# 03 — Narrative **The Curiosity Lab** Version: 0.1 Status: Working Draft Path: `/research/domains/03-Narrative.md` --- ## 1. Purpose This document defines the third research domain for The Curiosity Lab: **narrative**. Its purpose is to establish a working account of how narrative helps people organise events, objects, relationships and meaning, and what that implies for curiosity-oriented information architecture. This domain is central to the programme because The Curiosity Lab is not trying merely to present connected information. It is trying to investigate whether connected information can become a meaningful and memorable account of why an object matters. The current product direction has clarified an important distinction: > **Narrative does not require a linear interface.** The first prototypes treated narrative as a sequence through which the user was guided. The Living Narrative Canvas treats narrative as a persistent interpretive surface that becomes richer through exploration. This document asks: > **How can narrative organise connected meaning without forcing the user through a prescribed story?** --- ## 2. Current Status This is a Version 0.1 working draft. It is suitable for: - defining a provisional narrative model for the programme; - distinguishing narrative from linear sequence; - informing the Living Narrative Canvas; - identifying narrative entities and relationships for the ontology; - defining candidate design and evaluation principles; - guiding the Roman coin paper test; - identifying source-level Research Register entries. It is not yet suitable for: - final theoretical claims; - claims that narrative is always superior to exposition; - claims that narrative automatically creates curiosity; - automated narrative generation; - external publication; - a universal account of narrative. Current confidence: **Low to Moderate** Reason: - Narrative has a large and fragmented literature across psychology, narratology, rhetoric, education, communication, museums and interface design. - Several well-established theoretical strands are directly relevant. - The project has now produced useful practical observations about the failure of an overly linear narrative model. - The current Living Narrative Canvas model has not yet been paper-tested. - The project-specific concepts of `Meaning Accretion`, `Meaning Weave` and `Living Narrative Canvas` are not established theoretical terms. - Contradictory evidence and limits have not yet been reviewed comprehensively. --- ## 3. Domain Question The domain question is: > **How does narrative turn information and relationships into meaningful understanding, and how can that process be supported without prescribing a fixed path?** Supporting questions: 1. What distinguishes narrative from description? 2. What distinguishes narrative from chronology? 3. What distinguishes narrative from explanation? 4. Must narrative have a beginning, middle and end? 5. Can a narrative remain coherent when explored non-linearly? 6. How do readers construct causal, temporal, spatial and thematic coherence? 7. How do motifs and themes contribute to narrative meaning? 8. How does narrative interact with prior knowledge and inference? 9. When does narrative create curiosity? 10. When does narrative become manipulative or misleading? 11. How should narrative be represented in the ontology? 12. What should the Living Narrative Canvas test first? --- ## 4. Working Definitions The following definitions are provisional and programme-specific. --- ### 4.1 Description Description identifies characteristics of an object, event, person or setting. Example: ```text A bronze sestertius of Trajan, dated 103–111 CE, shows the emperor on one side and an arched bridge over a river, probably the Danube, on the other. ``` Description can be accurate and useful without yet explaining why the details matter together. For The Curiosity Lab, the initial object text is usually a description. It provides the stable starting point from which meaning can develop. --- ### 4.2 Fact A fact is a claim treated as established within a defined evidence context. Example: ```text The coin is made of bronze. ``` A narrative may contain facts, but a list of facts is not necessarily a narrative. --- ### 4.3 Event An event is a change, action or occurrence situated in time. Examples: ```text a coin was minted; an army crossed a river; a bridge was constructed; a coin entered circulation; a person exchanged it in a market. ``` Events are common narrative units, but an object-led narrative may also depend on states, systems, symbols and relationships. --- ### 4.4 Narrative Narrative is an organised representation of events, states, actors, objects and relationships that makes change or significance intelligible. For this programme, narrative is not defined only as a conventional plot. A narrative may organise meaning through: ```text time cause agency contrast change motif theme perspective consequence relationship ``` Working definition: > **Narrative is a meaning-bearing organisation of relationships that helps a person understand what happened, why it matters, or how one thing changes the interpretation of another.** --- ### 4.5 Story Story is often used interchangeably with narrative. For this programme, a practical distinction is useful: ```text story = represented events, actors and changes narrative = the particular organisation and presentation that makes them meaningful ``` This distinction is provisional. It should not be treated as a settled universal definition. --- ### 4.6 Plot Plot is an arrangement of events that creates intelligible progression. Plot often depends on: ```text sequence causality tension change resolution ``` The Living Narrative Canvas does not require every object to have a conventional plot. An object narrative may instead develop through cumulative interpretation. --- ### 4.7 Narrative Frame A Narrative Frame is an organising perspective that shapes how details are interpreted. Examples: ```text the coin as money the coin as portable authority the bridge as infrastructure the bridge as political memory the coin as an everyday handled object ``` The same fact may carry different meaning inside different frames. --- ### 4.8 Motif A motif is a recurring or emphasised element that gathers significance through repetition or connection. Examples in the Roman coin: ```text bridge crossing movement portrait circulation metal authority ``` A motif is usually more concrete than a theme. --- ### 4.9 Theme A theme is a broader organising idea that emerges across details, events or relationships. Examples: ```text portable authority infrastructure as power trust movement public memory ordinary life ``` A theme should emerge through evidence and associations. It should not merely be imposed as a topic label. --- ### 4.10 Narrative Anchor A Narrative Anchor is a word, phrase, object detail, person, date, place or concept that can open further interpretation. Examples: ```text Trajan bronze sestertius 103–111 CE bridge Danube coin ``` An anchor is part of the information architecture. It becomes narratively meaningful when its associations change how the object is understood. --- ### 4.11 Narrative Insert A Narrative Insert is a short authored passage that adds an explored meaning to the visible object narrative. Example: ```text The bridge therefore becomes more than scenery. It links military movement, engineering, public memory and the ordinary handling of money. ``` A Narrative Insert should not simply add another fact. It should integrate a fact or association into the object’s developing meaning. --- ### 4.12 Meaning Weave Meaning Weave is the project term for adding an explored Narrative Insert into the visible narrative. It is an interaction and a recorded relationship. It does not imply that the user authors the full story. --- ### 4.13 Meaning Accretion Meaning Accretion is the project term for the visible enrichment of a narrative as associations are explored and woven into it. This is not currently claimed as an established theoretical term. Working model: ```text initial description + contextual relationship + motif + theme + structural echo = expanded interpretive narrative ``` --- ### 4.14 Living Narrative Canvas A Living Narrative Canvas is a persistent object narrative that remains visible while the user explores meaningful associations around it. The narrative is not presented only at the beginning or reconstructed only at the end. It remains the surface on which exploration happens. --- ## 5. Core Distinctions --- ### 5.1 Narrative vs Description Description answers: ```text What is here? What does it look like? What are its basic facts? ``` Narrative begins to answer: ```text What happened? What changed? Why do these details belong together? Why does this matter? ``` The distinction is not absolute. A description can contain narrative potential. For the MVP, the initial description should contain enough tension, specificity or unusual detail to invite expansion. --- ### 5.2 Narrative vs Chronology Chronology orders items in time. Narrative makes temporal order meaningful. Example chronology: ```text 101–102 CE: first Dacian campaign 103–111 CE: coin dated 105–106 CE: second Dacian campaign 113 CE: Trajan’s Column dedicated ``` Narrative interpretation: ```text The coin belongs to a period in which Trajan’s military expansion, river crossing, engineering and public memory were being connected. ``` Chronology may support narrative, but chronology alone does not explain significance. --- ### 5.3 Narrative vs Explanation Explanation clarifies why or how something occurs. Narrative may contain explanation, but it usually situates explanation within events, perspectives, tensions or consequences. Example explanation: ```text Bridges allow movement across rivers. ``` Narrative interpretation: ```text On this coin, the bridge may make Roman movement and engineering part of a portable account of imperial capability. ``` A useful Living Narrative Canvas should preserve evidence and explanation. It should not substitute dramatic language for understanding. --- ### 5.4 Narrative vs Linear Navigation This is the most important distinction for v0.3. Narrative can be coherent without being delivered through a fixed sequence of screens. Linear navigation is an interface structure: ```text screen 1 → screen 2 → screen 3 ``` Narrative coherence is a meaning structure: ```text these details belong together because of time, cause, motif, theme or relationship ``` The first MVPs mistakenly treated linear navigation as the primary way to produce narrative. The Living Narrative Canvas tests a different possibility: ```text persistent narrative + non-linear exploration + controlled narrative updates ``` The user may explore associations in different orders while the system preserves coherence. --- ### 5.5 Narrative vs Topic Menu A topic menu begins with categories. Example: ```text Choose: Power Money Infrastructure Trust ``` This exposes the classification system before meaning has developed. The Living Narrative Canvas should instead allow themes to emerge from concrete details. Example: ```text bridge → crossing → Dacian campaigns → infrastructure as power ``` --- ### 5.6 Narrative vs User-Authored Reflection The user may contribute a thought or question. But the user should not be required to construct the entire narrative at the end. The system should make the meaning journey visible through the narrative it has assembled from explored associations. Reflection remains optional. --- ### 5.7 Narrative vs Persuasion Narrative can increase attention, absorption and persuasive force. This creates both opportunity and risk. A coherent narrative can make an interpretation feel true even when the evidence is limited. The project must therefore distinguish: ```text established fact supported interpretation working hypothesis analogy speculation ``` Narrative quality must not be used to conceal evidential weakness. --- ## 6. Major Theoretical Strands --- ## 6.1 Narrative as a Mode of Constructing Reality Bruner argues that narrative is not merely a decorative form placed around facts. It is one of the ways people organise experience and construct intelligible realities. Relevant implication: ```text meaning is shaped not only by which facts are present but by how events, intentions, norms and consequences are organised ``` For The Curiosity Lab, this supports the idea that an object can become meaningful when its details are organised into relationships. However, this does not mean every object requires a fictionalised story. The stronger design implication is: > **Information architecture always makes interpretive choices about which relationships become visible.** The Living Narrative Canvas should make those relationships explicit and revisable. --- ## 6.2 Story Structure and Recall Story-grammar research proposes that familiar structural patterns can guide how stories are encoded and recalled. Mandler and Johnson’s work is relevant because it suggests that people do not necessarily remember stories as unstructured sentence lists. They use organising structures. Potential implication: ```text coherent relationships may support memory better than disconnected facts ``` Important caution: The Roman coin experience is not a simple conventional story with a protagonist, goal, attempt and outcome. Story-grammar models should not be imported mechanically. The project should test whether thematic and causal coherence help users remember the object without forcing every object into a conventional plot. --- ## 6.3 Construction, Integration and Inference Narrative comprehension depends partly on what readers infer rather than only on what is explicitly stated. Graesser, Singer and Trabasso describe how readers construct knowledge-based inferences while building a representation of what a narrative is about. Kintsch’s construction-integration model, already used in the Knowledge and Meaning domain, is also relevant. Potential implication: ```text a narrative should provide enough structure for meaningful inference without requiring every relationship to be stated exhaustively ``` The Living Narrative Canvas can support this by: ```text showing a concrete anchor revealing a bounded association explaining the relationship basis adding a concise narrative insert leaving room for a next question ``` Risk: If an association is too implicit, the user may not understand why it exists. If it is too explicit, the experience becomes another lecture. --- ## 6.4 Situation Models and Narrative Worlds Situation-model research describes how readers build integrated mental representations of the situations described in discourse. Relevant dimensions may include: ```text time space actors goals causality objects events ``` The project implication is significant. The user should not merely remember sentences about the coin. They should form a more integrated model of: ```text the coin as an object its historical moment the people who may have handled it the systems of exchange around it the bridge as image and infrastructure the authority represented by the portrait ``` This connects directly to `Worldness`. A Context Spark is useful when it helps build the situation model of the object’s lived world. --- ## 6.5 Causal Coherence Narratives often feel coherent when events and states are linked through causes, goals and consequences. Causal coherence is useful for the project, but it requires caution. Example: ```text Trajan fought campaigns in Dacia. The coin is dated near those campaigns. The reverse depicts a bridge, probably over the Danube. ``` A supported interpretation may connect these details. But the interface should not imply a proven causal chain unless the evidence supports it. Design requirement: ```text relationship strength must be represented separately from narrative fluency ``` A sentence can sound coherent while remaining only a moderate-confidence interpretation. --- ## 6.6 Temporal Coherence Narrative often organises change through time. Time can help users understand why an object belongs to a particular moment. For the Roman coin: ```text before during after later remembrance ``` may help organise: ```text Dacian campaigns coin production bridge imagery Trajan’s Column imperial memory ``` But the Living Narrative Canvas should not become a compulsory timeline. Time is one relationship basis among several. --- ## 6.7 Motif and Thematic Coherence Narrative coherence can also be thematic rather than strictly causal. Repeated or connected motifs may accumulate significance. For the Roman coin: ```text bridge river movement circulation crossing frontier ``` can gather around the theme: ```text movement made political ``` Similarly: ```text portrait inscription metal recognition acceptance ``` can gather around: ```text portable authority and trust ``` This is especially relevant to the Living Narrative Canvas because themes can emerge in different exploration orders. --- ## 6.8 Narrative Tension and Curiosity Narrative often creates attention through uncertainty, conflict, anomaly or an unresolved question. The earlier prototype promoted Narrative Tension into a fixed stage. The current position is more cautious. Tension remains useful, but it should not become a compulsory screen. Potential tensions in the Roman coin include: ```text Why place a bridge on money? Why combine an emperor and infrastructure? How can an ordinary object carry an imperial claim? What made people trust this piece of bronze? ``` These questions may open curiosity. But the interface should not manufacture false mystery. Tension must arise from the object or relationship. --- ## 6.9 Narrative Transportation Green and Brock describe transportation as absorption involving attention, imagery and affect. Narrative transportation can increase story-consistent beliefs and evaluations. For the project, the important lesson is double-edged. Potential benefit: ```text a richer narrative may hold attention and make an object feel present ``` Potential risk: ```text absorption may reduce scrutiny and increase acceptance of weak interpretations ``` The Curiosity Lab should not optimise for transportation as an end in itself. The desired outcome is not merely immersion. It is: ```text engaged interpretation + evidential awareness + better questions ``` --- ## 6.10 Narrative Rationality Fisher’s narrative paradigm emphasises that people assess narratives through forms of coherence and fidelity. This is relevant because users may judge whether an object story: ```text hangs together feels internally consistent fits what they know resonates with wider experience ``` However, narrative fidelity is not a replacement for historical evidence. A compelling modern parallel can feel right while being historically misleading. Project implication: ```text narrative plausibility and evidential confidence must remain separate ``` --- ## 6.11 Narrative in Science and Public Communication Research on narrative communication suggests that narrative forms can support comprehension and engagement for nonexpert audiences. Dahlstrom also stresses that narrative communication brings ethical and epistemic choices. Relevant project implication: ```text narrative may help make complex relationships accessible but may also simplify, select and persuade ``` The interface should therefore show enough evidence structure to preserve trust. --- ## 6.12 Narrative Visualisation and Reader Agency Segel and Heer’s work on narrative visualisation is useful because it examines the design tension between author-driven communication and reader-driven exploration. This maps closely to the Living Narrative Canvas. Too author-driven: ```text the system dictates the story and order ``` Too reader-driven: ```text the user receives a graph of possibilities without interpretive help ``` The v0.3 design target is a hybrid: ```text authored base narrative + authored associations + user-selected order + controlled meaning weaving ``` This balance is central. The user has agency over exploration. The system remains responsible for coherence, evidence and wording. --- ## 6.13 Narrative and Object Interpretation Museum interpretation frequently uses objects as entry points into people, systems, events and meanings. An object can hold multiple narratives: ```text production use exchange discovery collection classification display political symbolism modern interpretation ``` The Curiosity Lab should avoid implying that an object has only one true story. Instead, it should present: ```text bounded interpretive narratives with visible relationship bases and evidence status ``` The Living Narrative Canvas is therefore not a mechanism for revealing “the story”. It is a mechanism for building **a supported interpretation**. --- ## 7. Working Model: Narrative as Connected Meaning The current working model is: ```text Object Description ↓ Narrative Anchors ↓ Meaningful Associations ↓ Motifs and Relationship Bases ↓ Narrative Inserts ↓ Visible Meaning Accretion ↓ Expanded Interpretation ↓ New Question or Transfer ``` This model differs from the retired linear journey. It does not require a single order. It requires: ```text a stable object centre bounded associations coherent inserts visible accumulation evidence-aware interpretation ``` The user can explore: ```text bridge first coin first Trajan first date first ``` The resulting narrative may differ in order and emphasis. But it should remain coherent. --- ## 8. Narrative and the Curiosity Cycle The project’s broader curiosity cycle can be expressed narratively as: ```text Observation → Curiosity → Connection → Understanding → Wonder ``` Narrative may support this cycle by: ### Observation The narrative begins with a concrete object description. ### Curiosity An unusual detail, gap or juxtaposition opens a question. ### Connection Associations connect the detail to wider contexts, motifs, systems or patterns. ### Understanding Narrative Inserts explain why those relationships matter together. ### Wonder The expanded narrative leaves the object more meaningful and opens further questions. The cycle should not be shown as a mandatory interface sequence. It is an analytical model. --- ## 9. Design Principles --- ### NP-001 — Narrative Without Forced Sequence Narrative coherence does not require linear navigation. --- ### NP-002 — Keep the Object Present The initial object narrative should remain visible while wider meanings are explored. --- ### NP-003 — Description Before Interpretation Begin with a stable account of what the object is before adding broader claims. --- ### NP-004 — Relationship Before Flourish Explain why details connect before using evocative language. --- ### NP-005 — Motifs Gather Meaning Allow concrete motifs to accumulate thematic significance. --- ### NP-006 — Themes Should Emerge Do not begin with an abstract theme menu. --- ### NP-007 — Every Insert Must Change Understanding A Narrative Insert should enrich interpretation, not merely lengthen the text. --- ### NP-008 — Preserve Evidence Status Narrative fluency must not disguise uncertainty. --- ### NP-009 — User Agency, Authorial Responsibility The user chooses what to explore. The system remains responsible for evidence, wording and coherence. --- ### NP-010 — Do Not Manufacture Tension Questions should arise from real features, gaps or relationships. --- ### NP-011 — Leave Productive Openings The narrative does not need to resolve every question. --- ### NP-012 — Narrative Growth Must Remain Legible Meaning Accretion should not become paragraph accumulation without structure. --- ### NP-013 — Multiple Interpretations May Coexist An object may support several evidence-aware narratives. --- ### NP-014 — Structural Echoes Need Return A cross-domain analogy must change how the original object is understood. --- ## 10. Architecture Implications The narrative domain suggests the following entities: ```text Narrative BaseNarrative CurrentNarrative NarrativeFrame NarrativeAnchor NarrativeInsert Motif Theme Event Actor Setting Perspective Tension Association RelationshipBasis MeaningWeave MeaningTrace Interpretation EvidenceCategory ConfidenceLevel ``` Candidate relationships: ```text Object has_base_narrative Narrative Narrative contains_anchor NarrativeAnchor NarrativeAnchor opens Association Association develops_motif Motif Association reveals_theme Theme Association provides_insert NarrativeInsert NarrativeInsert woven_into CurrentNarrative NarrativeInsert supported_by Evidence NarrativeFrame shapes_interpretation_of Object MeaningWeave updates CurrentNarrative MeaningTrace records MeaningWeave Interpretation has_confidence ConfidenceLevel Interpretation has_relationship_basis RelationshipBasis ``` Important architectural distinction: ```text content relationship ≠ narrative presentation ``` The same association might be presented differently in different narratives. The ontology should therefore avoid treating `narrative_insert` as the relationship itself. It is an authored expression of the relationship. --- ## 11. Living Narrative Canvas Implications The v0.3 prototype should test a very small narrative model. Minimum content: ```text one Base Narrative five to seven Narrative Anchors two to four Associations per tested anchor one concise Narrative Insert per association emergent Motifs or Themes one Meaning Trace ``` The interface should allow: ```text selection without fixed order preview before weaving visible distinction between base text and added meaning revisiting explored associations prevention of duplicate inserts optional next question ``` The first test should not attempt: ```text automatic narrative generation dynamic prose rewriting multiple objects branching fictional plots personalised storytelling AI narration full narrative graphs ``` --- ## 12. Narrative Coherence Rules for the MVP A woven narrative should pass these checks. ### Referential Coherence The reader can tell what each sentence refers to. ### Temporal Coherence Dates and event relations do not contradict each other. ### Causal Restraint The prose does not imply unsupported causes. ### Thematic Coherence Each insert contributes to one or more emerging themes. ### Object Return Each insert remains connected to the coin. ### Register Coherence The tone does not shift abruptly between museum label, academic note and promotional copy. ### Evidence Coherence Fact, interpretation and analogy remain distinguishable. ### Length Coherence The narrative does not become unusably long after several inserts. --- ## 13. Measurement Candidates The first paper test should not attempt to prove that narrative causes learning. It should gather practical signals. ### Behavioural candidates ```text number of anchors voluntarily explored number of associations previewed number of inserts chosen return to previously explored anchor request to explore another motif ``` ### Comprehension candidates ```text can the user explain why one association matters? can the user identify an emerging theme? can the user distinguish fact from interpretation? can the user explain the limit of a Structural Echo? ``` ### Narrative candidates ```text does the expanded text feel coherent? does each insert feel connected? does the narrative feel richer rather than merely longer? does the user notice how the interpretation changed? ``` ### Curiosity candidates ```text does the user ask a new question? does the user voluntarily follow another connection? does the user express surprise resolved into understanding? ``` ### Negative signals ```text the experience feels like a lesson anchors feel like menu buttons the story becomes repetitive associations feel arbitrary the user cannot remember the object the analogy becomes more memorable than the evidence the user believes an interpretation is proven fact ``` --- ## 14. Implications for The Curiosity Lab Hypothesis The narrative domain supports a refinement of the programme hypothesis. Earlier formulation: ```text Connected information can cultivate curiosity. ``` Narrative refinement: > **Connected information may cultivate curiosity when relationships are organised into an intelligible and visibly developing interpretation.** This adds an important constraint. Connections alone may produce: ```text a graph a menu a list a collection of trivia ``` Narrative helps organise those connections into: ```text significance coherence change in interpretation ``` The working proposition becomes: ```text Connection + Narrative Coherence + User Agency + Evidential Transparency → Candidate Meaningful Curiosity ``` This remains a hypothesis. --- ## 15. Known Weaknesses The current model has several weaknesses. ### 15.1 Narrative is broadly defined The working definition may be too inclusive. If every meaningful relationship becomes narrative, the term loses precision. ### 15.2 The model may privilege prose Meaning may also develop visually, spatially, materially or through interaction. The canvas should not assume that all meaning must become paragraphs. ### 15.3 Narrative may create false coherence A fluent account can make weak evidence seem settled. ### 15.4 Meaning Accretion may become text accumulation Adding more sentences is not automatically deeper understanding. ### 15.5 User agency may be superficial Choosing among authored associations is not the same as open exploration. ### 15.6 Narrative may suppress contradiction A clean story may hide uncertainty, competing interpretations or unresolved evidence. ### 15.7 The object may become a pretext Wider themes may overwhelm the specific object. --- ## 16. Contradictions and Cautions --- ### 16.1 Coherence vs Openness Narrative benefits from coherence. Curiosity benefits from open questions. The interface must not resolve everything too neatly. --- ### 16.2 Author Guidance vs User Agency Authored narrative protects coherence and evidence. User choice protects exploration and ownership. The design must balance both. --- ### 16.3 Engagement vs Scrutiny Absorbing narratives may increase engagement while reducing critical attention. Evidence cues and limits of interpretation are necessary. --- ### 16.4 Specificity vs Transfer Historical specificity protects accuracy. Structural Echoes support transfer across disciplines. Transfer should follow contextual grounding rather than replace it. --- ### 16.5 Narrative Growth vs Readability The story should become richer. It should not become an unedited transcript of everything explored. The MVP may eventually require: ```text grouping compression layering reordering summary ``` These should not be implemented before basic testing. --- ### 16.6 Multiple Narratives vs Product Simplicity An object may support several valid narratives. The first MVP should test one authored interpretive frame without implying exclusivity. --- ## 17. Research Debt | ID | Research debt | Priority | Status | |---|---|---:|---| | RD-NAR-001 | Define a narrower operational distinction between narrative, explanation and interpretation. | High | Open | | RD-NAR-002 | Review narratology sources on story/discourse, plot, focalisation and non-linear narrative. | High | Open | | RD-NAR-003 | Test whether users perceive visible narrative growth. | High | Open | | RD-NAR-004 | Define narrative coherence criteria for unordered inserts. | High | Open | | RD-NAR-005 | Test whether Narrative Inserts feel integrated or appended. | High | Open | | RD-NAR-006 | Review evidence on narrative persuasion and reduced scrutiny. | High | Open | | RD-NAR-007 | Determine how alternative interpretations should coexist. | Medium | Open | | RD-NAR-008 | Test whether motifs are clearer than themes as interaction anchors. | Medium | Open | | RD-NAR-009 | Determine when the narrative needs compression or summarisation. | Medium | Open | | RD-NAR-010 | Decide whether `Meaning Accretion` remains the programme term. | Medium | Open | | RD-NAR-011 | Add narrative-specific source entries to the Research Register. | High | Open | | RD-NAR-012 | Compare prose-based and visual narrative growth. | Low | Open | --- ## 18. Register Entries to Create Recommended first source-level entries: ```text RR-NAR-0001 — Bruner (1991), The Narrative Construction of Reality RR-NAR-0002 — Mandler & Johnson (1977), Remembrance of Things Parsed RR-NAR-0003 — Graesser, Singer & Trabasso (1994), Constructing Inferences During Narrative Text Comprehension RR-NAR-0004 — Zwaan & Radvansky (1998), Situation Models in Language Comprehension and Memory RR-NAR-0005 — Green & Brock (2000), Narrative Transportation RR-NAR-0006 — Fisher (1984), Narration as a Human Communication Paradigm RR-NAR-0007 — Dahlstrom (2014), Narrative Science Communication RR-NAR-0008 — Segel & Heer (2010), Narrative Visualization ``` --- ## 19. Immediate Architecture Promotions Candidate ontology promotions: ```text NarrativeFrame Motif Theme NarrativeCoherence Interpretation Perspective ``` Candidate relationship promotions: ```text develops_motif reveals_theme shapes_interpretation_of expresses_relationship has_narrative_frame has_coherence_basis ``` Do not promote all of these automatically. Each should be reviewed against the current v0.3 Ontology. --- ## 20. Immediate Pattern Promotions Candidate patterns: ```text Persistent Narrative Surface Description Before Interpretation Motif-to-Theme Reveal Narrative Insert With Evidence Cue Narrative Coherence Without Linear Navigation Productive Unresolved Question Alternative Interpretation ``` Candidate anti-patterns: ```text Linear Story Wizard Theme Menu Narrative Pile-Up False Resolution Fluent Overclaim Object as Pretext Compulsory Reflection ``` --- ## 21. Immediate MVP Promotions The Roman coin paper test should evaluate: ```text whether the base narrative remains cognitively present; whether motifs invite exploration; whether each insert changes interpretation; whether themes emerge naturally; whether the resulting text remains coherent; whether the user asks another question; whether evidence and analogy remain distinguishable. ``` The MVP should not yet attempt to test narrative transportation, persuasion or long-term recall. Those require more controlled research designs. --- ## 22. Relationship to Other Domain Papers ### Curiosity Narrative can open and sustain information gaps, but should not manufacture them. ### Knowledge and Meaning Narrative helps integrate information into situation models, causal relations and thematic structures. ### Museum Interpretation Narrative can reveal significance around an object while keeping the object central. ### Knowledge Representation Narrative requires entities and relations, but should not be reduced to a graph. ### Information Architecture The architecture must balance authored coherence with user-selected navigation. ### Patterns Across Disciplines Structural Echoes can become narrative expansions when they return meaning to the object. ### Connected Curiosity Narrative may be one mechanism by which isolated associations become a meaningful network. --- ## 23. Provisional Conclusion The current working conclusion is: > **Narrative is not primarily a sequence of screens. It is an organisation of relationships that makes change, context and significance intelligible.** For The Curiosity Lab, this means that a narrative interface should not simply guide a user through authored steps. It should help the user see how an object’s meaning develops as relationships are explored. The Living Narrative Canvas therefore tests a specific proposition: > **A persistent object narrative can remain coherent while users explore associations in different orders, provided that each association has a clear relationship basis and is woven back into the object’s interpretation.** The central design challenge is not: ```text How do we tell the user a story? ``` It is: ```text How do we make connected meaning visible without turning exploration into either a prescribed lesson or an incoherent collection of links? ``` Confidence: **Low to Moderate** --- ## 24. References Bruner, J. (1991). The narrative construction of reality. *Critical Inquiry*, 18(1), 1–21. https://doi.org/10.1086/448619 Dahlstrom, M. F. (2014). Using narratives and storytelling to communicate science with nonexpert audiences. *Proceedings of the National Academy of Sciences*, 111(Supplement 4), 13614–13620. https://doi.org/10.1073/pnas.1320645111 Fisher, W. R. (1984). Narration as a human communication paradigm: The case of public moral argument. *Communication Monographs*, 51(1), 1–22. https://doi.org/10.1080/03637758409390180 Graesser, A. C., Singer, M., & Trabasso, T. (1994). Constructing inferences during narrative text comprehension. *Psychological Review*, 101(3), 371–395. https://doi.org/10.1037/0033-295X.101.3.371 Green, M. C., & Brock, T. C. (2000). The role of transportation in the persuasiveness of public narratives. *Journal of Personality and Social Psychology*, 79(5), 701–721. https://doi.org/10.1037/0022-3514.79.5.701 Kintsch, W. (1988). The role of knowledge in discourse comprehension: A construction-integration model. *Psychological Review*, 95(2), 163–182. https://doi.org/10.1037/0033-295X.95.2.163 Mandler, J. M., & Johnson, N. S. (1977). Remembrance of things parsed: Story structure and recall. *Cognitive Psychology*, 9(1), 111–151. https://doi.org/10.1016/0010-0285(77)90006-8 Segel, E., & Heer, J. (2010). Narrative visualization: Telling stories with data. *IEEE Transactions on Visualization and Computer Graphics*, 16(6), 1139–1148. https://doi.org/10.1109/TVCG.2010.179 Zwaan, R. A., & Radvansky, G. A. (1998). Situation models in language comprehension and memory. *Psychological Bulletin*, 123(2), 162–185. https://doi.org/10.1037/0033-2909.123.2.162 --- ## 25. Next Action Do not expand this paper immediately. The next action should be to convert its strongest claims into a paper-testable narrative content document: ```text /mvp/Roman-Coin-Living-Canvas-Content.md ``` That document should define: ```text the stable Base Narrative; the Motifs and Narrative Anchors; the Associations each anchor opens; the Relationship Basis of each association; the Narrative Insert; the Evidence Category and Confidence Level; the theme or interpretation changed by the insert; the limits of each Structural Echo. ``` This will allow the project to test narrative coherence before spending more implementation tokens.