SOPhiA 2021

Salzburgiense Concilium Omnibus Philosophis Analyticis

SOPhiA ToolsDE-pageEN-page

Programme - Talk

Episodic Memory and Probabilistic Causation
(Philosophy of Mind, )

A Markus Werning recently brought probabilistic causation up to the debate about episodic memory arguing that if one defends that episodic memory is reliable then one should accept that there is some kind of causal connection between the episodic retrieval and the perceptual experience remembered. His reasoning is basically this: in order to be reliable, the perceptual experience should raise the probability of an episodic memory happening (Reliable Production). And he considers a causal connection to be one that raises the probability of an event happening (Reichenbach's Common Cause Principle). I will defend that this conclusion is problematic for two main reasons: first, establishing a causal connection is not the only option if someone considers episodic memory to be reliable. In Pearl's probabilistic causation approach, the alternative would be to consider the relationship between episodic retrieval and perceptual experience as one of dependence. An event x is dependent on an event y if when y happens it changes the probability of x happening. A causal connection implies that there is a dependence relation, but not all dependence relation implies a causal connection, as Werning concludes. And a dependence relation could also be considered epistemically reliable (Reliable Production), even if there is no causal connection between episodic retrieval and the perceptual experience remembered. Second, Werning's proposal of minimal traces fits better as a dependence (but not causal) relation than as a causal relation. To defend my argument, I will begin my talk by presenting the idea that to be reliable, episodic memory should be causally connected to the perceptual experience. Then, I will expose the problems of that argument through Pearl's account of probabilistic causation. After that, I will defend that accounts like the minimal traces or procedural causation fit better with a dependence relation. Thereafter, I will introduce the necessity of a probabilistic active path to a relation be considered as a causal one. Finally, I will argue that as far as we know from neuroscience studies, episodic memory does not have a probabilistic active path, therefore, should be considered as having a dependence (but not causal) relation to the perceptual experience.

Chair: Martin Niederl
Time: 19:30-20:00, 09 September 2021 (Thursday)
Location: SR 1.005
Remark: (Online Talk)

José Carlos Camillo  
(Universidade Federal de Goiás, )

Testability and Meaning deco