{"schema_version":1,"research_id":"eoa_inverse_innovation_exp05_external_evaluation_20260803","source_assessment_id":"negative_space_design__rhetoric:P3:v0","cell_id":"negative_space_design__rhetoric","search_queries":["site:gov.uk corrections social media communications policy correction official","research social media corrections visibility subsequent posts misinformation correction effect study","institutional correction policy social media pin correction best practice official","FCC emergency communications social media accessibility official required updates","official newsroom corrections policy social media correction prominence pin post","site:bbc.com editorial guidelines corrections social media accuracy correction","IPSO corrections prominence guidance official correction social media","site:help.x.com pin post profile official","study social media correction visibility feed subsequent posts prominence correction social media","correction prominence social media post same platform study visibility buried feed","social media post visibility decays over time research Twitter half life primary study","organizational social media correction after error audience trust experiment","official social media management pause suspend scheduled posts crisis mode Hootsuite","Sprout Social pause scheduled posts official pause all posts","Buffer pause queue scheduled social media posts official","social media crisis protocol pause scheduled posts official guidance"],"sources":[{"source_id":"S1","title":"Prominence in digital media guidance","publisher":"Independent Press Standards Organisation","url":"https://www.ipso.co.uk/resources/prominence-guidance-digital/","source_class":"GOVERNMENT_OR_REGULATOR","publication_date":"2024-01-24","accessed_at":"2026-08-03","claims_supported":["Editors must correct significant inaccuracies promptly and with due prominence.","Digital and social-media prominence is complex, and readers may not see corrections placed elsewhere.","A social-media inaccuracy generally requires a correction on the same platform and account.","Editors control institutional social accounts and are identifiable authorizers for remedial publication."]},{"source_id":"S2","title":"Corrections & Clarifications","publisher":"Australian Broadcasting Corporation","url":"https://www.abc.net.au/edpols/corrections-and-clarifications/13644514","source_class":"OFFICIAL_GUIDANCE","publication_date":"undated","accessed_at":"2026-08-03","claims_supported":["Senior editorial managers and policy advisers authorize specified corrective actions.","Social-media corrections normally use a new comment or post, subject to platform conventions and legal or editorial constraints.","Corrections should be placed where the affected audience is likely to encounter them.","Corrective repetition can be inappropriate when it recreates harm or offence."]},{"source_id":"S3","title":"The Corrections Dilemma: Media Retractions Increase Belief Accuracy But Decrease Trust","publisher":"Cambridge University Press for the American Political Science Association","url":"https://www.cambridge.org/core/journals/journal-of-experimental-political-science/article/corrections-dilemma-media-retractions-increase-belief-accuracy-but-decrease-trust/A6A9A16D97F22E1CA22B051D215E752B","source_class":"PRIMARY_RESEARCH","publication_date":"2023-03-17","accessed_at":"2026-08-03","claims_supported":["A randomized simulated-Twitter-feed experiment with 2,862 analyzed respondents found that media self-retractions improved belief accuracy.","Retractions produced a smaller adverse effect on outlet trust.","Mock social feeds are an established research method for testing correction exposure and interpretation.","The study did not test a quiet interval after a correction."]},{"source_id":"S4","title":"The Half-Life of a Tweet","publisher":"arXiv","url":"https://arxiv.org/abs/2302.09654","source_class":"PRIMARY_RESEARCH","publication_date":"2023-02-19","accessed_at":"2026-08-03","claims_supported":["Repeated impression-count observations indicated a median tweet half-life of about 79.5 minutes.","Approximately 95% of sampled tweets received no relevant additional impressions after 24 hours.","Social-post attention is temporally concentrated, supporting concern about rapid loss of visibility but not isolating displacement caused by an institution's own subsequent posts."]},{"source_id":"S5","title":"Pausing your queue","publisher":"Buffer","url":"https://support.buffer.com/article/515-pausing-your-queue","source_class":"OFFICIAL_PRODUCT_DOCUMENTATION","publication_date":"2026-04-29","accessed_at":"2026-08-03","claims_supported":["A channel operator can pause all queued and scheduled posts on a per-channel basis.","Paused posts remain recoverable and resume in their original order or can be rescheduled.","The product does not show a paused-queue indication in its calendar, illustrating an operator-state visibility risk.","The principal queuing operation in the proposal is already technically available."]},{"source_id":"S6","title":"During a Crisis: Social Media Strategy Guidelines","publisher":"United States Department of Defense","url":"https://media.defense.gov/2021/Aug/02/2002820335/-1/-1/1/DURING%20A%20CRISIS-%20SOCIAL%20MEDIA%20STRATEGY%20GUIDELINES.PDF","source_class":"OFFICIAL_GUIDANCE","publication_date":"2024-12 (last verified)","accessed_at":"2026-08-03","claims_supported":["Official public-affairs guidance already recommends pausing scheduled posts across platforms during a crisis.","The guidance pairs the pause with audience monitoring, defined team roles, approval coordination, and continuing updates.","This is close procedural prior art, although its trigger is a crisis and sensitivity risk rather than a completed factual correction."]},{"source_id":"S7","title":"How to customize your profile","publisher":"X Corp.","url":"https://help.x.com/en/managing-your-account/how-to-customize-your-profile","source_class":"OFFICIAL_PRODUCT_DOCUMENTATION","publication_date":"undated","accessed_at":"2026-08-03","claims_supported":["An account operator can pin a post so it is the first post profile visitors see.","Pinning is an implementable baseline and close rival to suppressing subsequent routine posts.","Pinning guarantees profile placement, not delivery in followers' algorithmic feeds."]},{"source_id":"S8","title":"Telling the Story: News Values and Principles","publisher":"The Associated Press","url":"https://www.ap.org/about/news-values-and-principles/telling-the-story/","source_class":"OFFICIAL_GUIDANCE","publication_date":"undated","accessed_at":"2026-08-03","claims_supported":["Serious errors must be escalated to supervisory editors and corrected as soon as possible.","Corrections must be visible to consumers and explicitly labeled.","AP uses channel-appropriate correction and retransmission practices rather than a documented post-correction publication freeze.","Existing correction governance supplies identifiable decision authority but not expressed demand for a landing window."]}],"problem_evidence":{"support":"MODERATE","rationale":"The underlying problem matters: correction exposure can improve belief accuracy, digital prominence is an acknowledged regulatory problem, and social-post impressions decay rapidly. However, no opened source measured how often institutions themselves displace corrections with routine posts or whether those subsequent posts causally change interpretation. The exact prevalence claim therefore remains an evidence gap.","source_ids":["S1","S3","S4"]},"stakeholder_evidence":{"support":"MODERATE","rationale":"IPSO-regulated editors, ABC senior editorial managers, AP supervisory editors, and institutional public-affairs teams are identifiable adopters or authorizers with expressed needs for prompt, visible corrections and controlled publication during crises. None expresses demand for a correction-specific quiet window, so stakeholder pull is inferred rather than directly demonstrated.","source_ids":["S1","S2","S6","S8"]},"prior_art":{"proximity":"ADJACENT_PRIOR_ART","closest_analogues":[{"name":"Crisis-triggered pause of scheduled institutional posts","similarity":"The Department of Defense already advises pausing scheduled posts, monitoring audiences, coordinating approvals, and continuing appropriate crisis communication.","remaining_difference":"The proposal narrows the trigger to an approved factual correction, predeclares a maximum interval, adds a public status cue and explicit urgent-message exemptions, preserves replies, and tests interpretation.","source_ids":["S6"]},{"name":"Recoverable per-channel queue pause","similarity":"Buffer already pauses scheduled and queued posts per channel and restores them afterward.","remaining_difference":"Buffer supplies the operation but not correction classification, accuracy-owner approval, an external status cue, emergency override roles, contextual rereview, or evaluation criteria.","source_ids":["S5"]},{"name":"Same-platform prominent correction plus pinning","similarity":"IPSO requires same-platform prominence, and X lets operators pin a correction above later profile posts.","remaining_difference":"These practices preserve placement but do not deliberately prevent institution-originated routine messages from competing in the initial interval.","source_ids":["S1","S7"]},{"name":"Institutional correction workflows","similarity":"ABC and AP already require labeled, prompt, audience-visible corrections with editorial escalation.","remaining_difference":"Neither opened policy specifies a bounded correction-triggered suspension of otherwise legitimate routine publishing.","source_ids":["S2","S8"]}],"distinctive_claim_remaining":"Compared with a pinned, properly labeled correction followed by the normal institutional queue, a correction-triggered, visibly signaled, maximum-four-hour pause of recoverable routine outbound posts will increase recipients' unaided identification of the corrected claim and reduce perceived tonal displacement without reducing perceived availability of urgent help or necessary context.","confidence":"MODERATE"},"implementation_evidence":{"support":"STRONG","rationale":"Core operations—pausing and restoring queues, per-channel control, pinning, editorial escalation, approval roles, audience monitoring, and crisis publication controls—are documented capabilities or practices. Institution-specific legal duties, platform coverage, accessibility review, exemption latency, and staff behavior under real pressure remain untested. A mock-feed and tabletop first step avoids interruption of live public-information channels.","source_ids":["S2","S5","S6","S7","S8"]},"scores":{"meaningful_impact":{"score":3,"rationale":"Corrections improve belief accuracy and visibility matters, but the incremental effect of silence beyond a prominent pinned correction is unknown.","source_ids":["S1","S3"]},"stakeholder_pull":{"score":3,"rationale":"Formal correction owners and publication authorizers exist, but no source requests this specific protocol.","source_ids":["S1","S2","S8"]},"incremental_advantage":{"score":2,"rationale":"Queue pauses, crisis freezes, pinning, and correction governance already exist; the unverified advantage is their correction-specific combination and measured quiet interval.","source_ids":["S5","S6","S7"]},"distinctiveness_plausibility":{"score":3,"rationale":"No direct match was found for the full correction-triggered protocol, but it is a narrow recombination of established practices and should not be treated as world novelty.","source_ids":["S1","S5","S6","S7"]},"technical_implementability":{"score":5,"rationale":"Commercial queue controls and native pinning make the mock and limited deployment technically straightforward.","source_ids":["S5","S7"]},"adoption_authority_feasibility":{"score":4,"rationale":"News and institutional communication policies identify supervisory or senior editorial authority, although exemption ownership and duties must be localized.","source_ids":["S1","S2","S6","S8"]},"evidence_readiness":{"score":2,"rationale":"Adjacent evidence and a suitable mock-feed method exist, but exact problem prevalence, adopter demand, optimal duration, and incremental effect are unmeasured.","source_ids":["S3","S4"]},"safety_net_benefit":{"score":3,"rationale":"Recoverable queues, time bounds, monitoring, and urgent overrides reduce blackout risk, but the status cue and exemption performance have not been tested with affected audiences.","source_ids":["S5","S6"]},"scalability":{"score":4,"rationale":"The protocol uses common channel-management functions and reusable governance templates; multi-platform differences and distributed authority may limit uniform deployment.","source_ids":["S1","S2","S5","S7"]}},"score_confidence":"MODERATE","costs":{"first_evidence":{"band_2026_usd":"UNDER_10K","scope":"Build three matched mock feeds, recruit approximately 90 consenting participants plus 6 channel operators, run interpretation and tabletop tests, and analyze predefined feasibility outcomes.","confidence":"MODERATE","assumptions":["Existing survey and prototyping tools are available.","Participant compensation averages $35-$50.","One researcher, one communications practitioner, and limited accessibility review contribute roughly 60-100 total hours.","This is a feasibility and signal test, not a powered estimate of small effects."],"source_ids":["S3","S5","S7"]},"initial_deployment_startup":{"band_2026_usd":"10K_TO_50K","scope":"Localize the protocol for one institution and up to five owned channels, map authority and exemptions, configure queue controls, prepare accessible templates, conduct legal and service review, and train operators.","confidence":"MODERATE","assumptions":["The institution already owns accounts and uses native or commercial scheduling tools.","No custom platform integration is required.","The range represents staff and adviser time as resource-equivalent cost, not only cash spending."],"source_ids":["S2","S5","S6","S7","S8"]},"operational_launch":{"band_2026_usd":"10K_TO_50K","scope":"Conduct shadow drills, instrument audit records and metrics, support the first several authorized uses, and review overrides and queued posts before broader rollout.","confidence":"LOW","assumptions":["Launch is limited to one communications unit and five or fewer channels.","No high-risk live use occurs until tabletop and mock-feed criteria are met.","Legal, accessibility, safety, and service owners participate in launch review."],"source_ids":["S2","S5","S6"]},"annual_recurring":{"band_2026_usd":"10K_TO_50K","scope":"Maintain templates and role rosters, train new operators, run two drills, review each activation and override, and cover incremental scheduler, monitoring, accessibility, and audit labor for one institution.","confidence":"LOW","assumptions":["Fewer than 20 qualifying corrections occur annually.","Existing communications staff and tooling absorb most execution.","High-volume, multinational, regulated, or round-the-clock operations could exceed this band."],"source_ids":["S1","S2","S5","S6","S8"]}},"verified_pipeline_gates":{"externally_supported_problem":{"status":"UNCERTAIN","reason":"Sources establish that correction prominence and rapid visibility decay matter, but not the prevalence or causal consequence of institutions posting routine content immediately after their own corrections.","source_ids":["S1","S3","S4"]},"externally_credible_adopter_or_authorizer":{"status":"YES","reason":"Regulated editors, senior editorial managers, supervisory editors, and public-affairs teams have documented authority over corrections or scheduled publication, although none has requested this exact intervention.","source_ids":["S1","S2","S6","S8"]},"distinct_testable_incremental_claim":{"status":"YES","reason":"The landing window can be compared directly with a pinned correction followed by normal posting using correction identification, interpretation, context, trust, and urgent-help measures.","source_ids":["S3","S5","S7"]},"bounded_next_evidence_step":{"status":"YES","reason":"Matched mock feeds and a tabletop operator exercise can test interpretation and workflow without modifying a live channel.","source_ids":["S3","S5","S6","S7"]},"no_unresolved_safety_or_authority_stop":{"status":"YES","reason":"The proposed first test is simulated; live use can remain prohibited pending localized authority mapping, urgent-message overrides, accessibility review, and stop criteria.","source_ids":["S2","S6","S8"]},"credible_cost_scope_and_range":{"status":"YES","reason":"The four bands are tied to explicit participant, channel, staffing, tooling, and activation assumptions. Confidence is reduced where no direct price source was used.","source_ids":["S3","S5","S6","S7"]}},"next_evidence_step":"Create three matched mock feeds containing the same fictional error, complete correction, evidence link, urgent-help route, and routine queue: (A) correction pinned with normal posting, (B) correction repeatedly promoted without a pause, and (C) correction pinned with a four-hour-equivalent landing window and status cue. Randomly assign approximately 90 consenting participants, 30 per condition, and measure unaided identification of the corrected claim, missing-context errors, perceived accountability versus evasion, tonal collision, trust, and perceived availability of urgent help. Separately, have six operators classify a mixed queue, invoke one urgent override, resume publication, and record disagreement, latency, and audit failures. Do not proceed to live use if condition C fails to improve correction identification directionally over A, produces materially more outage/evasion interpretations, causes any operator to withhold the simulated urgent message, loses queued content, or makes required context less recoverable. Treat results as feasibility evidence rather than a definitive impact estimate.","blocking_evidence":["No direct observational estimate of how often institutional corrections are immediately displaced by the institution's own routine posts.","No head-to-head evidence that a quiet interval outperforms pinning with normal publication or repeated promotion.","No adopter has expressed demand for a correction-specific landing window.","The optimal duration and proportionality rule by correction severity and normal channel cadence are unknown.","Audience interpretation of the status cue, especially as evasion, theatrical remorse, outage, or unavailable support, is unknown.","Urgent-message classification accuracy and override latency under real operating pressure are unknown.","Institution-specific legal, records-retention, accessibility, labor, and service-continuity requirements remain unmapped."],"research_disposition":"PARTNERED_RESEARCH_PROGRAM","world_novelty_boundary":"The bounded search found no direct source specifying the complete correction-triggered, maximum-four-hour, publicly signaled pause with recoverable queues, open replies, urgent exemptions, and comparative interpretation testing. This absence is not evidence of world novelty. Patentability, freedom to operate, market size, realized impact, and exhaustive global practice were not evaluated.","arm":"COMPLETE_PROPOSAL_PORTFOLIO","candidate_version":0,"controller_recommendation":{"action":"STOP_EMPIRICAL_RESEARCH_NEEDED","repairable":false,"material_progress_observed":true,"progress_targets":["Obtain direct prevalence evidence from owned-channel logs showing whether and how quickly routine posts follow substantive corrections.","Secure a named institutional communications or editorial partner willing to authorize a simulated test and articulate adoption criteria.","Complete the three-condition recipient experiment with predefined comprehension, context, trust, and urgent-help outcomes.","Demonstrate error-free urgent override, queue preservation, resumption, and audit logging in tabletop exercises.","Localize legal, accessibility, service-continuity, records-retention, and authority requirements before any live activation.","Estimate duration-response effects before treating four hours as a default rather than an experimental maximum."],"reason":"Web research establishes that correction prominence matters and that the operational components already exist, but it does not establish the exact problem's prevalence or the landing window's incremental benefit and safety. Those questions require channel-log analysis, participant testing, partner authority, and workflow exercises rather than further bounded web search."},"proposal_index":3}