Rebecca Dreier

The debate about episodic memory’s evolutionary function has given rise to three prominent views. The Mnemonic Function View (MFV) argues that episodic memory is for remembering the Future Function View (FFV) argues that episodic memory is for simulating the future and the Communicative Function View (CFV) argues that episodic memory is for communication. In the current debate, these views are presented as incompatible and sole functions of episodic memory. While some authors mention the possibility that episodic memory has multiple functions, a systematic account of how these functions might be related is currently lacking. In this paper, I argue that episodic memory could have evolved for all three functions, despite the prima facie tensions between them. I further show that the mnemonic function is episodic memory’s most basic or primary function. This is because the future and communicative functions depend, both metaphysically and historically, on the existence of an episodic memory capacity with the mnemonic function.

Episodic memories can fail to represent the past. We often forget, misremember, or make up histories we have never experienced. Yet according to the mnemonic function, episodic memory evolved for encoding, storing, and retrieving _tolerably_ accurate representations of the past to use in retroactive and unrestricted learning (Boyle, 2019, 2022 ; Brown, 2024 ; Schwartz, 2020 ). This appears to entail that episodic memory malfunctions regularly (De Brigard, 2014 ), whenever episodic memory is less than tolerably accurate. In this paper, I reject the intuitive inference from a trait’s failure to achieve its selected effect to a malfunction. Instead, I argue that inaccurate memories are often failures without being malfunctions. I exemplify this by considering the misinformation effect and the false memory paradigm. These are two common and extreme ways of misremembering, often cited to show the malleability of episodic memory (Loftus, 2005 ). However, I show that the best explanation of these memory errors is that they are “misfunctions” rather than malfunctions (Millikan, 2017 ; Sullivan-Bissett, 2024 ). To properly show that this is the best explanation of these memory errors, I exclude various alternative explanations. While these alternatives do not provide a good explanation of these two memory errors, their discussion is of vital importance for the aim of the paper. This is because they highlight that “mere failures to perform” (Sullivan-Bissett, 2024, p. 4) can be produced by a variety of causes. Thus, many failures of episodic memory to achieve an accurate representation of the past are not malfunctions.

The debate about episodic memory’s evolutionary function has given rise to three prominent views. The Mnemonic Function View (MFV) argues that episodic memory is for remembering the Future Function View (FFV) argues that episodic memory is for simulating the future and the Communicative Function View (CFV) argues that episodic memory is for communication. In the current debate, these views are presented as incompatible and sole functions of episodic memory. While some authors mention the possibility that episodic memory has multiple functions, a systematic account of how these functions might be related is currently lacking. In this paper, I argue that episodic memory could have evolved for all three functions, despite the prima facie tensions between them. I further show that the mnemonic function is episodic memory’s most basic or primary function. This is because the future and communicative functions depend, both metaphysically and historically, on the existence of an episodic memory capacity with the mnemonic function.

Court cases involving sexual assault and police violence rely heavily on victim testimony. We consider what we call the “Traumatic Untrustworthiness Argument (TUA)” according to which we should be skeptical about victim testimony because people are particularly liable to misremember traumatic events. The TUA is not obviously based in mere distrust of women, people of color, disabled people, poor people, etc. Rather, it seeks to justify skepticism on epistemic and empirical grounds. We consider how the TUA might appeal to the psychology and neuroscience of memory for empirical support. However, we argue that neither support the TUA.