Consolidation-like Effects in Flashbulb Memories: Evidence from September 11, 2001

Abstract

After September 11, 2001, we distributed flashbulb memory questionnaires at 5 different dates: within 48 hr (T1) and at 1 week (T2), 1 month (T3), 3 months (T4), and 1 year (T5). We scored responses for self-reported memory (veracity unverified), memory accuracy (recollection-matched T1 response), and memory consistency (recollection-matched prior responses other than T1). Self-reported memory and subjective confidence remained near ceiling, although the accuracy declined. However, memories given a week or more after September 11 were consistent throughout. We hypothesize that flashbulb memories follow a consolidation-like process: Some details learned later are incorporated into the initial memory, and many others are discarded. After this process, memories stabilize. Therefore, the best predictor of flashbulb memories at long intervals is not the memory as initially reported but memories reported a week or more after the event.