This study assessed the holistic and fragmented retention and forgetting of event models. We report four experiments that manipulated causality, co-reference, events versus objects, and description determinacy. While increased causal connections among events increased holistic remembering, there was no clear effect for manipulations of co-reference, events versus objects, or determinacy. Thus, our work suggests that there are limits to the extent to which different types of events are remembered and forgotten in a holistic or fragmented manner. That said, all of our event did show significantly greater than chance holistic remembering, suggesting that the very act of creating event models leads these memories to be remembered or forgotten as wholes to a greater extent.