Examples¶

Two complete worked reviews — one in each mode — followed by a gallery of finished lineage figures from real runs.

Topic mode: visual–cerebellar anatomy¶

Run from a single plain-English prompt, fully reproducible from the JSON files in the output directory.


Prompt	"anatomical connections between visual system and cerebellum, primate or human, any tractography method, back to the 1970s"
Output dir	`visual_cerebellum/`
Deliverable	`visual_cerebellum_bibliography.xlsx` (72 rows)
Agent search batch	42 papers, 1980–2025 ( cream rows)
Cross-citation batch	30 papers, 1944–2010 ( green rows)
Verifier corrections	3 fabrications caught: one paper had hallucinated authors (Schmahmann et al. 2025 returned as "Olson et al."), one DOI was off by a digit, one PMCID was invented
Wall-clock	~7 min, no PDF acquisition

The output directory holds the full audit trail: agent_out.json (raw agent return), verify_report.json (what was caught), xref_visual_cerebellum.json (the cross-citation frequency table), xref_picks.json (the 30 picked from it), and rows.json (what the spreadsheet renders from).

Lab mode: the Gallant lab in context¶


Prompt	"review the Gallant lab's human-imaging work in the context of the broader field"
Output dir	`gallant_lab/`
Front end	`lab_corpus.py` ingest → prune false-positives → derive themes & their drift
Deliverables	`gallant_lab_in_context_bibliography.xlsx` · contextualized lineage figure · AI-authored review `.docx`

Gallant lab themes over time — The lab's research themes and how their emphasis shifted across decades (L3).

Gallant lab work placed in the field — The contextualized lineage figure — lab papers (highlighted) within the surrounding literature.

What a finished review article looks like¶

From the complexity_representation run — AI-authored prose, canonical references, embedded figure, explicit disclosure.

Review title page — Title page: abstract, intro, and the AI-authorship disclosure note.

Canonical reference list — Canonical APA-7 references, pulled straight from the verified corpus.

Lineage figure gallery¶

Each figure groups a verified corpus into theoretical families on a citation-weighted timeline, with landmark papers auto-labelled. Click any to zoom.

Complexity representation — **How the brain represents complexity** — 6 families, 1948→2026, CDF-warped timeline.

Cognitive map representation — **The cognitive map** — spatial & relational representation.

**World models** — from internal models to modern AI.

**Distributed conceptual network** — semantic representation across cortex.

Structure representation — **How the brain represents data structures**.

**Comparative language network** — across species & methods.

Conversation brain recording — **Conversation & brain recording** — naturalistic language neuroscience.

The HTML version is interactive

These are static exports. Each run also produces an interactive HTML figure (hover for paper details, zoom, pan) plus SVG and PDF for publication.