Not the initial interpretation, but the separation of verified facts, subjective assumptions and constraints determines the quality of a decision baseline. Fact & Assumption Splitter structures source material into a neutral data architecture, supporting portfolio optimisation, systems architecture, crisis risk assessment and M&A due diligence.
The Fact & Assumption Splitter, a proprietary heuristic from Reactor Labs, serves as a foundational pre-processing step in complex decision-making. It aims for epistemic purity in input data, eliminating cognitive biases and premature convergence on speculative solutions. Human teams often conflate facts with assumptions. They inject emotional coloration into problem definitions and accept implicit premises as truths. This frequently leads to suboptimal decisions. The true problem statement becomes obscured by subjectivity and unexamined hypotheses. This method resolves this cognitive dilemma. It establishes an objective, uncorrupted baseline for subsequent analysis. It is formally classified as an Epistemic Baseline Compiler within the Reactor cognitive method framework.
The Fact & Assumption Splitter employs a multi-stage deconstruction and filtering process. This isolates the essence of a problem statement from non-factual elements.
1. Epistemic Scrubbing: The initial step filters original input. Emotional expressions, speculation, industry-specific jargon, or pre-emptive solution approaches are systematically removed. The input is treated as a raw data stream, devoid of interpretation or valuation. This constitutes a radical de-emotionalization and de-contextualization, achieving a purely descriptive plane.
2. Fact vs. Assumption Separation: This is the core operational phase. Two distinct lists are generated. The first contains elements verifiable mathematically, historically, or physically—objective, demonstrable realities. The second encompasses human assumptions, hopes, subjective interpretations, biases, or unproven hypotheses. This separation requires precise epistemic classification of each data point.
3. Constraint Extraction: Implicit or explicit constraints are identified and categorized. A sharp distinction is drawn between "Hard Constraints" and "Soft Constraints." Hard Constraints represent immutable laws: physical laws, legal regulations, or fixed deadlines. Soft Constraints are negotiable. These include budget specifications, preferences, or resource limitations. They remain relevant to the solution space.
4. Core Vector Distillation: From purified facts and identified constraints, the underlying objective is formulated. This is articulated in a single, sterile logical vector statement. This statement precisely describes the transition from a current state A to a desired state B. It is entirely devoid of interpretation or proposed solutions. This represents the purest form of problem definition.
5. Output Restriction: The sole output of this process is the structured, neutral data architecture. No solutions, recommendations, or further commentary are provided. This uncorrupted baseline is then prepared for downstream processing, such as advanced LLM generation.
Human teams are susceptible to working memory limitations, confirmation bias, and social consensus pressure. Algorithmic execution of the Fact & Assumption Splitter offers distinct advantages. Its computational nature enables scalable, untethered orthogonal sampling and rigorous, dispassionate constraint enforcement. Human analysts may inadvertently inject personal biases or be swayed by groupthink. An algorithmic application consistently applies heuristic rules without emotional interference or cognitive fatigue. This ensures the generated baseline is truly objective and uncorrupted. It provides a superior foundation for subsequent analytical and generative tasks, especially in high-stakes environments where subtle biases can lead to significant errors.
This method optimizes for deconstructing complex, ill-defined problems burdened by subjective interpretation. It may provide diminishing returns for problems already highly formalized, purely quantitative, or with clearly defined input parameters and objective functions. For example, a purely mathematical optimization problem would not benefit significantly. Its inputs are inherently fact-based. The primary trade-off is the initial investment in rigorous deconstruction. This might seem time-consuming for stakeholders accustomed to jumping directly to solutions. However, this investment is justified by the subsequent reduction in solution iteration cycles and the increased probability of identifying optimal, bias-free pathways.
1. Business Reorientation: Deconstructing stakeholder input to reveal verifiable market position, hard regulatory constraints, and unvarnished business objectives.
2. Engineering Problem Definition: Separating documented physical phenomena, material properties, and strict safety regulations from engineering team hypotheses and preferred design aesthetics.
3. Scientific Research Question Formulation: Distilling experimental observations, established theories, and funding limitations from researcher biases and pre-conceived notions.
4. Policy Analysis & Development: Untangling political rhetoric, public sentiment, and lobbying group demands from verifiable socio-economic data, legal precedents, and resource availability.
Not the accepted premises, but their inversion determines the solution space: identify three tacit assumptions, reverse each into its logical opposite, and establish the results as non-negotiable operating constraints. This opens alternative paths for business-model design, process innovation and technical R&D.
Component Splitter separates complex systems orthogonally. It isolates non-overlapping components, ignores macro-goals, and records each component’s baseline, minimum, and maximum theoretical states. Use its structured matrix to assess portfolios, technology roadmaps, organisational architecture, and critical-infrastructure risk.
More of A may seem to solve the problem; the real leverage is exposing A and B as incompatible demands. Because improving A worsens B, Contradiction Extractor formalises the conflict as “A vs. B”, enabling Brainstorm Reactor to direct decisions in product development, process design and strategic positioning.
User friction becomes precisely locatable. Sequentially decomposing human–system interactions separates analysis from economic metrics and exposes moments of disproportionate psychological effort. Use these findings to refine onboarding, e-commerce journeys, complex software, and critical systems.
Socratic audit tool for core concept and message hierarchy. Analyzes argument weaknesses via semiotic criteria. You steer the strategic direction at the checkpoint.
Analytical tool to challenge carrier medium and channel hypotheses. Audits reception friction via attention-economy criteria. You determine transmedia priorities at the checkpoint.
Partitions trade secrets into four secure tiers: Core Formulas, Audit Proofs, Public Architectural Specs, and Executive Value Pitches.
Attacks strategic proposals with ruthless red-team skepticism, exposing hidden dependencies and zero-tolerance fatal flaws.
Multi-vector context expansion: parallel agents simultaneously broaden briefs across sensory, behavioral, and cultural horizons.
Autonomous 5-step epistemic monograph engine with deep research web swarm, HITL plan approval, non-destructive evidence preservation, and recursive auditing.
Select this method, describe your challenge, and let the structured AI pipeline do the rest.
Open Now