01
Portable measurement
We are investigating whether wearable brain imaging can make repeated, real-world measurement more useful alongside established lab methods.
Research lab / 21.3
What if visual experience left a signal?
Analog Memory Aug 21.3 is an early-stage research lab exploring wearable brain imaging and AI methods for searchable records of visual experience.
Working premise
Curiosity is a method. The long-term idea is ambitious; the near-term work is evidence, prototypes, and transparent limits.
Wearable direction
Signal trace / 01Model sketch
02The question
A memory is personal. So should be the model that studies it.
Can portable, repeated measurements make visual recall more useful?
Conventional imaging and controlled experiments remain important baselines. We are asking what becomes possible when neural measurement follows an individual over time and personalized models learn from that context.
This is not a promise of a complete thought or memory backup. It is a research program for determining what can be measured, reconstructed, and compared with care.
Evidence before extrapolation
03How we learn
Three ways to keep the question honest.
Each step narrows the gap between an evocative possibility and a result another team can inspect.
02
Personalized models
The models are meant to learn an individual’s neural representations, with signal quality and uncertainty kept visible rather than abstracted away.
03
Searchable experience
The near-term question is bounded: can selected visual experiences be reconstructed or retrieved well enough to evaluate against clear benchmarks?
04A staged program
Build evidence. Share openly.
The path is intentionally incremental. A useful result can be a clear limit as often as it is a successful reconstruction.
01
Measure
Collect and characterize neural signals in defined conditions.
02
Model
Train a personalized representation without hiding uncertainty.
03
Compare
Evaluate reconstruction or retrieval against explicit baselines and benchmarks.
05Research posture
The unknowns are part of the work.
Measure first
Every useful claim starts with a reproducible measurement and a defined experimental condition.
Make uncertainty legible
Reconstruction quality, signal limits, and comparison baselines belong in the result.
Keep people in control
Sensitive neural data requires clear consent, careful governance, and participant agency.
06Open invitation
Bring a hard question.
Research collaboration, technical feedback, and future study participation all begin with a careful conversation about the data, the method, and the limits.