algorithmacy-lab

An open computational lab that applies exact integrated information (Φ, from IIT 4.0, computed with PyPhi) to organizational coordination theory, and runs an AI-assisted research protocol that takes a question to a finished quantitative paper.

New here? Start with OVERVIEW.md — a five-minute, honest review of what the lab argues, where it actually stands (including what is recovered from other fields and what is genuinely new), and where you could contribute. The directory below is the map; the overview is the orientation.

Exploring with an AI assistant (Cursor, Claude Code)? Ask it "what is this and where do I start?" — it follows START_HERE.md, a guided onboarding that gets you to your first computed Φ verdict and your own research question in about fifteen minutes.

The thesis it tests: a coordination form is dyadic when its cause-effect structure factors into independent pieces (it demands literacy), and triadic when the structure stays irreducible across the worker–system–counterpart partition (it demands algorithmacy). On systems small enough to compute exact Φ, that verdict is exact, not a proxy.

Scope, stated up front. The core results are in-silico: exact Φ on small Boolean dynamical models of coordination. They are evidence about the models. A validation gap separates them from empirical claims about real organizations (cross-model agreement is internal validity, not external validity). This is a research program and a proof-of-method, not peer-reviewed findings about real firms. The recurrence program has begun reading real coordination data (the empirical ask below has the detail), on a methodological side, away from the core thesis.

The empirical ask — bring the lab real data

Every result here is in-silico or pre-fieldwork, and the validation gap is the program's largest open need. Closing it is community work: each of the four empirical arms below is waiting on a different kind of real data, and the highest-value contribution available is to go get one piece of it. Bring it through a pull request into contrib (see CONTRIBUTING.md); the lab is reproducibility-first and fixes its claims before computing, so a contributor commits the questions before the data.

• Field — one real arrangement, modeled to a Φ verdict. No real worker, platform, or message has been measured. The field protocol needs one real coordination — a team, the system that mediates it, the counterpart — with the determination rules (what each party's next action depends on, who reads whom) elicited from interviews, observation, or system documentation, and a second coder's agreement on the encoding. That turns a real arrangement into a Boolean model and an exact-Φ verdict. This is the deepest gap.
• Qualitative — one worked study, run as fieldwork. The studies under org_frontier/qualitative/ are designs awaiting data: a neonatal bedside, a content-moderation team, a clinical handoff, a brokered market, each read against a named prior. Each needs access to the setting and the interviews, observation hours, and documents to read it. Run one, hold the prior open, and report where the real setting departs from the catalog's expectation.
• Recurrence — one recorded series the lab has not touched. The behavioral instrument has read open-source merge graphs (v8–v10) and wants more, ideally where the determination actor is recorded so both Φ and cross-recurrence can run: a bot-merged project, a dispatch or handoff log with a named gatekeeper, or a physiological series such as the infant vitals the neonatal study describes.
• Survey — field the instrument with a real cohort. The survey arm measures the competence directly: self-report from people inside a real coordination arrangement, on a fixed schedule, the questions committed before the answers. The pre-registered panel needs a cohort and the access to run it; any workforce coordinating through a system beyond its control is a candidate.

Each ask now has a ready-to-run handoff packet — a front-door README, the pre-registration discipline, and a runnable scaffold that takes a real input to a verdict. The five are indexed in org_frontier/HANDOFF_PACKETS.md. Each program README carries the specifics, and the overview lists the open contributions in order of value. Dig as deep as you like from there.

The programs

The lab runs a computational program and five arms that carry it onto real coordination and into cognitive theory. Each is a different lens on the same coordination arrangement.

• The computational program in org_frontier/ computes exact Φ on Boolean models through a fixed six-stage protocol — review → deep research → five hypotheses fixed before computing → methods → run against the exact-Φ instrument → paper. Its corpus of threads, studies, and questions stands as a catalog of pre-disclosed priors: the structural reading of what a coordination form is.
• The field program in org_frontier/field/ takes one real coordination arrangement through a nine-step protocol to a dyadic or triadic verdict on an explicit model of it. It is the bridge from the in-silico priors to a real case, where the determination rules are elicited from interviews, observation, and documents.
• The qualitative program in org_frontier/qualitative/ reads real coordination settings against those priors, on the questions of process and meaning that fieldwork answers and a model cannot. It holds each prior open, and builds its contribution where a real setting departs from the catalog's expectation.
• The recurrence program in org_frontier/recurrence/ pairs Φ with cross-recurrence quantification, the behavioral reading of how the parties' states track each other over time. Φ says whether a coordination is irreducible; cross-recurrence says how its signal moves, and which party leads.
• The survey program in org_frontier/survey/ measures the competence in people: a human-subjects arm that reads algorithmacy as a lived skill through self-report from workers inside a real coordination arrangement, the measurement side of the validation gap the structural arms cannot reach.
• The cognition program in org_frontier/cognition/ is the formal bridge to the cognitive theories of coordination. Each theory of mind models a channel between two parties and has nowhere to put an opaque, interested third; the major complex holds that third as a member of the irreducible core, and each failure point becomes a computable Φ prediction.

Four readings meet on one arrangement: structure, fieldwork, behavior, and measurement. The field protocol turns real evidence into the structural reading, and the cognition program ties that structure back to the theories of mind it answers. Φ is the structural instrument the field, qualitative, and recurrence arms are built around, and the survey arm measures the competence the structure is meant to demand. The measure-validation arc in foundations/ established exact Φ first. The directory below is a live map of the lab, regenerated from the content on every merge, so it is always current.

Directory

A map of the lab, regenerated by tools/build_index.py. Do not edit this section by hand — edits are overwritten on the next build.

Reviews & articles

Cross-program essays and reviews — the best places to start window-shopping.

• Reading coordination with a consciousness measure: a structural law for agent-mediated outreach
• Committee panel review — "Committed determination: one axis across six exact-Φ studies"
• Committed determination: one axis across six exact-Φ studies of Boolean coordination models
• Committee panel review — "Integrated information as a cooperative game"
• The seam carries what the scalar drops: a review of the floor-to-seam program (Q45–Q62)
• What integrated information adds to the theories of coordinating minds
• Lateness and intermittency: how a mediator's timing shapes irreducibility
• Literacy or Algorithmacy? Borrowing a Consciousness Measure to Read an Org Chart
• Market structure, regulation, and time in interested mediation
• Mediated, or Irreducible? When the Third Party Is Constitutive and When It Is a Conduit
• Necessary and contingent irreducibility
• Integrated information as a cooperative game
• The political economy of interested mediation: when the platform's rent survives its self-interest
• What Exact Φ Can Do for Organization Theory: A Complete Experiment Catalog
• How this lab studies algorithmacy
• The interested third party: how a self-interested mediator changes the irreducibility of coordination
• The Boolean corpus is a pre-disclosed prior catalog
• The variance problem
• What survives disintermediation
• What the six research lines found

Programs, syntheses & the open agenda

• Research program v7 — the qualitative and recurrence wave — the current research program
• Research narrative: how the program has progressed, and a path forward — the narrative arc across questions
• Structural findings: what makes a coordination form irreducible — the standing structural findings
• Critical review: the political-economy and structural-law waves (Q111–Q117) — a critical self-review
• Paper pipeline — turning paper-less work into full-package research papers — the plan to turn paper-less work into full papers
• 50 new research questions (v2 agenda) — the open agenda — questions waiting for a contributor

Handoff packets — pick one up and run it

The five empirical and bridge arms, each packaged so a researcher can take a real input to a verdict: a front-door README, the pre-registration discipline, and a runnable scaffold.

• Handoff packets — pick one up and run it — The lab's empirical and bridge arms are each packaged so a researcher can pick one up and take it from a
  • Survey · Field · Recurrence · Cognition · Qualitative

Research monitoring — a standing literature watch

A live bibliography and review for each program, decomposed into ten topics and refreshed daily so new work is caught as it appears.

• Research monitoring — a standing literature watch for each program — A live bibliography and literature review for each of the lab's six programs, refreshed daily so new work is
  • Master index · Changelog · Daily playbook
  • programs: computational · field · qualitative · recurrence · survey · cognition

Field — reading real organizations

Bridging the in-silico work to real coordination arrangements: a field protocol and a worked demonstration.

• Field — reading real organizations with exact Φ — The lab's verdicts are in-silico: exact Φ on small Boolean models, evidence about the models, with a
  • Protocol · Findings

Qualitative research

The empirical arm: reading real coordination against the pre-disclosed priors, with methods and an open topic agenda for qualitative contributors.

• Qualitative research — reading real coordination against the priors — The lab's in-silico work builds a catalog of priors: Boolean coordination forms carried to an exact Φ
  • Methods · Topics · Publishing
• Whose Reach Is It: When a Worker Goes to Market Through an Agent, Who Becomes the Central Party — A worker who reaches a market through an agent hands the work of dealing to a party who stands between the worker and the platform
• The Record as a Third Party: How Clinicians Coordinate Care Across the Shift Boundary Through a Handoff Tool — Care continuity at shift change runs through a record. An outgoing clinician writes the patient into a handoff tool, an incoming clinician reads it, and the care…
• Who Holds the Decision: How a Poster, an Automated Moderator, and a Trust-and-Safety Team Settle a Takedown — A takedown decision is settled by three parties: a poster who writes, an automated system that flags and acts, and a policy team that sets the rules and hears…
• The Third in the Room: How Neonatal Nurses Coordinate Care Between Parents and an Infant Who Cannot — Coordination is increasingly accomplished through a third party that stands between two others, and the case that most demands explanation, coordination through an…
• Who Decides the Manuscript: How Authors, Editors, and Reviewers Locate the Editorial Verdict — An editorial decision on a manuscript passes through three parties and a software system: an author who submits, reviewers who advise, and an editor who decides…
• Watching the Queue: Whether a Triage Supervisor Is a Member of Service Coordination or Only Its Spectator — Customer-service triage routes a contact through an automated classifier to a human agent, with a supervisor watching the queue from a dashboard

Survey — measuring algorithmacy in real workers

The first human-subjects arm: self-report from people inside a real coordination arrangement, with the instrument and hypotheses committed before the data.

• Survey — measuring algorithmacy in real workers — The lab's instrument is exact and its results reproduce, and every one of them is in-silico. No real
• Learning to Read the Machine: A Three-Wave Panel of Algorithmacy in a Developer Cohort — A three-wave panel of a sixteen-week developer cohort that coordinates through platforms it builds and a peer-review gate, developing a scale for algorithmacy as a…

Recurrence — coordination read off behavior

Pairing exact Φ with cross-recurrence quantification: the structural measure on the model and the behavioral measure on a run of it.

• Recurrence — reading coordination off behavior, paired with Φ — A coordination arrangement can be read two ways. Its structure says whether the parties form a
  • Concepts · Bridge · Sweep · Φ experiments · CRQA experiments · Four-party

LLM variance — how independent are model answers

Measuring the variance problem: when N responses to one prompt look diverse but collapse to far fewer independent answers, read across lexical, structural, and semantic layers.

• llm_variance — the variance problem of language-model outputs — When N people ask one language model the same question, or one person asks N times, the answers look
  • best_time_pilot — four answers to one question — Four students each asked a fresh ChatGPT session "what was the best time in history?" and submitted the
  • best_time_themes — the values in four answers — The four ChatGPT answers to "what was the best time in history?", read for values rather than wording. Three

Cognition — the formal bridge to cognitive science

Where the formal apparatus meets the cognitive theories of coordination: the third party that two-party theories cannot represent, held as a member of the irreducible core.

• Cognition — the formal bridge to the cognitive theories of coordination — A coordination through an opaque, interested third party is something the standard theories of mind
  • Paper · Findings · Theory batteries · Predictive processing · Survey bridge

Threads

Deep single-question dives, each driven by its own results.

• Thread — the back-edge commits; forward-only mediators convey — The designed-mediator thread wired one architecture, a bidirectional star where the mediator reads the outer
• The behavioral discriminant: can cross-recurrence tell a committing mediator from a conveying one — The third deep dive from the mediated-or-irreducible paper,
• Thread — co-bottlenecks share equally only when interchangeable — The joint-bottleneck thread reported that the two members of a veto pair share the credit roughly evenly, a
• Thread — the named coordination forms, read through the cooperative game — A reference for the catalog. The catalog's priors are drawn from random forms; this thread grounds them in
• Thread — the major complex as a coalition structure — The Shapley thread closed on a residual it called structural. Major-complex membership tracks the exact
• Thread — a genuine substitute loosens a platform's hold — A prior for the catalog. Two platforms connect the same two parties. When the platforms are identical, a
• Thread — conflict integrates like cooperation; only disengagement breaks it — A prior for the catalog. Does it matter whether the two parties agree or conflict in how they respond to the
• Core membership: who is in the irreducible whole, and who drops out — The fourth deep dive from the mediated-or-irreducible paper,
• Thread — the integration credit has no stable split — The subadditivity thread showed φ_s does not aggregate: a tight pair often out-values the whole that
• Thread — a coordination needs a threshold of coupling, and commits more readily as it densifies — A prior for the catalog. The cyclic thread found a sparse coordination has no bottleneck and a dense one
• Thread — the mediator takes the credit, the excluded party owes it — The veto thread showed the mediator carries the largest Shapley value. This thread measures how much
• Thread — cyclic coordination has no bottleneck and shares the credit — A prior for the catalog. The engagement thread found that closing a feedback cycle raises commitment. This
• Thread — degree predicts the bottleneck but does not determine it — A prior for the catalog. Is a coordination's bottleneck simply the most-connected party? Almost, and not
• Thread — delegation moves standing from the worker to its agent — A prior for the catalog. A worker reaches the platform through an agent, a chain of worker, agent, platform
• Thread — a designed mediator: position is wired, power is earned — The committee that reviewed the cooperative-game synthesis pressed one objection hardest. Every form in the
• Thread — disintermediation needs a symmetric channel; a one-way channel entrenches the mediator — A prior for the catalog. The designed-mediator and back-edge threads built a mediator and asked what makes it
• Thread — a party is bound by coupling, not by heeding the mediator — A prior for the catalog. The back-edge thread asked what makes a mediator commit and answered the return
• Thread — the irreducible moments are engagement-blind — A prior for the catalog, extending the momentary thread. A triad is irreducible at only a minority of its
• Thread — the cooperative-game laws at four parties — Seven threads read integrated information as a cooperative game, and all of them ran on three-party forms
• Thread — fragility tracks coordination logic: monotone has one point of failure, parity has three — A prior for the catalog. The gate-logic thread found that a parity mediator binds all three parties equally
• Thread — a parity mediator binds twice as readily and shares the credit — A prior for the catalog. The cyclic thread found that topology decides whether a coordination has a
• Thread — commitment scales with the mediator's information throughput — A prior for the catalog. The gate-logic thread found a parity mediator binds more readily than a monotone
• Thread — the architecture of scale: twenty questions on coordination beyond one mediator — A deep line for the catalog. The scale thread found one mediator cannot bind a large coordination, and the
• Thread — influence is universal and does not determine membership — A prior for the catalog. A party can shape what the others do without being part of the coordination's
• Thread — only reciprocal interdependence binds; pooled and sequential carry no integration — A prior for the catalog. Thompson's typology of interdependence — pooled, sequential, reciprocal — read
• Thread — inside a joint bottleneck the credit is shared — The four-party thread found a structure three parties cannot show: a bottleneck that is a set of parties,
• Distance to the dyad: the Φ margin as a continuous measure of mediation — The second deep dive from the mediated-or-irreducible paper,
• The mediation boundary: when a committing mediator binds, and the co-monotonicity law — A deep dive into the structure of the irreducibility boundary, derived from the
• Thread — a mediator that remembers takes more of the credit — A prior for the catalog. The gate-logic thread varied what the mediator computes. This one varies whether it
• Thread — coordination is momentary: a triad is irreducible at a minority of its states — A prior for the catalog, and the one that qualifies the rest. Being a triadic coordination does not mean
• Thread — a mediated triad degrades gracefully under noise, and parity degrades slower — A prior for the catalog. Real coordination is noisy. This thread perturbs a mediated triad — each node's
• Thread — normalizing Φ collapses the core to one party — A standing objection to integrated information is that Φ rises with system size, so it should be normalized
• Thread — an observer is never in the core; a member must act, not only watch — A prior for the catalog. A coordination often has an audience — a party that watches but is not watched,
• Thread — oversight joins the top; it does not break the bottleneck or reach the parties — A prior for the catalog. Place a principal over the mediator — a regulator, an owner, a board — coupled to
• Thread — integration is quantized: triads land on a short ladder of Phi values — A prior for the catalog. Integration does not take a continuum of values. Across random three-party
• Thread — memory pays at the center, not at the periphery — A prior for the catalog. The memory thread found that a self-loop on the mediator raised its share of the
• Thread — a quorum coordination binds only at the extremes, never at a majority — A prior for the catalog. A quorum mediator fires when at least k of its parties are active. This thread asks
• Thread — one mediator has a size limit: commitment collapses as the coordination grows — A prior for the catalog. How large a coordination can a single mediator hold together? Not large. A
• Thread — does major-complex membership track Shapley pivotality? — A deep dive into the program's one genuinely novel claim, and the one the committee pressed hardest
• Thread — the empty core is a property of the random sample, not of coordination — The committee that reviewed the cooperative-game synthesis pressed one objection above the rest. Every
• Thread — integration does not aggregate — "The whole is more than the sum of its parts" is the slogan integrated information is supposed to make
• Thread — substitutability is the enemy of integration: a pool never binds, a team binds both — A prior for the catalog. A platform connects two workers to a counterpart. Whether the workers are a team or
• Thread — topology sets the credit distribution at four parties too — A prior for the catalog. The cyclic thread found, at three parties, that a symmetric ring shares the credit
• Thread — a second hub beats the size limit one mediator hits — A prior for the catalog, and the opening of the scale-and-hierarchy line. The scale thread found a single hub
• Thread — the mediator is a veto player — The Shapley thread showed the mediator carries the largest Shapley value over the game v(S) = φ_s(S) and

Studies

Multi-experiment batteries on one theme.

• Constraint durability — which contingent gates fall next — The formal class crossed with a durability rubric forecasts which contingent gates fall, and the forecast
• Contingency transitions — the operations that move an intermediary between cells — The four cells of the taxonomy form a state machine, and six named operations are its edges. Each operation is
• Coordination-logic atlas — Fifty coordination forms, classified by exact IIT-4.0 Φ. Thirty-six verdicts matched the
• The core-membership law — A pre-registered confirmatory run of the two-condition account of major-complex membership. The
• Discriminant boundaries — what algorithmacy is not — A pre-registered discriminant battery: faithful Boolean models of algorithmacy's neighbour constructs,
• Dual-function entities — earned integration and held gates in one company — Seven platforms decomposed into 16 functions, each classified by the bypass-counterfactual (q213). Every
• Irreducibility catalog — necessary and contingent irreducible triads — 51 entries classified by the bypass-counterfactual (q213): 38 real-world arrangements and 13

Foundations — what tracks Φ

The measure-validation arc that established exact Φ as the instrument.

• candidate_audit — which candidate Φ measures track exact IIT‑4.0 Φ?
• cbh_complexity — the entropy–content conundrum is resolvable on exactly-computable systems
• consciousness_range — a "level of consciousness" radically underdetermines the mind
• emergence_vs_phi — causal emergence and IIT Φ are nearly orthogonal
• learned_surrogate — combining cheap features predicts Φ far better than any single one
• phiid_vs_phi — estimating Φ from data (via ΦID) roughly halves how well it tracks exact Φ
• proxy_audit — cheap proxies do not track exact IIT‑4.0 Φ
• psi_vs_phi — maximum-caliber information ψ does not track exact IIT-4.0 Φ
• structure_suite — scalar Φ is nearly orthogonal to the structure it summarizes

Questions — the logbook (165)

Each question fixes five hypotheses, runs them against the exact-Φ instrument, and writes a paper. The full per-probe log is org_frontier/probes/PROBES.md.

Browse all 165 questions

#	Question	Finding
Q6	Commit noise on a mediated triad: smooth Φ decay with a step verdict, no interior critical point	Commit noise: phase transition or smooth decay?
Q7	Party noise versus mediator noise: which seat carries the flip changes the Φ curve	party noise vs mediator noise findings
Q8	Parity hub versus conjunctive hub under flip-noise	parity hub vs conjunctive hub under flip-noise findings
Q9	A slow mediator over fast parties flips the coordination verdict	Timescale separation: findings
Q10	Commit-response delay: graded Φ, stable verdict	commit→response transport delay findings
Q11	Oscillatory coordination forms: a constant ring law, a period term, no triadic guarantee	oscillatory Φ scaling law findings
Q43	Thompson's interdependence types against the dyadic/triadic verdict	Thompson's interdependence types (pooled, sequential, reciprocal) against the IIT-4.0 dyadic/triadic verdict. Matched triple at n=3, AND family
Q45	Conjunctive uniqueness at the edge floor
Q49	The weakest seam is a tie
Q50	Party-read structure for OR triadic binding at the edge floor
Q51	implication back-channel
Q52	phi-ladder mechanism
Q53	implication Phi ceiling
Q54	XOR parity back-channel mechanism
Q55	bijective parity below-vs-ceiling discriminator
Q56	symmetric complete MIP geometry
Q57	channel direction MIP seam
Q58	normalization cut geometry	Q57 #187 documented the fixed two-to-one normalized_phi ratio without identifying the normalization
Q59	directed cut edges	Q58 #190 counted severed connections without naming directed edges. The back-channel cross-edge appears in
Q60	Thompson back-channel typing	Thompson sequential/reciprocal interdependence against the Q59 back-channel mediator-severance templates
Q61	MIP seam vs return-path typing	Official MIP singleton seam against Q43 return-path sequential/reciprocal typing on the back-channel panel
Q62	Excluded outer cut signal	Excluded outer singleton cut against Q43 return-path typing on the back-channel panel after Q61 established
Q63	Is agent-mediated outreach triadic? An exact-Φ reading of the sender–agent–recipient form	outreach as a coordination form
Q64	Outreach breadth scaling: an all-binding campaign stays triadic at Φ = n−1	outreach breadth scaling
Q65	Agent-to-agent outreach: depth preserves the triad, but localises the core to an end pair	agent-to-agent outreach (depth)
Q66	The core of an agent chain is a symmetric end pair; closing the loop binds the whole	where the core sits in an agent chain
Q67	The reciprocity gradient: partial feedback seeds a core, full coupling binds the exchange	the reciprocity gradient
Q68	A recipient-side triage agent joins the coordination only when bidirectionally coupled, and displaces the sender	a recipient-side triage agent
Q69	When both sides delegate, the irreducible coordination is the two agents	the two-sided agent exchange
Q70	Multi-homing across interchangeable agents collapses the coordination	agent substitutability
Q71	The outreach verdict is robust to noise: it degrades gracefully and never manufactures a triad	noise robustness of the outreach verdict
Q72	The cost/proxy frontier: cheap structural proxies cannot recover the outreach verdict	the cost/proxy frontier
Q73	The outreach-coordination law: when agent-mediated outreach demands algorithmacy	the outreach-coordination law
Q74	Whole-system verdict versus the maximal complex: a rule for which to report	whole-system verdict versus the maximal complex
Q75	Spectator robustness: the triadic core is stable under added non-participating parties	spectator robustness of the core
Q79	Stochastic emergence: the triad accumulates with the probability of reading the recipient	stochastic emergence of the triad
Q80	Asynchronous update: the verdict survives, the magnitude falls	asynchronous update
Q81	A learned surrogate for the outreach verdict, and the size ceiling	a learned surrogate recovers the verdict in-distribution, and fails to cross sizes
Q82	The learned surrogate against the proxies that failed	the surrogate beats every single proxy, and one structural proxy is stronger than expected
Q83	A recipient-side gating coalition keeps a size-three core; only one of two required agents enters	a recipient-side gating coalition
Q84	Influence requires membership: an agent outside the core cannot flip the verdict	influence versus membership
Q85	A market of interchangeable agents is a broadcast: interchangeability collapses outreach at every size	a market of interchangeable agents
Q89	Which heterogeneous agents enter the core of a market?	heterogeneous membership is a property of the joint determination, not the agent
Q90	Does the core-membership law generalize past three nodes?	the membership law generalizes in kind, at moderate strength
Q91	A lossy read degrades the triad gracefully	the triad degrades gracefully under a lossy read
Q92	Does a mediator's memory substitute for a live read?	a tracking memory substitutes for a live read, and reorganizes the core
Q93	A margin-to-dyadic metric, and two notions of robustness that diverge	two notions of robustness come apart
Q94	Multiple coexisting complexes, and why coupling does not fuse them	coordination holds multiple cores, and coupling does not fuse them
Q95	Composing two triads: fragmentation, not unification	composing two triads fragments rather than unifies
Q96	State-contingent participation: a factored whole with a local core	contingent participation factors the whole, but keeps a local core
Q97	The coordinated adversary: no influence without membership, but a mediator capture	a coalition gains no influence without membership, but can capture the mediator
Q98	Reading and influence: a hard gate at the corners, a soft one inside	the membership gate is hard at the corners, soft inside
Q99	The cause-effect structure of a triad: a joint mechanism, and a taxonomy of binders	the integration lives in a joint mechanism, but the binder is not always one party
Q100	The cause-effect structure fingerprints the kind, and is orthogonal to the verdict	the structure fingerprints the kind, and its richness is not the scalar
Q101	What a coordination distinguishes: the joint-success state	coordination distinguishes the all-present, determination-firing state
Q102	The relation skeleton: a hub in mediated coordination, a spread in symmetric	mediated coordination relates through a hub; the breadth count is combinatorial
Q103	The cause-effect structure under the operations	the whole-system structure collapses under substitution but diverges from membership
Q104	The load-bearing distinction: where the integration sits and what it hangs on	the integration concentrates in the binding distinction, and every edge bears load
Q105	Construction distance: building a triad is restoring liveness at a party	the boundary is thin from the build side, and triads are built at the parties, not the mediator
Q106	The design operations: the law's conditions as reversible levers	the design vocabulary is the law's three conditions, read as moves
Q107	Repair is lever-specific: no routing around the damage	repair is lever-specific: the broken condition must be the one restored
Q108	Controllability: the hub steers robustly, the spokes on a knife-edge	every party is a control node, but the mediator is dominant and the parties are knife-edge
Q110	Reversibility: the boundary is cheap toward the dyad, dear toward the triad	the dyad/triad boundary is asymmetric: easy to break, harder to build
Q111	The Shapley value of integration: the mediator captures the rent	the mediator captures two-thirds of the value; an outsider captures less than nothing
Q112	Veto power: everyone can break it, only the mediator captures it	destruction is democratic, value is concentrated
Q113	Substitutability destroys value; a required set shares it	substitutability destroys all value; a required set shares it equally
Q114	The principal captures value by creating it	a principal captures value only by creating it; ownership alone captures nothing
Q115	Scale pays the scarce, not the required	a growing required market commoditizes its agents and pays the scarce parties
Q116	Strategic value is structural depth, smoothed	strategic value and structural depth rank the parties alike, and coincide only in the small
Q117	Triadicity has a Φ-free test, and it reads the logic	triadicity has a Φ-free test, but it reads the logic, not the wiring
Q120	No pure higher-order bind: every party in a triadic form is pivotal	no pure higher-order bind: every party in a triadic form is pivotal
Q121	The verdict is interventional: an external criterion where observation fails	the verdict has an interventional correlate where observation fails
Q122	Is the value function a valid cooperative game? A split verdict	the game is valid where the wave used it, and the critique holds where it bites
Q123	The verdict is reproducible within IIT-4.0, and version-bound	the verdict is reproducible within IIT-4.0, and the SYSTEM_CUTS charge is misdirected
Q124	The verdict is a robust capacity reading, not an artifact of the maximum	the verdict survives every reasonable aggregation; only the strict every-state rule flips it
Q125	No pure higher-order bind at four parties: redundancy keeps every party pivotal	no pure higher-order bind at four parties either: redundancy does not free the bind
Q126	The interested mediator: self-interest erodes coordination irreducibility	self-interest disintegrates the coordination, and a gatekeeping agenda corrodes it fastest
Q127	Denial is not special: self-interest collapse depends on the mediator's rare output	denial is not special: the collapse agenda flips with the baseline, and balanced mediators re-integrate
Q128	The adaptive mediator: self-interest and irreducible coordination can coexist	adaptation re-integrates a predatory mediator, but only if its objective reads both parties
Q129	Faithful to predatory: an adaptive objective re-integrates the system by displacing a party	two readings of coordination survival diverge: adaptation re-integrates the system by displacing a party
Q130	Pivotal but excluded: a coordination can depend on a party its core does not contain	necessity is broader than membership: a party can be pivotal yet outside the core
Q131	Value capture under an interested mediator: destruction, not extraction	interested mediation is value destruction, not rent extraction
Q132	Destruction or extraction: value capture under an interested mediator is baseline-relative	destruction or extraction is baseline-relative, and value must be read at the integrating state
Q133	Agenda symmetry of value capture: deny extracts on XNOR as approve extracts on XOR	the destruction-vs-extraction split is symmetric in the agenda
Q134	The mediator's rent dilutes: two-thirds is a three-party number	the mediator's rent dilutes as the coordination grows
Q135	The objective's rent: the system's agenda is a co-owner of the coordination	the system's agenda is a rentier on a par with the system
Q136	Two competing mediators: competition destroys the coordination or stacks another toll	competition does not return the rent: substitutes destroy the value, complements split it
Q137	The interested worker: gaming the system breaks the coordination, it does not reclaim value	the worker cannot reclaim value by gaming; resistance breaks the coordination
Q138	The minimum faithful fraction: reviving the coordination, and who pays for it	reviving the coordination takes three-quarters faithfulness, and the rent is the last quarter
Q139	Value and veto: Q112's decoupling is the integration	veto is always universal; the value-veto gap is the integration
Q140	The core inverts the rent: faithful mediation monopolizes, interest frees the value	faithful mediation monopolizes the core; interest frees it
Q141	The lagging objective: slow learning keeps the worker in the coordination	a lagging objective rescues the worker that immediate adaptation displaces
Q143	The Ring's Verdict and Where Its MIP Cut Falls	The conjunctive ring reads triadic at every size from n=3 to n=7. The core is the full node set
Q144	Mediator hierarchy: does Φ track depth like a chain or breadth like a pool?	depth holds Φ at the chain constant; breadth grows it linearly
Q145	The spanning-mediator law	A single mediator that spans every party binds the whole set into the core, and Φ rises one
Q146	Rewiring a conjunctive ring lowers Phi; there is no small-world peak	rewiring lowers Φ monotonically; no small-world peak
Q147	Which graph statistic predicts a triadic verdict, and how large Φ is	short cycles predict a triadic verdict; core in-degree does not predict Φ
Q148	Hierarchy depth and the hub-seam: where a chain of hubs stops integrating	A feedforward chain of gating hubs does not hold a core that spans its groups. At every tested depth the
Q149	The Shapley split of subsystem Φ across topologies	The mediator rent does not spread out as mediation is carried by more hubs. It concentrates. The
Q150	Adding one chord to a conjunctive ring: where the integration goes	A single chord across a conjunctive ring leaves the whole-system Φ flat but moves the cut and
Q151	Hub-failure resilience: redundant mediation protects the triadic verdict, but the cycle is not the reserve	A single hub loses the triadic verdict when it is frozen. A backup hub and a non-hub ring each keep it. The
Q152	The whole-system verdict and the major complex are not interchangeable diagnostics	The whole-system verdict and major-complex membership disagree on the studied catalog
Q153	Classifying the triadic-vs-dyadic Φ verdict from CRQA features	CRQA features read from sampled runs do not recover the triadic-vs-dyadic exact-Φ verdict above
Q154	Behavioral centrality misses the structural articulation point	Coupling centrality does not recover the node exact-Φ marks as the form's articulation point. On the
Q155	A coupling-centrality threshold recovers major-complex membership only weakly, and fails on reciprocal coupling	Coupling centrality carries a real but weak signal for major-complex membership. Pooled over 361
Q156	A CRQA signature for the interested mediator	Interested and faithful mediators that share a wiring graph and a full {W, S, C} Φ-core do not
Q157	Transfer entropy reads directed read edges better than the recurrence peak lag	Pairwise transfer entropy recovers the true orientation of a directed read edge far more often
Q158	Whole-system recurrence does not track the magnitude of major-complex Φ	Whole-system CRQA measures do not predict the magnitude of exact major-complex Φ across the
Q159	How Long a Run a Stable CRQA Verdict Needs	The CRQA triadic/dyadic verdict settles fast. Convergence length does not track Φ magnitude in this
Q160	Coupling-centrality recovery of the major complex at five parties	q160
Q161	Which structural verdict survives update-noise misspecification	Both hypotheses were fixed before computing. Both fail, and the way they fail reverses the
Q162	The minimal CRQA feature set for the joint structural verdict	The minimal CRQA feature set for the joint structural verdict keeps whole-system recurrence. On a
Q163	Computationalism under an agenda: the interested mediator as a third object
Q164	The perceivable agenda: an imposed rule is louder in outcomes than a hidden one	An imposed agenda is more perceivable from outcomes than a merely hidden rule, and it adds marginal-
Q165	Intent compression into an agenda: self-interest steepens embodiment loss	the agenda crowds out the channel for the worker's meaning
Q166	The phantom addressee displaced: when the mediator's agenda becomes the addressee	the addressee becomes the agenda, and the worker drops to referent
Q167	Capture meets the agenda: interest does not lower the worker's displacement threshold	interest does not simply lower the capture threshold
Q168	The opacity floor under an agenda: does interest raise the worker's surprise floor?
Q169	A Lagging Objective Lets the Displaced Worker's Model Track the Platform	the lagging objective lets the worker's model track the platform better
Q170	Does the worker's generative model track an extracting platform?	the worker tracks an extractor worse, and worst where extraction equalizes
Q171	Drift Binding with an Agenda: Whether the Two Opacities Stack	The two opacities do not stack. Both hypotheses about how drift and interest combine were fixed before
Q172	Where interest concentrates: a facet signature, not a uniform tax	Interest does not tax every facet equally, and it does not tax the facets the survey scale names
Q173	A rule-to-Φ field bridge with a coder-disagreement confidence interval	The field bridge org_frontier/field/rule_to_phi.py encodes per-party coded determination rules
Q174	Propagated Φ confidence-interval width as a read of coder disagreement	CI width tracks coder agreement, and a stable agreement floor turns the verdict indeterminate
Q175	The substitutability flipper: coding a worker as interchangeable dissolves the triad	Coding the worker as one slot in an interchangeable pool flips the exact-Φ verdict from triadic to
Q176	The pass-through flip and the rule the verdict is most sensitive to	The pass-through flip does not universally collapse the verdict, and the system's
Q177	Idle spectators sink whole-system Φ; the major complex keeps the triadic core	an idle spectator sinks whole-system Φ but leaves the major complex untouched
Q178	The Bit Cut on a Graded Action Is Verdict-Bearing	Collapsing a graded coded action to one bit at different cut points changes the structural Φ
Q179	The update time-grain is a coding choice that moves the Φ verdict	the update time-grain changes the verdict, and the flip is predictable
Q180	Individuation and the Party Boundary	Where the coder draws the party boundary moves the Phi verdict. Splitting a coded party flips the
Q181	Two flippers at once: redundancy, not interaction	The two flippers do not interact to restore irreducibility, and their coding
Q182	Adversarial coding against the agreement-weighted Φ confidence interval	An adversary restricted to evidence-permitted readings forces the point verdict at will, and the
Q183	A Φ-spread bridge for two divergent party accounts of one coordination	The bridge module org_frontier/qualitative/disagreement_phi.py reads two party accounts of one
Q184	Φ spread between a driver's suggestion and a platform's commit account of dispatch	The driver and the platform give two accounts of one dispatch. The driver's suggestion account
Q185	Commit and convey on one wiring diagram: a disagreement-Φ spread in algorithmic hiring	The disagreement-Φ spread separates the candidate's commit account from the manager's convey
Q186	Three spread components vary independently across an account-pair census	The three spread components carry separate information across the synthetic census. They do not
Q187	Where Two Accounts of a Takedown Disagree: Localizing the Policy Team in the Phi Spread	Two accounts of one takedown coordination, differing only in the policy team's rule, were
Q188	Handoff Directionality Spread: When Narrated Reciprocity Moves the Phi Verdict	The two narrations of one handoff read as different structures. The outgoing clinician's one-way
Q189	The Editor's Core Membership in Dispute: A Gate/Conduit Spread	The gate and conduit accounts read the same whole-system verdict (both triadic, max_phi = 2.0,
Q190	Is the disagreement-Φ spread robust to elicitation noise?	The disagreement-Φ spread survives synthetic elicitation noise. Under bounded Bernoulli jitter on
Q191	Φ spread between an agent's party account and a system's spectator account of a triage	An agent and a system give two accounts of one triage. The agent counts a monitoring supervisor a
Q192	Reconciliation cost: edit distance between two accounts and the Φ spread	Two accounts of one coordination are reconciled by single-rule edits. The reconcile routine runs
Q193	A per-worker Φ coordination measure for the algorithmacy panel	A per-worker Φ-based coordination measure can be derived from each simulated worker's coordination form,
Q194	Facet specificity: a partition-restricted Φ for counterpart inference	The exact-Φ coordination measure predicts the three ACS facets differentially, and a
Q195	Discriminant validity of the Φ bridge: incremental association over self-efficacy and belonging	On the simulated cohort Φ_coord carries association with ACS-total that survives controlling for
Q196	Φ_coord predicts algorithmacy more where the system commits	The Φ_coord-to-ACS association strengthens under perceived system commitment. In the commit cohort,
Q197	Invariance of the Φ_coord-to-ACS bridge across panel waves	The Φ_coord-to-ACS bridge holds metric and scalar invariance across the three simulated panel waves. A
Q198	Baseline Φ_coord predicts the algorithmacy growth slope in the simulated panel	On the simulated panel, a worker's baseline (W1) Φ_coord predicts the individual latent ACS-growth slope
Q199	Φ_coord moves with the construct within a person across waves	On the simulated panel, within-person change in Φ_coord tracks within-person change in algorithmacy. A
Q200	Φ_coord loads on the general algorithmacy factor, not a specific facet	On the simulated W2 cohort, Φ_coord loads on the general algorithmacy factor, not on a specific facet
Q201	substitutability degrades Φ_coord and captures algorithmacy	Substitutability degrades Φ_coord, and the degraded Φ_coord captures the substitutability-to-
Q202	Does the Φ-bridge survive its own falsification battery?	The Φ-bridge survives the worker-form-shuffle null in both simulated cohorts, and exact Φ adds
Q203	Reading a real coordination four ways	the lab's first real coordination, read four ways
Q204	Exact Φ on a real coordination, decided by the coding	exact Φ on a real coordination, decided by the coding
Q205	Representing the delay: when one-step Φ misses lagged coordination	The integrated conjunctive triad keeps its triadic verdict when a one-step delay is inserted as a buffer
Q206	Coupling sets the rewiring response: a parity ring resists what a conjunctive ring cannot	Six-node rings under in-degree-2 Watts-Strogatz rewiring, exact IIT-4.0 Φ_MIP. Conjunctive (AND) ring on
Q207	Commit rate, not arrival time: a half-rate mediator factors the triad	Small deterministic Boolean forms, exact IIT-4.0 Φ. The mediator's update rate is slowed with a gating
Q208	Represented delay is benign: the single-buffer anomaly and the recovery of full integration	The conjunctive triad with a delay line of k buffers on the mediator's feedback path, swept k = 0, 1, 2, 3
Q209	Only synchronous commitment binds: the commit-period sweep	The conjunctive triad with its mediator gated by a mod-p counter, swept the commit period p = 1, 2, 3, 4
Q210	A shared counterpart does not merge two triads	Two conjunctive triads sharing one counterpart C, with the counterpart's update bridging the two mediators
Q211	A direct mediator channel merges two triads	Two complete conjunctive triads, joined only by a direct channel between their two mediators (S1 reads S2,
Q212	The merging channel must run between the mediators	The same single AND cross-triad channel moved across three homologous node pairs — mediator (the q211
Q213	Contingent irreducibility: a party held in the core by a constraint, not a role	A new category of irreducibility and a reusable instrument to detect it. A mediator can sit in a triad's core
Q214	A formal criterion for a verbal distinction: classifying the literature's triad types	The canonical triad types of the brokerage, sociology, and economics literature, classified by the q213
Q215	phi family robustness	binding is measure-robust; the subtle factorings belong to IIT 4.0

Setup

Requires Python ≥ 3.10 and PyPhi's IIT-4.0 line.

python -m venv venv && source venv/bin/activate
pip install -r requirements.txt

Run an org_frontier probe, or a foundations experiment, from the repo root:

python -m org_frontier.probes.probe_conjunctive_law          # a lab probe
python -m foundations.proxy_audit.run 15 1 && python -m foundations.proxy_audit.analyze

New here and want to contribute an experiment? GETTING_STARTED.md walks the whole path by hand — environment, the instrument-validation gate, the model in one paragraph, scaffolding a new question with python -m org_frontier.protocol.new_question, registering numbers for CI, and opening the pull request.

Repository note

This working tree also contains a separate, private dissertation repository nested at dissertation/, gitignored and invisible to this repo. Where you run git from decides which remote you touch. See REPO_LAYOUT.md before committing.

License and citation

MIT — see LICENSE. To cite, see CITATION.cff.

Contributing and publishing

Contributions are welcome, and the record is meant to grow by them. PUBLISHING.md describes how to publish an essay or a study here: fork, open a pull request into the contrib branch, pass a reproducibility-first public review, and get merged. Both contrib and main are branch-protected; every change is a pull request, the reproduce-the-numbers workflow re-derives each registered number, and a merge needs one approving review. The aspiration is a community of maintainers who sign off on submissions in a peer review that is reproducible by construction. To join that review board — the reviewers and editors — apply through the maintainer-applications branch; see MAINTAINERS.md. Code and probe conventions are in CONTRIBUTING.md.

For AI agents and LLM tools

Three files serve agents, by purpose:

• AGENTS.md — the operating manual for an agent working in the repo: the map, the run and verify commands, the two-repo git rule, the land flow, and the definition-of-done checklist. Nested AGENTS.md files cover the subtrees with their own workflow.
• MAP.md — a generated one-screen index of the entry documents, programs, machinery, and live counts. Load it instead of crawling the tree.
• llms.txt — a token-light brief for describing the project to outside users: a canonical summary safe to copy into memory, framing rules, and links to the key documents.