{"closest_prior_art":[{"name":"Scalable Reading of Structured Data","overlap":"Combines dataset-level statistical analysis with systematic, reproducible selection of distinctive individual cases for close interpretation—the proposal's core macro-to-micro review pattern.","remaining_difference":"It does not preserve a tested stable literary-period claim through an explicit passage-to-work-to-corpus licensing rule, display uncertainty-bearing narrative-mode distributions, or test whether the crosswalk reduces unsupported work-level generalizations.","source_ids":["SRC2"]},{"name":"WordSeer","overlap":"Supports collection-wide distributions, sentence-level context, navigation back to original texts, detailed document reading, tagging, and annotation, closely matching the proposed distributional dashboard and drill-down review.","remaining_difference":"It is an exploratory environment rather than a decision protocol: it lacks equal-work aggregation, stability bands, predeclared excursion criteria, a level-of-analysis crosswalk, stratified center/tail review, and the proposed reviewer outcome test.","source_ids":["SRC1"]},{"name":"Voyant Tools Trends","overlap":"Provides established first-party functionality for switching between corpus-level distributions across documents and within-document distributions, making macro and local variation inspectable.","remaining_difference":"It does not encode which claims each scale licenses, preserve a validated macro claim while qualifying member-level descriptions, carry annotation uncertainty, or require human review of sustained excursions.","source_ids":["SRC4"]},{"name":"CLSInfra corpus-design guidance","overlap":"Requires explicit eligibility, scope, sampling, proportions, metadata, representativeness, bias, and canonicity decisions for computational literary corpora; its examples monitor variation by work, period, and other metadata.","remaining_difference":"It governs corpus construction and monitoring, not the downstream inferential error of applying a valid corpus profile to each work or the proposed corrective crosswalk and reviewer experiment.","source_ids":["SRC3"]}],"contrastive_claim_falsifier":"The contrastive claim would be falsified if reviewers using an existing scalable-reading workflow—configured with corpus/document distributions, systematic selection of distinctive cases, and a short level-of-analysis disclaimer—make no more unsupported work-level inferences than reviewers using the full Period–Work Narrative Mode Crosswalk, under the same masked summaries, rubric, evidence access, and review-time constraint.","contrastive_claim_remaining":"Adjacent practice already connects corpus summaries to individual texts, but the bounded search did not find a method that specifically retains a stable, equal-work literary-period profile while formally limiting its implications through an uncertainty-bearing passage-to-work-to-corpus crosswalk, predeclared excursion rules, stratified close reading, and a behavioral test of unsupported work-level inference.","experiment_id":"eoa_inverse_innovation_exp13_second_slot_policy60_20260806","gates":{"adequate_source_search":{"rationale":"The search covered the proposal directly, macroanalysis/distant-reading and scalable-reading terminology, corpus representativeness and monitoring, first-party corpus/document visualization tools, and combinations involving annotation, distributions, passages, and individual works. All four retained sources were opened; they span four publishers and include primary research, official guidance, and first-party documentation.","source_ids":["SRC1","SRC2","SRC3","SRC4"],"status":"PASS"},"bounded_next_test":{"rationale":"The proposed 24-novel, 480-passage comparison is finite, read-only, reversible, and records disagreement, unsupported inference counts, time, and halt conditions. Existing work shows that systematic case selection, passage annotation, and corpus/document distribution inspection are implementable. The pilot must predeclare its stability band and consequential-departure rubric before analysis.","source_ids":["SRC1","SRC2","SRC4"],"status":"PASS"},"distinct_testable_claim":{"rationale":"The candidate predicts fewer unsupported work-level generalizations than an aggregate-profile baseline while preserving comparable evidence requirements. This separates the claimed inferential benefit from the already-established ability to visualize multiple scales or select cases for close reading.","source_ids":["SRC1","SRC2","SRC4"],"status":"PASS"},"no_obvious_safety_or_authority_stop":{"rationale":"A reversible analysis of public-domain novels presents no obvious safety or authority stop when interpretive and selection authority remains with designated humans, annotator rankings are not published, sensitive attributes are not inferred, and small cells or unresolved disagreement trigger withdrawal of pilot flags. Corpus-boundary, metadata, and terms-of-use checks remain necessary.","source_ids":["SRC3"],"status":"PASS"},"supported_problem":{"rationale":"The literature and tools visibly distinguish collection-wide evidence from individual-text or passage evidence and motivate systematic movement between those scales. This supports the possibility and methodological importance of level overreach, although the retained sources do not measure how often period summaries are actually applied uniformly to novels; the specific prevalence and consequence therefore remain only partly established.","source_ids":["SRC1","SRC2","SRC4"],"status":"PASS"}},"prior_art_disposition":"ADJACENT_PRIOR_ART","problem_evidence":{"finding":"The problem is partly visible: established methods treat aggregate and individual-text views as different analytical scales, and scalable reading was developed to connect overall patterns with systematic close interpretation of distinctive cases. However, no retained source directly documents anthology or syllabus editors treating a stable period-level narrative-mode profile as a uniform description of every included novel.","source_ids":["SRC1","SRC2","SRC4"],"status":"PARTLY_SUPPORTED"},"research_id":"eoa_inverse_innovation_exp13_light_screen_20260806","schema_version":1,"screen_id":"E13P082","screen_survival":true,"search_lanes":{"component_combination":{"no_result_note":"No retained source combined equal-work period-profile stability, per-work and within-work narrative-mode distributions, uncertainty, excursion alerts, stratified close reading, and an explicit claim-licensing crosswalk in one protocol.","queries":["literary studies dashboard corpus individual text distributions close reading stratified sampling","computational literary studies levels of analysis corpus text passage","literary corpus annotation uncertainty inter annotator disagreement passage close reading","computational literary studies corpus-level results individual works variation"],"source_ids":["SRC1","SRC2","SRC3","SRC4"]},"direct_problem_and_intervention":{"no_result_note":"Direct searches found adjacent macro/micro and scalable-reading methods but no exact Period–Work Narrative Mode Crosswalk or direct evaluation of unsupported work-level generalization from a valid period profile.","queries":["literary period corpus aggregate individual works distribution close reading distant reading narrative mode","computational literary studies corpus-level claims individual texts heterogeneity aggregation","literary studies computational analysis overgeneralization corpus individual works","distant reading aggregation hides variation individual texts criticism"],"source_ids":["SRC1","SRC2"]},"products_practices_and_standards":{"no_result_note":null,"queries":["Voyant Tools corpus document trends individual documents distribution official","WordSeer literary corpus representative passages corpus-wide individual texts visualization","literary corpus representativeness sampling metadata guidelines corpus design","literary corpus annotation visualization document distribution official"],"source_ids":["SRC1","SRC3","SRC4"]},"synonyms_and_historical_terms":{"no_result_note":null,"queries":["scalable reading literary corpus close distant reading individual texts","zoomable reading literary studies corpus visualization individual works","macroanalysis literary history individual texts close reading distribution","distant reading ecological fallacy individual texts aggregate"],"source_ids":["SRC1","SRC2","SRC4"]}},"sources":[{"claims_supported":["WordSeer was designed with literary scholars to investigate hypotheses across text collections.","It presents collection-wide distributions while allowing users to open sentence occurrences in original texts.","It supports detailed document reading, tagging, notes, and annotation, but reported limitations included passage-example-based exploration."],"publisher":"University of California, Berkeley School of Information / HCIR workshop authors","source_id":"SRC1","source_type":"PRIMARY_RESEARCH","title":"WordSeer: Exploring Language Use in Literary Text","url":"https://people.ischool.berkeley.edu/~hearst/papers/hcir2011.pdf"},{"claims_supported":["Scalable reading combines statistical analysis of an entire dataset with close interpretation of individual data points.","Its workflow uses distant analysis to select distinctive cases for close reading systematically and reproducibly.","The work is identified as peer-reviewed research published in Programming Historian."],"publisher":"Programming Historian (ProgHist)","source_id":"SRC2","source_type":"PRIMARY_RESEARCH","title":"Scalable Reading of Structured Data","url":"https://pure.au.dk/portal/en/publications/scalable-reading-of-structured-data/"},{"claims_supported":["Computational literary studies requires explicit corpus eligibility, scope, sampling, proportions, metadata, and reflection on representativeness, bias, and canonicity.","The guidance distinguishes designed and growing corpora and describes metadata-based monitoring.","Its literary examples operationalize variation by work, genre, period, publication history, and other corpus dimensions."],"publisher":"CLSInfra","source_id":"SRC3","source_type":"OFFICIAL_GUIDANCE","title":"Data Sharing – Planning and Designing Data","url":"https://data.clsinfra.io/3.1.Planning.html"},{"claims_supported":["Voyant Trends visualizes a term's distribution across a corpus or within a document.","Corpus mode distributes results across documents, while document mode displays within-document distributions.","The tool permits restriction to selected document identifiers and supports relative or raw distributions."],"publisher":"Voyant Tools / LINCS","source_id":"SRC4","source_type":"FIRST_PARTY_PRODUCT","title":"Voyant Tools Help: Trends","url":"https://voyant.lincsproject.ca/docs/Tools.Trends.html"}],"world_novelty_boundary":"This bounded public-web screen establishes only coarse researchability and adjacent prior art. It cannot establish world novelty, patentability, freedom to operate, market size, expert acceptance, realized scholarly value, or absence of relevant material in patents, paywalled books, non-indexed projects, other languages, or local scholarly practice."}