Towards a unified theory of false memory for similar episodes

Abstract Important advances have been made in understanding the roles that recollection and familiarity, true and false memory, as well as pattern separation and pattern completion, play in episodic memory, using experimental paradigms that examine memory for lures that are similar – but not identical – to prior episodes. These include the Process Dissociation Procedure, the Deese-Roediger-McDermott paradigm, and the Mnemonic Similarity Task. However, research using each of these paradigms has remained largely isolated, and a coherent theoretical integration is lacking. In this perspective paper, we argue that memory for similar lures in all these paradigms can be understood within a unified theory in which memory performance reflects the operation of three distinct processes: false recollection, false familiarity, and recollection rejection. Our proposed approach leverages the theoretical insights provided by examining recognition confidence ratings and analyzing false memory receiver operating characteristics. Using extant data from paradigms across all three literatures that collected recognition confidence judgments, we show that this analytic approach can be used to measure each process. Our findings bridge the existing literatures by showing that these three memory processes are functionally distinct and that they underlie performance in each experimental paradigm. This new approach overcomes key limitations of the earlier methods by separating common memory and decision processes across paradigms, and it provides new opportunities to further understand why factors such as healthy aging disrupt the ability to discriminate between similar episodes.