🔬 Complete Methodology v2.0

The Institute Framework

Rigorous, reproducible, systems-based methodology for consciousness research.

"You cannot prove anything true. You can only fail to disprove it."
📋 Core Principles ⚗️ 3-Stage Process 🎓 Learn the Skills
↓
0
Core Principles
0
Foundation Stages
0
Total Stages (Full Method)
🧠
Did You Know?
Karl Popper revolutionized science by showing that theories can never be proven true—only falsified. This means every scientific "fact" you know is actually just something we've repeatedly failed to disprove. That's not weakness—it's intellectual honesty.
The Problem

Why Traditional Methods Fail

Standard scientific method has critical limitations for consciousness research.

LIMITATION 01
Many Discoveries Miss
Many major scientific discoveries don't follow the standard method. Rigid protocols suppress creative inquiry—penicillin, X-rays, and the microwave oven were all "accidents."
LIMITATION 02
Consciousness is Subjective
First-person experience ("what it's like") requires first-person data. Objective-only methods miss the phenomenon entirely—like studying music without ever hearing it.
LIMITATION 03
Complex Systems Need Systems Thinking
Feedback loops, emergence, multi-scale dynamics—consciousness can't be reduced to linear cause→effect. Your brain has roughly 86 billion neurons and an estimated 100 trillion connections.
This framework addresses these gaps while maintaining empirical rigor — documented stages, sourced claims, and results open for anyone to check.
🔄 Interactive: The Attention Degradation Loop
Click each node to see how the system reinforces itself

This is why systems thinking matters. Linear "phone bad" thinking misses the reinforcing feedback loop.

Foundation

5 Core Principles

Non-negotiable standards that make consciousness research actually work. Click cards to flip.

PRINCIPLE 01
Falsifiability Over Proof
You can never prove a hypothesis true. You can only fail to disprove it. Every claim must be stated so it could potentially be proven wrong.
Falsifiable Example
"Users show ≥20% improvement in metacognitive accuracy after 30 days"
👆 Click to see why this matters
WHY IT MATTERS
The Black Swan Problem
No matter how many white swans you see, you can't prove "all swans are white." But ONE black swan disproves it. Science works by elimination, not confirmation.
PRINCIPLE 02
Reproducibility Across Contexts
Findings must replicate in diverse settings by independent researchers. Not just lab reproducibility—real-world generalization.
Gold Standard
Multiple labs + varied conditions + independent verification
👆 Click for the replication crisis
THE CRISIS
Much Research Fails to Replicate
Psychology's replication crisis showed many published findings don't reliably hold up, and large replication projects have reproduced only a fraction — the Open Science Collaboration's Reproducibility Project in Science (2015) found 36% of 100 psychology studies replicated. We require convergent validation: if it's real, multiple independent methods should find it.
PRINCIPLE 03
Systems-Based Understanding
Consciousness exists in complex systems where parts interact to create emergent properties. Map feedback loops, multi-scale interactions, emergence.
Requirement
Components → Relationships → Mechanisms → Feedback → Emergence
👆 Click for the numbers
SCALE
100 Trillion Connections
Your brain: 86B neurons, 100T synapses, operating across milliseconds to decades. Linear cause→effect thinking is hilariously inadequate.
PRINCIPLE 04
First-Person + Third-Person Integration
Subjective experience (first-person) must integrate with objective measurement (third-person). Consciousness is inherently first-person—that data is valid.
Convergence
Self-reports align with behavioral performance gains
👆 Click for the hard problem
THE HARD PROBLEM
What It's Like
Third-person methods can't capture "what it's like" to see red. Dismissing first-person data dismisses the very phenomenon we're studying. Both required.
PRINCIPLE 05
Operational Definitions
Abstract concepts must be defined in measurable, observable terms. Enables independent verification and prevents moving goalposts.
Example
"Metacognitive awareness" = accurate self-monitoring + attention redirection
👆 Click for the woo problem
ANTI-WOO
No Unmeasurable Magic
"Raising your vibration" or "universal consciousness" without operational definitions = unfalsifiable = not science. If you can't measure it, you can't know it.
⚗️
Historical Fun Fact
Ignaz Semmelweis discovered handwashing could prevent maternal deaths in 1847. He was ridiculed, fired, and died in an asylum—because he couldn't explain the mechanism (germs weren't discovered yet). This framework requires mechanism specification so your good ideas don't die with you.
The Process

7-Stage Methodology

Stages 1-3 complete. Map the system BEFORE testing it.

Stage Details
Hover over or click on a stage to view its details. Use arrow keys to navigate between stages.
1
Pattern
Recognition
2
Question
Formation
3
Framework
Building
4
Hypothesis
Formation
5
Measurement
Design
6
Data &
Analysis
7
Synthesis &
Iteration
1

Pattern Recognition

"What's happening?"

Key Skills

  • Attention to anomalies—notice what doesn't fit
  • Multi-context observation—patterns across settings
  • Subjective + Objective—internal experience AND behavior
  • Systematic documentation—time-stamped, contextual

Common Errors

  • Confirmation bias—seeing only supporting evidence
  • Expectation effects—perceiving what you expect
  • Pattern imposition—creating patterns from noise
  • Salience bias—noticing dramatic, missing subtle

✓ Output

Raw observations: Specific descriptions, context, conditions, frequency/intensity data, initial curiosities sparked.

2

Question Formation

"What specifically do we want to know?"

Essential Components

  • Specific: Exact phenomenon, not vague
  • Measurable: Observable indicators exist
  • Bounded: Clear what IS and ISN'T included
  • Testable: Could be proven wrong
  • Relevant: Answer matters

8-Step Process

  • Extract core phenomenon from Stage 1
  • Generate 5-10 candidate questions
  • Choose ONE question
  • Operationalize (specific + measurable)
  • Define boundaries (IS about / NOT about)
  • Apply falsification test
  • Feasibility check
  • Connect to larger framework

✓ Output

Single well-formed question: Specific, measurable, bounded, testable, relevant + clear falsification criteria + feasibility confirmed.

3

Framework Building

"How does this system actually work?"

Core Concepts

  • Emergence: System behavior from interactions
  • Feedback Loops: Reinforcing (amplify) + Balancing (stabilize)
  • Multi-Scale: Neural → Cognitive → Behavioral → Social
  • Mechanisms: HOW X causes Y (actual processes)
  • Leverage Points: Where interventions work best

10-Step Process

  • Identify all components
  • Map relationships (visual diagram)
  • Specify mechanisms
  • Catalog feedback loops
  • Multi-scale integration
  • Identify leverage points (Donella Meadows)
  • Predict unintended consequences
  • Review existing knowledge
  • Synthesize coherent model
  • Generate 5-10 testable predictions

✓ Output

Complete framework document: System map + components + mechanisms + feedback loops + multi-scale integration + leverage points + testable predictions.

🎯
Leverage Point Insight
Donella Meadows identified 12 leverage points in complex systems. The highest-leverage interventions? Changing paradigms and goals—not tweaking parameters. That's why "use phone less" (parameter) fails while "I'm someone who protects their attention" (paradigm) succeeds.
Evidence Standards

How We Grade the Evidence We Cite

The bar published research must clear before this program leans on it — and the bar any future original study of ours would face.

The Institute synthesizes and analyzes existing published research; it does not run its own laboratory trials. The three standards below are how we weigh the studies we cite — not a description of in-house experiments. How we validate our own published claims is the next section.

🧠
Subjective
First-Person Experience
  • Validated self-report scales
  • Trial-by-trial assessment
  • Multiple converging measures
  • Acknowledge limitations
📊
Objective
Third-Person Performance
  • Validated behavioral tasks
  • Real-world outcomes
  • Multiple timepoints
  • Control conditions
⚡
Convergent
Gold Standard
  • Subjective + objective align
  • Multiple methods consistent
  • Replicates across samples
  • Persists in varied contexts
Transparency

How We Screen and Verify Our Own Claims

Two tiers, named so the label is the truth: AI-screened (an automated pre-filter) is never called "verified" — only a named human's check earns that word.

Verification pending

Our standard is that every public-facing claim about a named source is checked against its primary source, cited, and the result recorded — and we publish the live progress against that standard rather than asserting it is finished. We hold a hard line between two states: an AI-screen (an automated, fabrication-prone pre-filter — never labeled "verified") and human-verified (a named person checked the claim against its source). An automated claim scanner runs on every build, and the production build fails if it finds a single CRITICAL claim (currently zero). Reviewed pages are logged in a ledger with run identifiers, timestamps, and a revalidation cadence. The headline below is generated from that ledger at build time — no hand-typed numbers — and at launch the human-verified count is near zero: the human review is just beginning, and we show that plainly rather than round it up.

0 human-verified · 8 AI-screened · 566 pending (of 574 content pages)AI-screened is an automated pre-filter, not verification; every screened page is queued for human verification. Target: 100% human-verified.

Epistemic Honesty

Strength of Claims

Never overclaim. Communicate uncertainty honestly.

This is the rubric we use when reading and weighing the studies we cite — and the bar any future original claim of ours would have to meet. It is not a badge the published corpus currently carries; a project-wide grading scheme is under review, not yet adopted.

Preliminary
Weak Evidence
  • Single study
  • One measurement type
  • Limited sample
  • Lab context only
Supportive
Moderate Evidence
  • Multiple studies
  • Both subjective + objective converge
  • Diverse samples
  • Some real-world validation
Established
Strong Evidence
  • Extensive replication
  • Multiple measurement types
  • Diverse populations and contexts
  • Real-world outcomes confirmed
  • Independent labs verify
Avoid These

Common Pitfalls

Red flags to watch for and how to prevent them.

⚠️

Confirmation Bias

Only looking for evidence supporting hypothesis

✓ Pre-register predictions, seek disconfirming evidence
⚠️

P-Hacking

Running multiple analyses until finding significance

✓ Pre-register analysis, report all tests performed
⚠️

HARKing

Hypothesizing After Results Known

✓ Distinguish exploratory vs. confirmatory, label post-hoc
⚠️

WEIRD Samples

Only Western, Educated, Industrialized, Rich, Democratic

✓ Recruit diverse participants, acknowledge limitations
⚠️

Measurement Reactivity

Measuring changes what you're measuring

✓ Unobtrusive measures, multiple methods, control groups
⚠️

Correlation ≠ Causation

Assuming X and Y together means X causes Y

✓ Experimental design, test temporal order, third variables
Self-Audit

Interactive Checklists

Click items to track your progress.

🔍 Stage 1 Complete?

  • Systematic observations documented
  • Context and conditions recorded
  • Patterns identified across instances
  • Initial curiosities articulated

❓ Stage 2 Complete?

  • Specific, measurable question formed
  • Bounded scope clearly defined
  • Falsification criteria stated
  • Feasibility confirmed
  • Relevance established

🗺️ Stage 3 Complete?

  • All major components identified
  • Relationships mapped visually
  • Mechanisms specified
  • Feedback loops cataloged
  • Leverage points identified
  • 5-10 testable predictions generated
Resources

Where to Go Next

This page is the full methodology. Related tracks:

Version 2.0 • November 2025 • CC BY-SA 4.0

Built to Be Checked

A transparent, reproducible method — every claim attributed to a cited source, every stage documented, published under CC BY-SA 4.0 for anyone to verify.

📄 Read the Investigations 🎓 Learn the Skills 🤝 Collaborate