Multiple Ways of Understanding Evidence Accumulation Models: A Comment.

Shiffrin et al. (2026) explore "illusions of understanding" in science, primarily focusing on linear regression. This comment extends their ideas to Evidence Accumulation Models (EAMs) in cognitive science. I examine four distinct ways EAMs can be understood: as cognitive processes, neural mechanisms, statistical descriptors, and choice rules. I argue that while EAMs may foster illusions of mechanistic transparency, they remain vital "common languages" for bridging behavioral and neural levels of analysis.