LAB / WMSI-EXP-036
Persistent Auditory World Model.
Can a continuously listening local system build persistent models of recurring sound sources and contexts?
01 / QUESTION
Auditory Experience and Social Conditioning
Can a continuously listening local system build persistent models of recurring sound sources and contexts?
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
Stream controlled audio environment over days and test source/context tracking.
Controls
Clip-only model; shuffled identity labels; quiet baseline.
What we would measure
Source tracking, recurrence memory, context prediction, false associations.
Access and instruments
Physical audio or robotics equipment. This describes access requirements, not verified readiness on our Spark.
Microphones; local ASR/audio embeddings; event DB.
Before a full run
Feasibility=HARDWARE; priority=P1. 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
Acoustic confounds, TTS artifacts, semantic leakage, speaker imbalance.
What would weaken the idea
If continuous history adds no value over isolated clips, persistent auditory modeling is weak.
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
Raw audio, source IDs, timestamps, transcripts, embeddings/features, model outputs and latencies.
Planned visual evidence
Event raster; source timeline; embedding trajectory.
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