{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp04_retrieval_first_paired20_20260802","cell_id":"negative_space_design__tech_ethics_ai_governance","hypothesis_id":"H4","search_queries":["adaptive decluttering safety critical display aviation temporary hide controls operator performance","emergency mode interface declutter noncritical controls human factors display","patent context sensitive display decluttering safety critical controls","human AI decision support interface cognitive forcing override wrong recommendation simulation","site:nrc.gov NUREG 0700 declutter noncritical information display human system interface","dynamic adaptive interface emergency display hide nonessential information safety critical operator study","site:ntrs.nasa.gov 20250009504 Human-Machine Interface High-Stress Environments","site:mdpi.com/2226-4310/8/3/71 safety risks small unmanned aircraft HMI"],"sources":[{"source_id":"S1","title":"Appendix F: Display Standard","publisher":"NASA","url":"https://www.nasa.gov/reference/appendix-f-vol-2/","source_class":"STANDARD","claims_supported":["NASA directs designers to limit displayed information to what the task and situation awareness require while presenting task-needed information without additional navigation.","The standard requires continuously visible key information, distinct mode presentation, automation-state indication, access to automation inputs and configurations, and consequence information before override or shutdown."]},{"source_id":"S2","title":"US20240011796A1 — Display Systems and Methods for Aircraft","publisher":"Bombardier Inc. via Google Patents","url":"https://patents.google.com/patent/US20240011796A1/en","source_class":"COMMERCIAL_FIRST_PARTY","claims_supported":["The patent discloses condition-triggered decluttering during time-critical flight by moving, removing, replacing, compressing, or fading displayed information to expose essential visual cues.","It discloses immediate reversion from the decluttered display to the prior or normal layout when operational triggers such as go-around or landing occur."]},{"source_id":"S3","title":"To Trust or to Think: Cognitive Forcing Functions Can Reduce Overreliance on AI in AI-Assisted Decision-Making","publisher":"Harvard Intelligent Interactive Systems Group","url":"https://www.eecs.harvard.edu/~kgajos/papers/2021/bucinca2021trust.shtml","source_class":"PRIMARY_RESEARCH","claims_supported":["A controlled experiment with 199 participants found that interface interventions can reduce acceptance of incorrect AI recommendations.","The intervention effects varied with users' cognitive motivation and involved usability tradeoffs, supporting subgroup and unintended-effect testing rather than relying on aggregate override speed alone."]},{"source_id":"S4","title":"Inhibitory Cognitive Control Allows Automated Advice to Improve Accuracy While Minimizing Misuse","publisher":"Psychological Science / SAGE Publications","url":"https://journals.sagepub.com/doi/10.1177/09567976211012676","source_class":"PRIMARY_RESEARCH","claims_supported":["In a timed simulated air-traffic-control task, incorrect automated advice impaired both accuracy and response time.","The experiment measured responses to correct and incorrect automation across 24 participants and 2,400 trials per participant, closely paralleling the hypothesis's concern with timely correction of bad automated advice."]}],"proximity":"SUBSTANTIAL_COLLISION","closest_analogues":[{"name":"Bombardier condition-triggered aircraft-display decluttering and reversion","similarity":"Very high mechanism-level similarity: during a time-critical safety task, the display conditionally suppresses, compresses, moves, or fades competing information and immediately restores the normal layout when a reversion trigger occurs.","remaining_difference":"It serves approach and landing cues rather than accept-reject-override decisions about AI recommendations, and it does not report the proposed randomized rehearsal outcomes.","source_ids":["S2"]},{"name":"NASA task-focused crew-display and automation-interface requirements","similarity":"High guardrail similarity: restrict information to task need, preserve continuously needed key information, expose automation status and rationale, distinguish modes, and support informed override or shutdown.","remaining_difference":"The standard does not itself specify the nominated transient chrome-hiding treatment or establish a 10% correct-override-time benefit in AI-decision rehearsals.","source_ids":["S1"]},{"name":"AI cognitive-forcing interfaces for reducing overreliance","similarity":"High problem and evaluation similarity: interface design is experimentally manipulated to improve rejection of wrong AI advice, with attention to heterogeneous effects and usability costs.","remaining_difference":"The interventions force deliberation rather than create reversible negative space by hiding only noncritical interface chrome.","source_ids":["S3"]},{"name":"Timed automated-advice failure simulation in air-traffic control","similarity":"High workflow similarity: operators make speeded safety-relevant decisions with imperfect automated advice, and researchers jointly measure response time and accuracy when advice is wrong.","remaining_difference":"The experiment does not compare a decluttered focus mode against a full interface or test control restoration and evidence-omission failures.","source_ids":["S4"]}],"overlapping_components":["Condition- or phase-triggered temporary decluttering","Removal, compression, movement, or fading of secondary display information","Protection of task-critical and continuously monitored information","Explicit automation and mode-status presentation","Human override or shutdown authority","Immediate restoration of the normal display through a reversion trigger","Timed simulation of incorrect automated advice","Joint accuracy and response-time measurement","Testing for intervention tradeoffs and heterogeneous user effects"],"remaining_contrastive_claim":"In timed rehearsals of high-risk AI-assisted decisions, temporarily hiding only noncritical chrome while continuously exposing evidence, uncertainty, override, and emergency controls—and restoring all hidden controls instantly on demand—improves correct-override time without evidence-omission or recovery failures, a complete empirical combination not found in this bounded search.","claim_falsifier":"The contrastive claim would be falsified by a pre-existing study, deployed product, patent, or binding interface specification that applies reversible noncritical-control suppression to AI-recommendation override scenarios and evaluates correct-override latency together with evidence inspection, mode awareness, and control recovery; the treatment claim itself would also fail if a powered randomized simulation does not achieve the stated 10% improvement or worsens either safety-error measure.","problem_support":"STRONG","recommendation":"RESEARCH","world_novelty_boundary":"The novelty boundary excludes task-focused displays, safety-critical decluttering, protected essential information, automation override, instant layout reversion, and timed wrong-advice simulations individually or in ordinary combinations; only the narrow AI-governance rehearsal configuration and its joint latency, evidence-retention, and recovery test remain potentially distinctive."}