Scientific record
No methodology or model architecture is adopted. Concept formation precedes selecting a computational implementation.
Questions before protocol design
- Define productive spontaneous internally generated cognition and distinguish it from deliberate imagination, unproductive mind-wandering, generic spontaneous thought, generative search, and stochastic model behavior.
- Specify what counts as internal generation without immediate external task demand, and how spontaneity, productivity, and external conditions would be represented and evaluated.
- Determine whether observable tasks or process evidence could test the phenomenon without equating random or diverse output with the capability.
- Identify conditions, comparison cases, data requirements, and evaluation criteria only after the proposed distinction can be justified.
- State limits on any inference from computational behavior to a human cognitive capability.
These are unresolved design questions, not an approved protocol. No data collection, benchmark, metric, experiment, or training plan is selected here. If an operationalization is adopted, update the research question and status, record the reasoning, and register subsequent experiments with the required reproducibility details.