Why episodic memory may not be for communication.

Three serious challenges to Mahr & Csibra's (M&C's) proposal are presented. First, we argue that the epistemic attitude that they claim is unique to remembering also applies to some forms of imaginative simulations that aren't memories. Second, we argue that their account cannot accommodate critical neuropsychological evidence. Finally, we argue that their proposal looks unconvincing when compared to more parsimonious evolutionary accounts.