{"path":"research/curiosity-as-connection-zurn-bassett.md","content":"# Curiosity as Connection — Zurn & Bassett, read against the corpus\n\n**Date**: 2026-08-30 · **Type**: source reflection · **Source**: Overthink podcast ep. 62\n(2022-10-11), interview with Perry Zurn (philosophy, American University) and Dani S. Bassett\n(network science / bioengineering, Penn), co-authors of *Curious Minds: The Power of Connection*;\nfull transcript retrieved from the podcast's official site, stored privately in\n`sources/podcasts/`. Founder-flagged 2026-08-30 as relevant to Deliberus \"on different planes of\nabstraction.\"\n\n---\n\n## 1. The headline: curiosity IS edgework, and they measured it\n\nThe corpus's curiosity doc ([curiosity-as-growth-fuel.md](curiosity-as-growth-fuel.md)) is built on\nKahan (curiosity reverses motivated reasoning), PACE (closable gaps), and Loewenstein (epiphany).\nZurn & Bassett supply the piece none of those have: **a theory of what curiosity *does*, and it is\ngraph-building.** They reject the standard \"acquisitional\" framing — curiosity as a desire to\nacquire information — for curiosity as **\"a practice of connection\"**: connecting what you know to\nwhat you are about to know, building *\"some kind of scaffold of knowledge.\"* Knowledge as network,\ncuriosity as the act of adding edges. Deliberus's contribution mechanic is not merely *compatible*\nwith curiosity on this theory — **the contribution act and the curious act are the same act.**\nBassett's Dewey citation makes the reuse argument from the pedagogy side: knowledge is \"a network of\ninterconnections\" where any past experience offers \"a point of advantage\" for the next problem —\nthe reuse flywheel, stated in 1916.\n\nTheir \"blew my mind\" analysis is the epiphany factor with a mechanism: what a great class does is\nnot deliver a fact but **rearrange the whole structure** — one input restructuring many held\nconnections. That is cross-root fan-out, the curiosity-predictive quantity the corpus already\nregistered as *not yet computed* (the hinge measures single-root sensitivity; §9.1 of the curiosity\ndoc). Their frame independently confirms that fan-out, not depth, is where the felt \"wow\" lives.\n\n## 2. The three styles are traversal signatures, and they are measurable in our graph\n\nBusybody (many diverse hops) · hunter (focused deep tracking) · dancer (connects two structures so\nthat **both change shape**). Not personality types — measured **walk patterns**: in their Wikipedia\nstudy (15 min/day, 21 days), step-distance between visited pages quantified where each person sat\non the busybody–hunter axis, with styles stable-but-flexible over weeks. Three consequences:\n\n1. **Deliberus can compute the same signature** from claim-visit trails — step distance in the\n   graph — which would let surfaces serve the style in front of them (a hunter wants the descent;\n   a busybody wants lateral richness; forcing either into the other's path is the\n   one-size-descent UX failure). Session assertion, unruled, and firmly parameter-tier.\n2. **The dancer is the analogy daemon's human twin.** \"Takes some structure of information and\n   connects it up with a different structure such that the two may change in shape\" — that is\n   cross-domain structural kinship, the mother-claim analogy act, the thing cosine cannot see\n   ([the-analogy-daemon.md](the-analogy-daemon.md)). Bassett says the mathematics for detecting\n   dancer-moves is in progress — a literature to watch for the daemon's evaluation pass.\n3. **Style diversity is a register warning**: \"how people put concepts together is really, really,\n   really different,\" with explicit reference to neurodivergent thinkers. A descent UX tuned on\n   one cognitive style is an induced taxonomy waiting for its next register.\n\n## 3. Curiosity has politics, and it is our threat model in another vocabulary\n\nZurn's earlier book (*Curiosity and Power*) supplies the dark half. **Curiosity formations**\nposition who may ask about whom — the \"curiosity of the master\" (dissects, organises, opens the\nother to the gaze) versus connective curiosity. And their citation-network research measures the\nlaid-down-track problem: scholars moving through ideas are \"tracing back over structures of\nprivilege,\" with under-citation of women and scholars of colour quantified against\npublication-matched baselines. In corpus vocabulary: **frame lock-in and ingestion asymmetry,\nmeasured in the scholarly graph** — the reuse flywheel's shadow operating on citation lines. Their\nanswer (deliberately \"channel a curiosity that breaks the boundaries of the lines laid down\")\nis the re-framing act the residual-error taxonomy says has no mechanism here yet.\n\nThe algorithmic version came up unprompted: the *feeling* of free inquiry while \"following a path\nlaid out ahead of time by algorithms\" — the feed-steering worry, stated as a curiosity problem. A\nfeed that serves the styles (§ 2) without disclosing its layout is exactly this.\n\n## 4. Two smaller keepers\n\n- **Curiosity = attention + care** (the \"care killed the cat\" etymology — the proverb predates\n  \"curiosity\" in the slot). The attunement pole hiding inside the analytic drive: you are only\n  curious about what you care about. One more independent arrival at analysis-needs-attunement.\n- **The master's curiosity as decomposition warning**: Peña-Guzmán's own distinction — a curiosity\n  that \"dissects and cuts and organizes... no matter the cost.\" The universal acid wielded without\n  the proportionality principle is master-curiosity; the sacred-domain ruling (analysis may go\n  anywhere, owing attunement in proportion to depth) is the existing answer, and this gives it a\n  name from outside.\n- **Competing knowledge structures**: Zurn's closing point that there is not one network being\n  built but \"competing structures of knowledge\" with hubs that \"may not fit together,\" requiring\n  Lugones-style *world-traveling* — the worldview-lens architecture described from the\n  network-science side, and a caution against the one-graph reading that staged maturity already\n  answers (share the code, compartmentalise the contents).\n\n---\n\nCross-references: [curiosity-as-growth-fuel.md](curiosity-as-growth-fuel.md) (this doc's findings\nmirror there) · [the-analogy-daemon.md](the-analogy-daemon.md) (the dancer) ·\n[the-reuse-flywheel-prior.md](the-reuse-flywheel-prior.md) (Dewey) ·\n[worldview-lenses.md](../worldview-lenses.md) (world-traveling) ·\n[the-residual-error-taxonomy.md](the-residual-error-taxonomy.md) (frame lock-in; ingestion\nasymmetry) · [engagement-gradient-priors.md](engagement-gradient-priors.md) (styles as a\nwithin-tier axis)\n"}