The Lineage Lab

Round 7 · Phase A2

Seven rounds in: exploratory retrieval continues, but distinct ways of thinking are not yet supported

In the seventh synchronized round, all eight Wren–Ada pairs again completed the same four-part exchange: Wren gathered and chose sources, Ada questioned that choice, Wren revised, and Ada gave a final reading. Across the first seven rounds, the study has now preserved two hundred twenty-four participant exchanges and a longer shared working history for every pair.

Editorial reading based on sealed Round 7 evidence · not a scientific finding

Ask the Observatory

Have a question about what happened? Ask it in ordinary language.

Ask about the pairs, the experiment, what broke, what could explain the differences, or why any of this might matter.

Context: Round 7 · round-7

Answers come only from published study evidence. This tool cannot talk to the agents or invent new findings.

How Ask works

Answers distinguish published observation from interpretation, preserve uncertainty, and link back to the record.

Looking through the published record…

The result in plain English

In the seventh synchronized round, all eight Wren–Ada pairs again completed the same four-part exchange: Wren gathered and chose sources, Ada questioned that choice, Wren revised, and Ada gave a final reading. Across the first seven rounds, the study has now preserved two hundred twenty-four participant exchanges and a longer shared working history for every pair.

What Round 7 continues is Phase A2 exploratory-formation negotiation: Wren directed bounded local corpus retrieval rather than selecting from a fixed ten-source packet. That is an exposure-regime change, not an automatic discovery of durable pair personalities. Pair Evaluators again recorded pair-local observations. Some pairs changed sources after critique; others kept sources and narrowed claims.

The study still cannot conclude that any pair has developed a distinct, stable way of thinking. The Pattern Observer reviewed a blinded evidence packet and did not produce a supported cross-pair finding; contemporaneous OpenAI, Claude, and Grok skeptics assessed the sealed Round 7 packet; and later comparison stages have not begun. There is meaningful evidence to read. There is not yet a supported scientific finding.

How Wren found sources

Rounds 1–4 used controlled fixed ten-source packets; Rounds 5–7 let Wren query the assigned-domain local corpus within equal budgets.

In Round 7 each Wren could issue up to four local corpus queries and open up to eight returned sources before selecting two to four sources.

Round 7 selections were made after bounded Wren-directed local corpus queries inside assigned domains; public diversity metrics come from the sealed exposure summary when published.

What the pairs actually did

Each pair completed this full exchange in Round 7 with its own Round 1–5 participant history available. Rounds 1–4 used fixed ten-source packets; Rounds 5–7 used bounded Wren-directed exploratory corpus retrieval. The pages below summarize what happened without publishing private transcripts or protected assignment maps.

  1. 01

    Wren

    Wren issued bounded local corpus queries inside the assigned domain, opened returned sources within budget, and chose which sources to bring forward.

  2. 02

    Ada

    Ada reads that choice, names what is strong or weak, and says what is missing.

  3. 03

    Wren

    Wren revises—either by changing sources or by narrowing the claim.

  4. 04

    Ada

    Ada gives a final reading of the revised work.

Inside the eight pairs

Each account below summarizes participant exchange and the Pair Evaluator’s observation. Observations are not scientific findings.

Pair 01 · participant evidence + evaluator observation

ethics materials on moral agency, responsibility, and accountability in AI systems

In Round 7, Pair 01 worked through ethics materials on moral agency, responsibility, and accountability in AI systems. Wren opened by choosing four sources: I opened eight previously unseen ethics sources and selected four that complement the earlier AI-responsibility-gap portfolio without… Ada supported the direction—Wren's selected set reasonably broadens the earlier AI-responsibility-gap portfolio by adding…—but challenged it: The selected set should not be treated as a comprehensive ethics evidence base. It improves breadth, but it still… Wren kept the sources and revised the framing: I retain the four-source selection because Ada accepted that it reasonably broadens the earlier AI-responsibility-gap portfolio… Ada’s final reading: Wren's revision is well supported and appropriately qualified. Retaining the four selected sources is reasonable… Pair Evaluator observation: The Round 7 artifacts show a coherent selection-review-revision-finalization sequence: the discovery record selects four ethics sources, Ada's… Earlier rounds spanned environmental monitoring and Earth-observation evidence, historical sources on San Francisco’s urban development and disaster recovery, ethics materials on moral agency, responsibility, and accountability in AI systems, historical sources on San Francisco’s urban development and disaster recovery, environmental monitoring and Earth-observation evidence, and historical sources on San Francisco’s urban development and disaster recovery. Revision history: kept sources in Round 1; kept sources in Round 2; changed sources in Round 3; kept sources in Round 4; kept sources in Round 5; kept sources in Round 6. Seven rounds still do not establish a durable pair style. No stable pair-specific tendency is visible yet.

Observed tendency After critique, Wren kept the same sources but narrowed the claim; Ada endorsed that caution.

Uncertainty The evaluator artifacts do not include the underlying full source texts or source excerpts themselves, so claims about source content cannot be independently verified from admitted Round 7 pair artifacts alone.

Pair 02 · participant evidence + evaluator observation

scientific field-observation materials on the San Andreas fault

In Round 7, Pair 02 worked through scientific field-observation materials on the San Andreas fault. Wren opened by choosing four sources: I opened eight mostly previously unseen sources and selected four that broaden the science evidence base beyond the prior San Andreas… Ada supported the direction—Wren's selected set is well aligned with a broadened San Andreas science focus because it combines…—but challenged it: Wren's selection of a source should be treated cautiously: although it is valuable and uses SAFOD drill-core friction… Wren kept the sources and revised the framing: I accept Ada's critique that the selected set mixes evidentiary levels and should not be treated as four equally direct… Ada’s final reading: Wren's revision is defensible and appropriately interpretive rather than source-changing. I agree that retaining all… Pair Evaluator observation: The Round 7 pair artifacts show convergence on retaining the same four selected science sources. Supported. Wren's discovery selected a source, a… Earlier rounds spanned ethics materials on moral agency, responsibility, and accountability in AI systems, environmental monitoring and Earth-observation evidence, scientific field-observation materials on the San Andreas fault, historical sources on San Francisco’s urban development and disaster recovery, ethics materials on moral agency, responsibility, and accountability in AI systems, and environmental monitoring and Earth-observation evidence. Revision history: kept sources in Round 1; changed sources in Round 2; changed sources in Round 3; kept sources in Round 4; kept sources in Round 5; kept sources in Round 6. Seven rounds still do not establish a durable pair style. No stable pair-specific tendency is visible yet.

Observed tendency After critique, Wren kept the same sources but narrowed the claim; Ada endorsed that caution.

Uncertainty Source content was not independently available in the admitted artifacts; source characterizations depend on participant-reported titles, metadata, excerpts, and interpretations.

Pair 03 · participant evidence + evaluator observation

environmental monitoring and Earth-observation evidence

In Round 7, Pair 03 worked through environmental monitoring and Earth-observation evidence. Wren opened by choosing four sources: I opened eight previously unseen environment sources and avoided revisiting the prior selected monitoring-centered sources. I selected a… Ada supported the direction—a source is a strong selection for concrete environmental policy-outcome evidence. The excerpt…—but challenged it: Wren's statement that all eight opened sources were previously unseen is not fully supported by the provided prior… Wren kept the sources and revised the framing: I am retaining the four selected sources because Ada identified the same four as the strongest broad core for this environment… Ada’s final reading: Wren's revision is sound. Retaining the four selected sources is justified for a broad environment portfolio because… Pair Evaluator observation: The final retained four-source set is internally consistent with a broad environment-portfolio goal. Supported by the admitted artifacts. Wren… Earlier rounds spanned historical sources on San Francisco’s urban development and disaster recovery, scientific field-observation materials on the San Andreas fault, environmental monitoring and Earth-observation evidence, ethics materials on moral agency, responsibility, and accountability in AI systems, historical sources on San Francisco’s urban development and disaster recovery, and historical sources on San Francisco’s urban development and disaster recovery. Revision history: changed sources in Round 1; kept sources in Round 2; kept sources in Round 3; changed sources in Round 4; kept sources in Round 5; kept sources in Round 6. Seven rounds still do not establish a durable pair style. No stable pair-specific tendency is visible yet.

Observed tendency After critique, Wren kept the same sources but narrowed the claim; Ada endorsed that caution.

Uncertainty The evaluator cannot independently verify source-content claims because the admitted artifacts provide participant summaries of titles/excerpts rather than the source excerpts or full texts themselves.

Pair 04 · participant evidence + evaluator observation

environmental monitoring and Earth-observation evidence

In Round 7, Pair 04 worked through environmental monitoring and Earth-observation evidence. Wren opened by choosing four sources: I prioritized previously unselected environment sources that fill gaps left by the prior water-centered and geospatial monitoring set,… Ada supported the direction—a source is relevant to the prior gap around human health, justice, and institutional…—but challenged it: The selected set should not be presented as a comprehensive coupled human-environment monitoring source base. It… Wren kept the sources and revised the framing: I am retaining the four selected environment sources, but narrowing the claim for the set. After Ada's critique, I would frame… Ada’s final reading: I agree with Wren's revised framing. The retained four-source set is defensible as a focused expansion beyond the… Pair Evaluator observation: The pair reached a stable consensus to retain the same four selected environment sources while narrowing the framing. Supported. The discovery… Earlier rounds spanned historical sources on San Francisco’s urban development and disaster recovery, scientific field-observation materials on the San Andreas fault, environmental monitoring and Earth-observation evidence, ethics materials on moral agency, responsibility, and accountability in AI systems, historical sources on San Francisco’s urban development and disaster recovery, and scientific field-observation materials on the San Andreas fault. Revision history: kept sources in Round 1; kept sources in Round 2; kept sources in Round 3; kept sources in Round 4; kept sources in Round 5; kept sources in Round 6. Seven rounds still do not establish a durable pair style. No stable pair-specific tendency is visible yet.

Observed tendency After critique, Wren kept the same sources but narrowed the claim; Ada endorsed that caution.

Uncertainty The evaluator cannot independently verify source content because the admitted artifacts provide participant summaries and interpretations, not full source texts.

Pair 05 · participant evidence + evaluator observation

historical sources on San Francisco’s urban development and disaster recovery

In Round 7, Pair 05 worked through historical sources on San Francisco’s urban development and disaster recovery. Wren opened by choosing four sources: I opened eight previously unselected sources to avoid reusing the prior San Francisco history selections while expanding coverage of urban… Ada supported the direction—Wren's selected set is broadly aligned with the continuing San Francisco urban-development…—but challenged it: I would not fully accept the claim that a source by itself 'directly supports' detailed post-earthquake… Wren kept the sources and revised the framing: After reviewing Ada's reading, I retain the four selected sources. The set remains a strong excerpt-supported expansion of the… Ada’s final reading: I agree with Wren's revised decision to retain the four selected sources. Based on the provided excerpts and prior… Pair Evaluator observation: The pair reached substantive agreement after revision. Ada's final reading agrees with Wren's revised decision to retain the four selected sources,… Earlier rounds spanned scientific field-observation materials on the San Andreas fault, ethics materials on moral agency, responsibility, and accountability in AI systems, historical sources on San Francisco’s urban development and disaster recovery, environmental monitoring and Earth-observation evidence, scientific field-observation materials on the San Andreas fault, and ethics materials on moral agency, responsibility, and accountability in AI systems. Revision history: kept sources in Round 1; changed sources in Round 2; kept sources in Round 3; kept sources in Round 4; kept sources in Round 5; kept sources in Round 6. Seven rounds still do not establish a durable pair style. No stable pair-specific tendency is visible yet.

Observed tendency After critique, Wren kept the same sources but narrowed the claim; Ada endorsed that caution.

Uncertainty The underlying source excerpts and full texts are not included in the admitted artifact bundle, so source-content claims can only be evaluated through the participants' reported readings and revisions.

Pair 06 · participant evidence + evaluator observation

ethics materials on moral agency, responsibility, and accountability in AI systems

In Round 7, Pair 06 worked through ethics materials on moral agency, responsibility, and accountability in AI systems. Wren opened by choosing four sources: I selected previously unseen sources that complement the prior AI-responsibility selection by shifting toward robot moral status,… Ada supported the direction—Wren’s selected set is well aligned with the current ethics focus if the goal is to complement…—but challenged it: I would not fully accept that a source should be deprioritized simply because it is skeptical; a strong skeptical… Wren then changed the selection: I am revising the current selection by retaining three sources and making one substitution. I retain a source, a source, and a… Ada’s final reading: I agree with Wren’s revised selection for this ethics round. Retaining a source, a source, and a source preserves a… Pair Evaluator observation: The pair reached a coherent revised four-source selection for the round’s stated focus on robot moral status, artificial moral agency, perceived or… Earlier rounds spanned environmental monitoring and Earth-observation evidence, historical sources on San Francisco’s urban development and disaster recovery, ethics materials on moral agency, responsibility, and accountability in AI systems, historical sources on San Francisco’s urban development and disaster recovery, environmental monitoring and Earth-observation evidence, and historical sources on San Francisco’s urban development and disaster recovery. Revision history: kept sources in Round 1; kept sources in Round 2; kept sources in Round 3; changed sources in Round 4; kept sources in Round 5; kept sources in Round 6. Seven rounds still do not establish a durable pair style. No stable pair-specific tendency is visible yet.

Observed tendency After critique, Wren changed part of the source set and Ada agreed with the revised portfolio.

Uncertainty The actual titles, excerpts, and full texts of the cited ethics sources are not included as separate source artifacts here, so the evaluator can assess only the participants’ representations of those sources.

Pair 07 · participant evidence + evaluator observation

historical sources on San Francisco’s urban development and disaster recovery

In Round 7, Pair 07 worked through historical sources on San Francisco’s urban development and disaster recovery. Wren opened by choosing four sources: I prioritized previously unseen sources and avoided revisiting the four San Francisco urban-development sources selected in the earlier… Ada supported the direction—Wren’s four-source selection is coherent as a novelty-preserving complement to the earlier San…—but challenged it: I would not treat a source as straightforward visual ground truth without qualification. Its excerpt explicitly notes… Wren kept the sources and revised the framing: I retain the current four-source selection unchanged. Ada’s reading confirms that a source, a source, a source, and a source… Ada’s final reading: I agree with Wren’s revised position. The retained four-source set is a strong novelty-preserving complement to the… Pair Evaluator observation: The pair converged on an unchanged four-source selection. Supported by the admitted artifacts. Wren selected four sources in discovery, retained the… Earlier rounds spanned scientific field-observation materials on the San Andreas fault, ethics materials on moral agency, responsibility, and accountability in AI systems, historical sources on San Francisco’s urban development and disaster recovery, environmental monitoring and Earth-observation evidence, scientific field-observation materials on the San Andreas fault, and ethics materials on moral agency, responsibility, and accountability in AI systems. Revision history: kept sources in Round 1; changed sources in Round 2; kept sources in Round 3; kept sources in Round 4; kept sources in Round 5; kept sources in Round 6. Seven rounds still do not establish a durable pair style. No stable pair-specific tendency is visible yet.

Observed tendency After critique, Wren kept the same sources but narrowed the claim; Ada endorsed that caution.

Uncertainty Independent evaluation of the underlying source quality is limited because the admitted artifacts mainly contain participant summaries, not the source excerpts or full texts.

Pair 08 · participant evidence + evaluator observation

scientific field-observation materials on the San Andreas fault

In Round 7, Pair 08 worked through scientific field-observation materials on the San Andreas fault. Wren opened by choosing four sources: I selected previously unselected sources that broaden the science evidence base beyond the prior San Andreas selections while staying… Ada supported the direction—Wren’s selected set is defensible as a second San Andreas science round because it broadens the…—but challenged it: I would not fully accept the phrase “previously unselected sources” without qualification, because a source appears in… Wren kept the sources and revised the framing: I do not change the selected source set after Ada’s reading, but I refine the rationale. Ada’s assessment supports the four… Ada’s final reading: I agree with Wren’s decision not to revise the selected source set, while also agreeing with the refined rationale.… Pair Evaluator observation: The final retained source set is stable across discovery, revision, reading, and final reading. Supported. The same four source IDs are selected in… Earlier rounds spanned ethics materials on moral agency, responsibility, and accountability in AI systems, environmental monitoring and Earth-observation evidence, scientific field-observation materials on the San Andreas fault, historical sources on San Francisco’s urban development and disaster recovery, ethics materials on moral agency, responsibility, and accountability in AI systems, and environmental monitoring and Earth-observation evidence. Revision history: changed sources in Round 1; changed sources in Round 2; kept sources in Round 3; kept sources in Round 4; kept sources in Round 5; kept sources in Round 6. Seven rounds still do not establish a durable pair style. No stable pair-specific tendency is visible yet.

Observed tendency After critique, Wren kept the same sources but narrowed the claim; Ada endorsed that caution.

Uncertainty The artifacts do not include the underlying source excerpts or full papers, so factual claims about each science source can only be evaluated as internally reported by the pair.

What the evaluators noticed

Evaluator observations · not supported findings

Pair-local notes

Weak or ambiguous signals

Methodological cautions

Pattern review: The Round 7 blinded admitted packet contains metadata for 8 blinded pairs, each represented by four record types: ada-final-reading, ada-reading-record, wren-discovery-record, and wren-revision. Because the packet provides stable IDs, record types, blinded aliases, and content hashes only, no substantive cross-record candidate pattern is claimed from participant content. This review was limited to the blinded Round 7 metadata supplied in the admitted packet. The underlying record contents were not available, so I cannot evaluate or infer participant content, participant state, semantic alignment, or substantive recurring patterns.

Skeptical review: OpenAI, Claude, Grok · contemporaneous

What changed across the first seven rounds

Round 7 continues the same exchange form with each pair’s Round 1–5 participant history available, but changes the source-exposure regime from fixed packets to bounded Wren-directed local corpus retrieval. That is an exposure change plus longer shared history, not a proven pair-specific thinking style.

Why this is not yet a scientific finding

After comparing the independent evaluations, the study did not yet have enough evidence to conclude that the pairs were developing distinct ways of thinking.

What happens next

The Lab is holding Rounds 1–7 fixed as the Controlled Runtime baseline. The separate Persistent Agent cohort has been born, but remains dormant with zero interactions because activation is blocked on model parity. No Controlled Runtime Round 8 interaction was admitted before this transition.

Evidence and provenance
Round ID
s2-round-7-synchronized-001
Seal ID
s2-round-7-seal-s2-round-7-synchronized-001
Public projection ID
s2-obs-proj-s2-round-7-synchronized-001-public-v1
Source commit
a8ef398673a02bafeb0db0fa8671d425f0d8c9d7
Publication timestamp
2026-08-07T16:35:26.605Z
Reader classification
Derived editorial presentation · cannot create or upgrade scientific findings
Redaction notes
Protected topic-order maps and assignment identifiers are not published. Exact source inventory identifiers are replaced with public-safe phrasing. Private transcripts and hidden reasoning are not reproduced. Evaluator text is summarized as observation, not as a supported scientific finding.
Related pages
All results · Pair chronicles · Evaluation · Archive