Do FRBs form inside neutron-star magnetospheres or in outer shocks?
In plain words
The radio flash could be made close to the neutron star, inside its cloud of magnetic field and plasma (the magnetosphere), or far out where an ejected shell hits surrounding gas.
Precise statement
Decide between emission at $r < \sim 1e10\,\mathrm{cm}$ inside a magnetosphere (coherent curvature or plasma emission) and emission at $r \sim 1e13\ \text{to}\ 1e15\,\mathrm{cm}$ by synchrotron maser emission at a relativistic magnetized shock. Discriminants: emission-region size from scintillation, polarization-angle swings within bursts, microsecond structure and spectral bandwidth. Answer: the site for each burst subclass, with the evidence.
What would settle it
Scintillation-based size limits and polarization-angle measurements on a sample of repeaters and one-off bursts, compared with the predictions of both models.
Status in the literature
Unverified note
2025: scintillation of FRB 20221022A limited its emission region to below about $3e9\,\mathrm{cm}$, and a pulsar-like polarization-angle swing of about 130 degrees was found in the same burst, favoring a magnetospheric origin for that event; whether this holds for other bursts is open.