The thing I never understood was why the Aspen sensor would over-read even in flight, and a ~5C error was commonly reported.
One would think the airflow would remove any temp rise due to solar radiation.
OTOH most light GA planes don't have a lot of internal "padding" so if the OAT is say -10C and the cockpit is say +20C, it will take only a little bit of heat loss from the cockpit (unavoidable) to lift the roof temperature.