Historically, mental imagery has been defined as an experiential state—as something necessarily conscious. But most behavioural or neuroimaging experiments on mental imagery—including the most famous ones—do not actually take the conscious experience of the subject into consideration. Further, recent research highlights that there are very few behavioural or neural differences between conscious and unconscious mental imagery. I argue that treating mental imagery as not necessarily conscious (as potentially unconscious) would bring much needed explanatory unification to mental imagery research. It would also help us to reassess some of the recent aphantasia findings inasmuch as at least some subjects with aphantasia would be best described as having unconscious mental imagery.This article is part of the theme issue ‘Offline perception: voluntary and spontaneous perceptual experiences without matching external stimulation’.