LAB / WMSI-EXP-052
Long-Horizon Reanalysis of Dismissed Evidence.
Can a persistent research agent correctly reopen previously rejected leads when new evidence changes their status?
01 / QUESTION
Research Workflow and Meta-Experiments
Can a persistent research agent correctly reopen previously rejected leads when new evidence changes their status?
This is a protocol scaffold, not a completed experiment or a frozen execution plan. No measured effect or research conclusion is claimed here.
Hypothesis to test
The manipulation described below produces a reproducible, condition-specific change in the measured system behavior or internal state that is not explained by the ordinary controls.
Ordinary competing explanation
The observed pattern is explained by ordinary sampling variation, prompt/representation artifacts, surface lexical cues, evaluator bias, or the known product/runtime mechanics captured by the controls.
02 / PLANNED DESIGN
How we intend to test it
Seed a case with weak/rejected leads, then add discriminating evidence later.
Controls
No-reanalysis workflow; periodic full-review baseline.
What we would measure
Correct reopenings, false reopenings, time/cost, recall.
Access and instruments
Ordinary software or API access. This describes access requirements, not verified readiness on our Spark.
Research-agent harness; evidence ledger.
Before a full run
Feasibility=NOW; priority=P0. Before confirmatory execution, run a small pilot to estimate variance/cost, freeze the scoring rules, and assign a run-count target.
Model revisions, final prompts, sample sizes, cost limits, and scoring rules must be settled through the pilot and protocol-freeze process.
03 / INTERPRETATION
What could fool us
Source drift, unequal time budgets, evaluator knowing condition.
What would weaken the idea
If reanalysis creates more confirmation bias than recovered signal, the workflow is harmful.
Observation, inference, speculation, unknowns, and apparatus errors will be labeled separately. One striking transcript cannot establish a general effect.
04 / RECORD
Evidence we plan to preserve
Source corpus snapshot, task ledger, provenance, human interventions, model/tool logs, final reports.
Planned visual evidence
Lead-state timeline; precision/recall.
No research runs or raw datasets are published for this experiment. The September 26, 2026 registry and protocol compendium (version 1.0) are the sources for this planning record.
Back to all 90 experiments