Episodic memory traces: Causal, content and epistemic aspects of the link between experience and recall

许多记忆理论家假设记忆痕迹是一种存储有关过去经历的事件的信息的方法，以便在以后的时间点上可以回忆起这些信息以便于记忆。在这种精神下，情节记忆的因果理论假定记忆痕迹不仅是经验和记忆之间的因果联系，而且是表征内容的保护者。该项目的主要目的一方面是批判因果理论所支持的记忆痕迹的保守主义概念，并在生成性情景记忆的框架内用记忆痕迹的最小生成性概念取而代之。另一方面，该项目也反对模拟理论提出的彻底否定记忆痕迹的观点，即认为记忆根本不需要通过记忆痕迹与经验相耦合，甚至不需要通过因果联系，而只是由对过去事件的模拟组成。在拒绝因果理论的同时，项目中将要发展的最小生成主义接受了因果理论的四个主要愿望：(I)记忆的内容应该与所经历的内容具有适当的内容关系。(Ii)记忆应该是关于过去的特定事件--而不仅仅是一种类型的事件--因此必须有参照物。(Iii)任何情节记忆理论都应该使情节记忆在认识论上是可靠的，因为情节记忆是知识的真正来源。(4)任何情景记忆理论都应避免重新学习的问题：曾经经历过的事件已经被遗忘，后来由于外部信息源的原因，相关信息被重新获得，不应被视为记忆的情况。最后，正如cheng&Werning(2016)提出并被几位作者采用的任何情节记忆理论的一个总体限制是，它应该对应于一种自然类型。也就是说，对情节记忆的描述应该解释为什么情节记忆的实例共享一组特征的属性，因为一些潜在的统一的因果机制。项目沿着六个工作包进行：在(WP1)中，我们将发展一个最小的生成性概念，即记忆痕迹作为没有表征内容的因果联系。在(WP2)中，我们将演示最小的记忆痕迹如何在经验和记忆之间提供适当的内容关系，并确保记忆的参照。我们将在(WP3)中展示最小记忆痕迹如何使情节记忆在认识论上可靠，并在(WP4)中展示最小生成主义如何避免重新学习的问题。在(WP5)中，我们将根据最小生成主义修改我们之前给出的关于情景记忆的分析--序列分析，并证明它成功地识别了一种自然类型。最后，在(WP6)中，我们将概述最小生成主义相对于因果理论和模拟理论的优势。

Many memory theorists have postulated memory traces as a means by which information about events experienced in the past is stored such that this information, at a later point in time, can be recalled for the purpose of remembering. In this spirit, the Causal Theory of Episodic Memory postulates memory traces not only as a causal link between experience and remembering, but also as a preserver of representational content. The main goal of the project, on the one hand, is to criticize the preservationist notion of memory traces, as defended by the Causal Theory, and to replace it with a minimal generative notion of memory traces in the framework of Generative Episodic Memory. On the other hand, the project also opposes the radical denial of memory traces proposed by the simulation theory, i.e., the view that remembering need not be coupled to experience through a memory trace at all, not even through a causal link, but merely consists of a simulation of a past episode.While rejecting the Causal Theory, Minimal Generationism, the view to be developed in the project, accepts four main desiderata of the Causal Theory: (i) What is remembered should bear an appropriate content relation to what was experienced. (ii) A remembering should be about a particular event in the past – not just a type of event – and thus must have reference. (iii) Any theory of episodic memory should render episodic memory epistemically reliable because episodic memory is a genuine source of knowledge. (iv) Any theory of episodic memory should avoid the problem of relearning: A situation where a once experienced event has been forgotten and later, due to an external information source, the relevant information is re-acquired should not count as a case of remembering. Finally, an overall constraint on any theory of episodic memory, as has been proposed by Cheng & Werning (2016) and taken up by several authors, is that it should correspond to a natural kind. That is, an account of episodic memory should explain why instances of episodic memory share a characteristic set of properties because of some underlying uniform causal mechanism.The projects proceeds along six work packages: In (WP1) we will develop a minimal generative notion of memory traces as causal links without representational content. In (WP2) we will demonstrate how minimal memory traces provide an appropriate content relation between experience and remembering and secure the reference of remembering. We will show in (WP3) how minimal memory traces render episodic memory epistemically reliable and in (WP4) how Minimal Generationism avoids the problem of relearning. In (WP5) we will revise a previous analysis we gave on episodic memory – the Sequence Analysis – in the light of Minimal Generationism and show that it succeeds in identifying a natural kind. Finally, in (WP6) we will outline the strengths of Minimal Generationism vis-à-vis the Causal Theory and the Simulation Theory.