{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp04_retrieval_first_paired20_20260802","cell_id":"invariant_mode_decomposition_design__political_science","hypothesis_id":"H4","search_queries":["audit targeting interagency referral network centrality public sector accountability","complaint referral network eigenvector centrality inspector general audit selection","public administration network intervention centrality target nodes governance audit","responsibility shifting blame network centrality government agencies accountability","patent complaint routing network centrality audit risk government","\"key player problem\" network intervention centrality organizations","inspector general audit planning risk assessment official complaint referrals","complaint management software referral analytics bottleneck government oversight"],"sources":[{"source_id":"C1","title":"Network Interventions: Applying Network Science for Pragmatic Action in Public Administration and Policy","publisher":"arXiv","url":"https://arxiv.org/abs/2109.08197","source_class":"PRIMARY_RESEARCH","claims_supported":["Defines network interventions as purposeful use of network data to identify strategies for changing behavior or performance.","Explicitly transfers network-intervention concepts to public-sector interorganizational and governance networks.","Leaves the particular intervention objective, network construction, and outcome validation to the application."]},{"source_id":"C2","title":"El problema del actor clave","publisher":"Redes: Revista Hispana para el Análisis de Redes Sociales, Universitat Autònoma de Barcelona","url":"https://revistes.uab.cat/redes/article/view/v24-n2-borgatti","source_class":"PRIMARY_RESEARCH","claims_supported":["Formalizes selection of key nodes for network diffusion or disruption.","Shows that ordinary off-the-shelf centrality rankings need not identify the nodes whose removal best changes a network.","Closely anticipates the hypothesis's move from descriptive centrality to outcome-specific node perturbation."]},{"source_id":"C3","title":"The Network Organisation of Consumer Complaints","publisher":"arXiv","url":"https://arxiv.org/abs/1008.1815","source_class":"PRIMARY_RESEARCH","claims_supported":["Constructs a complaint network from Brazilian Department of Justice records.","Relates complaint-network structure to resolution and uses dimensional reduction on weighted complaint features.","Demonstrates that administrative complaint records were already being analyzed as a network, although not as interoffice responsibility transfers."]},{"source_id":"C4","title":"The Department Must Improve Processes for Addressing and Managing ‘H’ Referrals","publisher":"Oversight.gov / U.S. Department of Commerce Office of Inspector General","url":"https://www.oversight.gov/reports/department-must-improve-processes-addressing-and-managing-h-referrals","source_class":"GOVERNMENT_OR_REGULATOR","claims_supported":["Documents an operational population of OIG hotline complaints referred to departmental offices for inquiry and response.","Finds ineffective processes and internal controls over referrals.","Recommends Department-wide procedures, status tracking, prompt processing, objective review, and enforcement of response deadlines, directly supporting the nominated problem."]},{"source_id":"C5","title":"Audits and Evaluations FAQs","publisher":"U.S. Department of Commerce Office of Inspector General","url":"https://www.oig.doc.gov/about/faqs/faqs-audits-evaluations/","source_class":"OFFICIAL_GUIDANCE","claims_supported":["Describes annual risk assessment as the basis for audit planning.","Lists legal mandates, requests, management challenges, prior findings, and higher-risk programs as audit-selection factors.","Does not describe referral-network spectral rankings or node-perturbation simulations as established audit-selection practice."]},{"source_id":"C6","title":"The Anatomy of Blame: A Network Analysis of Strategic Responsibility-Shifting After a Systemic Disaster","publisher":"arXiv","url":"https://arxiv.org/abs/2510.21681","source_class":"PRIMARY_RESEARCH","claims_supported":["Models accusations among public, private, and regulatory organizations as a directed network.","Finds reciprocal and clustered responsibility shifting rather than isolated assignments.","Establishes a close conceptual precedent for treating accountability diffusion as network structure, but does not use administrative complaint referrals to select audits."]},{"source_id":"C7","title":"Improving Collaboration in the Public Sector Audit System in Brazil","publisher":"OECD","url":"https://www.oecd.org/en/publications/auditing-decentralised-policies-in-brazil_30023307-en/full-report/improving-collaboration-in-the-public-sector-audit-system-in-brazil_4f1436fc.html","source_class":"OFFICIAL_GUIDANCE","claims_supported":["Treats public audit institutions as a governance network in which responsibility allocation, information exchange, and coordinated audit selection require explicit design.","Describes risk matrices, integrated audit information, audit-selection committees, and cross-institutional sharing of potential audit topics.","Provides a professional-practice analogue for network-informed audit planning without prescribing spectral analysis of responsibility-transfer records."]}],"proximity":"SUBSTANTIAL_COLLISION","closest_analogues":[{"name":"Public-sector network-intervention framework","similarity":"Already proposes using governance-network data to identify nodes or structural strategies that improve public-sector performance.","remaining_difference":"It does not specify a longitudinal complaint-referral operator, responsibility-diffusion eigenmode, or prediction of unresolved cases for audit targeting.","source_ids":["C1"]},{"name":"Key Player Problem","similarity":"Already selects nodes by their counterfactual effect on a network rather than by raw degree or generic centrality, closely matching the proposed perturbation ranking.","remaining_difference":"Its canonical objectives are diffusion and fragmentation, not reduction of an empirically defined responsibility-transfer mode tied to audit outcomes.","source_ids":["C2"]},{"name":"Network analysis of complaint and blame structures","similarity":"Prior studies already turn complaint records or interorganizational responsibility assignments into networks and relate their structure to resolution or accountability diffusion.","remaining_difference":"The retrieved studies do not combine interoffice referral edges, modal participation, node intervention simulations, and prospective audit-target validation.","source_ids":["C3","C6"]},{"name":"Risk-based, coordinated public-sector audit selection","similarity":"Official practice already combines risk assessment, shared audit information, interinstitutional coordination, and referral tracking to choose or manage oversight work.","remaining_difference":"The opened guidance does not rank auditees by marginal reduction of a dominant referral-network mode or validate that ranking against future unresolved transfers.","source_ids":["C4","C5","C7"]}],"overlapping_components":["Directed organizational-network representation of complaints, referrals, or blame","Network-based identification of structurally important actors","Outcome-specific node removal or perturbation analysis","Public-sector governance-network intervention","Administrative complaint-network analysis","Risk-based audit planning","Cross-agency audit-information integration and coordinated target selection","Tracking of referral timeliness, status, and resolution"],"remaining_contrastive_claim":"In longitudinal agency referral networks, ranking audit targets by each node's marginal reduction of the leading responsibility-transfer mode predicts future unresolved or repeatedly transferred cases better than complaint volume, degree centrality, and existing risk scores.","claim_falsifier":"A preregistered temporal holdout across multiple oversight systems showing no reliable incremental predictive value over the named baselines—or an earlier disclosed system implementing the same referral-operator, modal-perturbation, audit-selection, and prospective-validation pipeline—would falsify the remaining claim.","problem_support":"STRONG","recommendation":"RESEARCH","world_novelty_boundary":"As a bounded ordinary-web search through 2026-08-02, the review found established complaint-network analysis, responsibility-shifting networks, key-player perturbation methods, public-sector network interventions, and network-informed risk-based audit planning; it did not find an opened research paper, official practice, product, or patent that implements the complete longitudinal referral-eigenmode and prospective audit-targeting pipeline, so novelty can attach only to that narrow integration and empirical performance claim, not to its component methods."}