How behavior works, and how to change it.
Mechanistic Mindset uses ideas from computation to explain everyday behavior. The articles follow what causes a result so you can identify what to change.
Browse the articlesWhen should you follow an established method?
A trail may go around a hill because the climb was unsafe. If the conditions change, the detour may no longer help. The same problem appears when we follow a method without knowing why it was chosen.
Read Paths and Trailblazing
The route shows where to go, but not why it goes that way.Why everyday tasks become difficult
Difficulty often comes from how a task is set up. These articles examine the work required to start and the conditions that make progress easier to repeat.
Activation Energy
Getting ready can take more effort than the task itself. This article explains what contributes to that starting cost.
Read the article ↗Why does changing the setting help?Nature Alignment
An environment makes some actions easier than others. This article explains how to use that difference.
Read the article ↗How can a useful change last?Ratchet
A test or a physical boundary can preserve a change even when your attention moves elsewhere.
Read the article ↗Articles by topic
Start Here: The Foundation3 articles
Moralizing vs Mechanistic
Describes behavioral outcomes through their causes so that a character judgment can become an actionable diagnosis.
Meta-Pattern
Traces the common pattern across moralized terms and their computational translations.
Glossary
Translates moralistic descriptions into the corresponding mechanisms used throughout the wiki.
Central Concepts: How Behavior Works6 articles
State Machines
Treats behavior as discrete states, probabilistic transitions and default scripts that run automatically.
Activation Energy
Explains starting costs through the Boltzmann distribution and thermodynamic constraints on behavior.
Willpower
Uses a daily budget of 10–15 units to account for depletion and the cost of decisions.
Expected Value
Explains motivation through (reward × probability) / (effort × temporal_distance).
Working Memory
Connects a capacity of 4–7 items to task overload and the need to store context outside the mind.
Epistemic Contamination
Examines why information integrates automatically and how prevention avoids exposure that cannot be made read-only.
Practical Implementation: How to Build Systems7 articles
Prevention Architecture
Removes decisions at 0 cost instead of requiring 2–3 units of resistance each time they recur.
30x30 Pattern
Uses 30 consecutive days to reduce activation cost from 6 units to 0.5 through neural pathway caching.
Tracking
Records observable system state so diagnosis does not depend on remembering it accurately.
The Braindump
Clears working memory in a daily 10-minute session, reducing the cost of starting work.
Discretization
Divides a continuous process into countable units with identifiable completion points.
Rhythm
Arranges work and rest in temporal patterns that support continued execution.
Zeitgebers
Uses light, food and temperature as external cues for circadian alignment.
Meta-Frameworks: How to Think9 articles
Question Theory
Treats questions as compulsory computational operations, connecting search strategies with Cypher query optimization.
Language Framework
Chooses language suited to a domain and resolves paradoxes caused by mismatched syntax.
Signal Theory
Distinguishes authentic, generative Alpha signals from market-responsive Beta signals.
Gradients
Follows energy, learning and information gradients, then asks how changing their conditions changes behavior.
Nature Alignment
Builds on existing momentum, thermodynamic gradients and the paths previous actions have made available.
Pedagogical Magnification
Matches explanatory resolution to available computation and causal control, distinguishing two kinds of overthinking.
Superconsciousness
Describes an operator mode with kernel privileges, invoked through spells to deploy resources deliberately.
Autodidact Framework
Organizes self-directed learning around practice before theory, suitable textbook density and repeated verification.
Representational vs Real Constraints
Tests whether an apparent limit comes from the model of a situation or from the system itself.
Intelligence as Water: Exploration, Selection, and Embodiment19 articles
Intelligence Is Water
Treats intelligence as an intentionless fluid whose result depends on force × channeling structure, making the riverbed the object of design.
Branching and Convergence
Treats divergence and convergence as the two basic operations of all cognitive work; shipping requires the often-missing step of merging alternatives.
Selection Over Design
Shows how repeated sampling, selection and promotion can produce the appearance of genius design, with reality selecting what a person generates.
Macrostate Engineering
Defines the desired large-scale state and lets computation resolve its detailed configuration.
Order of Determination
Resolves the representation with the most degrees of freedom first, locating anxiety in upstream choices left unconstrained.
Container Design
Builds bounded contexts that change the probability of activities by loading particular scripts.
Ignition
Separates starting pressure from judging pressure and uses warm-ups to load an identity when starting is the limiting step.
Accrual Substrate
Explains why actions need a persistent place to accumulate visibly, linking memory to agency.
Agent Body
Defines an agent's body through constraints, state, persistence and a surface for contact with reality.
Reality Contact Metabolism
Treats contact as a nutrient, anxiety as a deficiency signal and intelligence used before contact as a liability.
EV Sensor Calibration
Explains motivation as a sensor reading changed by lived action–outcome pairs, never by information alone.
Generative vs Retrospective
Separates two irreconcilable ways of experiencing an unfolding process: living through it and judging the story left afterward. Retrospective language brings self-judgment into the process.
Upstream Router
Places the self upstream of existing structures, routing computation with a light touch while consciousness audits it.
Be the Sun
Develops an agency arrangement in which a person generates and surrounding systems capture the output.
Bounded Search
Gives search the budget it needs to converge and identifies unbounded searching as paid avoidance of selection.
Structure Over Request
Builds depth, authority and voice through the computation graph that produces them.
Memory Is the Substrate
Treats data as code and processes as temporary executions over persistent state, making substrate design the design of the system.
Coherence Is Not Evidence
Separates internal consistency from truth and requires sampling before a belief is promoted.
Taste Compilation
Converts scarce felt judgment into language and then machinery so that a system can retain each improvement.
Retention, Boundaries, and Adjudication18 articles
Ratchet
Retains closed errors by blocking exits from good states; persistence pays only when something survives between trials.
Lockfile
Pins the meaning of good with a versioned skill, golden set, verifier and rejection ledger kept together.
Basin
Gives a pursuit a persistent question, retrieval boundary and living synthesis so that information has somewhere to collect.
The Membrane
Draws the ownership boundary around what stays open and retains state after attention leaves, creating a countable unit.
Afferents
Supplies inbound channels that wake a system without a human relaying the event; a system that waits for its owner dies.
Mode-Lock
Explains how an active behavioral mode reinforces itself in either direction and collapses when its keystone anchor is removed.
Spell-Packet
Judges a statement by its effect on the receiving agent rather than its truth-value, allowing deliberate invocation.
The Simulated Other
Distinguishes a mind's rendering of another person from a response obtained by actually asking them.
Provenance
Traces how a value was produced to determine what evidential weight it deserves.
Counterparty
Places an external agent on the other side of a loop so that deadlines bind and judgments of clarity can converge.
Artifact-Mode
Begins contact with a completed artifact for a named receiver, treating seeing as the only channel through which learning occurs.
Planning Euphoria
Traces the loss of next-day motivation when planning itself delivers the reward expected from shipping.
The Precondition Trap
Distinguishes doing an act from endlessly building what would enable it. Repeated attempts create the need for a system; feelings alone never build one.
The Mastery Ladder
Builds capability in rungs that can support later work only once repeated intervention is no longer needed.
Invokable Structures
Gives agency callable external structures; a goal becomes valid when something can respond to its error signal.
Error Signal
Gives a system a measurable difference between intended and actual state. Most domains lack this signal, so it must be constructed before resolved errors can be retained as rules.
Legibility
Makes state readable to agents that otherwise cannot act on it. Hidden conditions get mistaken for character defects; instrumentation reveals what persuasion cannot.
Brand Invisibility
Separates the person from the surface publishing their work so that rejection reaches the venture and further attempts remain possible.
Cross-Disciplinary Imports10 articles
Computation as Core Language
Develops computation as a universal language for describing mechanisms.
Computational Literacy
Teaches from natural patterns and physical computation before introducing syntax and digital machines.
Cybernetics
Examines feedback, control and goal-directed behavior with limited resources.
Statistical Mechanics
Transfers the Boltzmann distribution, entropy and thermodynamic constraints as metaphors for behavior.
Information Theory
Weighs uncertainty reduction against the cost of obtaining information.
Optimal Foraging Theory
Balances exploration and exploitation when search has a resource budget.
Predictive Coding
Connects prediction error, the six-layer cortical architecture and temporal causality.
Neural Positivism
Explains absence as computed prediction error in a brain that processes only positive signals.
AI as Accelerator
Accelerates tested paths while retaining the need for temporal exposure and encounters with unknown unknowns.
Digital Daoism
Connects Daoist thought with an understanding of computational substrates.
How this wiki is written
The articles develop models from Will’s observations and experiments. Examples show how an idea is meant to work so you can decide whether it fits your situation.
About Mechanistic Mindset