{"schema_version":1,"assessment_id":"eoa_inverse_innovation_exp03_opportunity320_20260801","source_experiment_id":"eoa_inverse_innovation_exp03_full320_20260801","cell_id":"invariant_mode_decomposition_design__archaeology_paleontology","archetype_slug":"invariant_mode_decomposition_design","domain_slug":"archaeology_paleontology","title":"Modal decomposition of coupled taphonomic distortion","opportunity_summary":"Test whether a scoped transition operator and SVD-derived modes can correct consequential joint preservation and recovery distortions more effectively than direct comparison or a hierarchical category-level model, while exposing residual distortion and instability. Effectiveness, demand, transportability, and novelty remain unverified.","adopter_authorizer":"A project research lead and responsible repository could authorize a read-only analytical pilot; sampling or interpretive changes additionally require applicable heritage, repository, and community governance.","scores":{"meaningful_impact":{"score":3,"rationale":"Avoiding preservation-driven errors in behavioral, ecological, or mortality reconstruction could be consequential, but the packet supplies no evidence about problem prevalence or realized improvement."},"stakeholder_pull":{"score":2,"rationale":"Relevant researchers, field staff, repositories, and affected communities are identified, but no adopter request, workflow dissatisfaction, willingness to participate, or resource commitment is reported."},"incremental_advantage":{"score":3,"rationale":"Joint modal correction could capture structured cross-category distortion missed by direct comparison and category-level hierarchical models, but no blinded result yet demonstrates superior reconstruction or abundance-contrast recovery."},"distinctiveness_plausibility":{"score":3,"rationale":"Outcome-sensitive selection of paired transition modes, residual checks, and drift limits form a coherent incremental claim, but prior art is explicitly unsearched and literature-level distinctiveness cannot be inferred."},"technical_implementability":{"score":3,"rationale":"The read-only analytical workflow is computationally plausible and has explicit diagnostics, but known inputs may be unavailable and sparse compositional data, nonlinearity, path dependence, and nonidentifiable gains could prevent a usable operator."},"adoption_authority_feasibility":{"score":4,"rationale":"Authority for an analytical pilot is mapped to the research lead and repository, with consequential actions reserved for existing governance; feasibility is favorable although actual authorization has not been obtained."},"evidence_readiness":{"score":3,"rationale":"The proposal supplies comparators, metrics, falsifiers, halt criteria, and a held-out design, but it does not identify an available known-input dataset, a qualified partner, sample adequacy, or preregistered thresholds."},"safety_net_benefit":{"score":4,"rationale":"A read-only pilot, explicit exclusions, uncertainty-adjusted retention, residual and drift monitoring, and rollback to baseline substantially limit harm, though erroneous correction of genuine structure remains possible."},"scalability":{"score":2,"rationale":"Each regime may require defensible known inputs and separate validation, while site, protocol, and preservation differences threaten transfer; no reusable operator or cross-context performance is established."}},"score_confidence":"MODERATE","costs":{"first_evidence":{"band_2026_usd":"50K_TO_250K","scope":"Preregister, curate, analyze, and independently review one known-input dataset with one blinded held-out context, comparing the baseline, hierarchical rival, and modal method.","confidence":"LOW","assumptions":["An existing dataset can be accessed without new destructive or extensive fieldwork.","Costs include domain and statistical labor, data harmonization, governance coordination, computation, and evaluation.","Specimen-level records require material quality control but no major digitization campaign."]},"initial_deployment_startup":{"band_2026_usd":"50K_TO_250K","scope":"Convert a successful pilot into a documented reproducible pipeline for one bounded research program, including validation thresholds, reporting, training, and repository workflow coordination.","confidence":"LOW","assumptions":["Pilot code and curated data are reusable.","Deployment remains analytical and does not alter excavation, sampling, or specimen disposition.","One repository or project environment is supported initially."]},"operational_launch":{"band_2026_usd":"250K_TO_1M","scope":"Validate and launch the method across several compatible contexts or projects with independent review, governance, training, monitoring, and integration into interpretation workflows.","confidence":"LOW","assumptions":["Multiple contexts have adequate calibration information and compatible records.","Separate transfer and drift testing is required for each represented regime.","No new destructive experiments or large-scale field recovery program is included."]},"annual_recurring":{"band_2026_usd":"50K_TO_250K","scope":"Maintain an active multi-project analytical service, including data curation, reruns, uncertainty review, mode-drift monitoring, governance documentation, software maintenance, and analyst support.","confidence":"LOW","assumptions":["A modest number of projects are assessed annually.","Major new actualistic experiments and broad repository digitization are excluded.","Human review remains necessary before interpretive use."]}},"research_burden":"HIGH","earliest_credible_horizon":"3_TO_12_MONTHS","pipeline_gates":{"recognizable_externally_supportable_problem":{"status":"YES","reason":"The candidate specifies a recognizable mechanism, observable structured residuals, consequential reconstruction errors, and a test that could falsify coupled distortion in a scoped setting; prevalence remains unmeasured."},"identifiable_adopter_or_authorizer":{"status":"YES","reason":"The project research lead and responsible repository are explicitly identified as potential analytical-pilot authorizers, with broader governance identified for consequential changes."},"distinct_testable_incremental_claim":{"status":"YES","reason":"The modal method claims better blinded reconstruction and prespecified abundance-contrast recovery than both direct/category-specific correction and a hierarchical category-level rival, within stated uncertainty."},"bounded_next_evidence_step":{"status":"YES","reason":"A preregistered read-only study on one known-input dataset and one held-out context is bounded, safe, comparative, and linked to explicit failure criteria."},"no_unresolved_safety_or_authority_stop":{"status":"UNCERTAIN","reason":"The analytical design has exclusions and rollback, but the packet does not establish that repository access and all applicable heritage or community governance are actually present."},"implementation_cost_scope_and_range":{"status":"YES","reason":"The one-dataset read-only pilot and excluded actions provide enough scope for broad resource-equivalent bands, although dataset condition, access, and partner requirements make confidence low."}},"blocking_evidence":["Availability and governance-approved access to a defensible known-input dataset and genuinely held-out context.","Evidence that coupled models outperform category-independent recovery models and leave less consequential residual structure.","Blinded superiority of the modal method over both the baseline and hierarchical rival within prespecified uncertainty.","Stability and identifiability of retained modes and gains across resamples, compositional treatments, and the held-out context.","Evidence delimiting transfer across sites, protocols, and preservation regimes.","Scoped prior-art findings sufficient to delimit any distinctive contribution.","Direct evidence of adopter demand and willingness to support the required analytical workflow."],"next_evidence_step":"With a repository and domain partner, preregister a read-only analysis of one accessible actualistic or otherwise defensible known-input dataset: fit on a calibration portion, blind one held-out context, and compare the ordinary baseline, hierarchical category-level rival, and modal method on reconstruction error, residual structure, and recovery of prespecified abundance contrasts. Reject the problem claim if coupled models add no held-out predictive value; reject the intervention if the modal method fails to outperform both comparators within uncertainty or if retained gaps and modes are unstable.","research_questions":["Which accessible datasets provide defensible deposited or earlier-stage inputs, adequate category coverage, and a genuinely held-out context?","Does repeatable cross-category residual structure remain after category-independent and hierarchical recovery modeling?","Are transition gains and retained modes identifiable and stable under resampling, compositional alternatives, sparse-category handling, and plausible model misspecification?","Does the modal method improve blinded reconstruction and prespecified abundance contrasts over both comparators without suppressing genuine behavioral or ecological structure?","How sharply must validity be bounded by site, material, protocol, and taphonomic regime?","What existing multivariate taphonomy, inverse preservation, latent-factor, compositional, and transition-model approaches overlap with the proposed contribution?","Which repository, heritage, and community approvals are required even for read-only analysis and later interpretation?","What level of workflow burden and evidentiary gain would make research leads and repositories willing to adopt the method?"] ,"recommendation":"PARTNERED_RESEARCH","uncertainty_constraints":["Closed-book assessment provides no basis for claims about prevalence, market size, adopter demand, prior art, or world novelty.","The deposited state is usually unobserved, so known-input calibration availability is uncertain.","Path dependence, nonlinearity, sampling design, and assemblage covariance may make a local operator or causal modal interpretation invalid.","Cost bands are resource-equivalent planning ranges based on the bounded activities described, not observed quotations or exact estimates.","No performance, transportability, governance approval, or operational adoption evidence is present."],"closed_book_prior_art_boundary":"The packet supports only a structurally coherent and testable hypothesis. Prior art is unsearched, so no claim is made that SVD-based transition modes, multivariate taphonomic correction, their components, or their proposed combination are novel, uncommon, or superior in existing practice."}