience Loop models improve longitudinal prediction and whether specific interventions widen decision-space under stated boundary conditions.

Stage 4 — Adversarial mechanism tests. Compare neural-only and Decoupling models of developing intention. Require preregistered discriminators and independent replication.

Stage 5 — Frame constraints. Formalize what developmental requirements imply for classes of lawful environments and compare those restrictions with known RF₀ physics.

Stage 6 — Extraordinary boundaries. Only after methods and governance mature, conduct tightly controlled research on information or continuity beyond ordinary biological channels.

The ordering matters. A theory should demonstrate discipline on accessible claims before asking the public to trust it on inaccessible ones.

A staged research program

• 1 · Definition — Fix working terms, postulates, dependencies, and rival accounts.
• 2 · Measurement — Validate narrow constructs and first-person methods.
• 3 · Practical prediction — Test Experience Loop models across time and contexts.
• 4 · Mechanism — Compare neural and Decoupling accounts through differing predictions.
• 5 · Frame constraints — Derive restrictions on possible lawful environments.
• 6 · Continuity — Investigate claims beyond the interface with mature controls.

Table 5 · A staged research program. DOT begins with definitions and measurable constructs before approaching its most extraordinary claims.