Behavioural evidence favours image psychology
Image psychology predicts image rotation and scanning in addition to explaining them. Propositional theory, by contrast, does not predict image rotation and scanning.
At best, a propropositional theory can be patched together to explain these phenomena, but only in an ad hoc way that could just as well explain opposite results.
Stephen Kosslyn, Stephen Pinker, George E. Smith and Steven P. Schwartz (1979).