Additional data for distinguishing behavioral multiscale and standard SEIR models

Identify which additional data sets are needed to determine whether population-level disease trajectories are better described by a standard SEIR-like model without explicit behavior change or by a multiscale behavioral model.

Background

The paper notes that epidemic curves generated by the symptom-score-driven multiscale model can closely resemble those of a standard SEIR model, creating an identifiability problem: population-level disease trajectories alone may not reveal whether illness-driven behavioral change is present.

The authors specifically point to symptom-severity and corresponding behavioral-survey data as possible supplementary observations beyond the within-host viral-load and population-level case-count data commonly used for fitting multiscale models. Determining the informative value of such additional, potentially sparse data is left unresolved.