ABSTRACT
The tenns implicit and explicit memory are used in two logically different ways; both to classify different memory tasks and to describe the memory processes underlying these tasks. Confusion arises when theories at the process level are tested with reference to data collected at the task level, without specifying the relation between processes and tasks. The view that implicit memory tasks draw exclusively on implicit memory processes, and explicit tasks on explicit processes (the Transparency Assumption), is shown to be false, since implicit and explicit tasks appear to share at least some processes in common. We conclude that theories must describe the fonn of the relation between hypothetical processes and overt task perfonnance in order to be testable.
