ABSTRACT
A fault found in most of the current stock of computer simulations of human memory is that they have forgotten that people forget. In the few programs that do forget (e.g., Reitman, 1965), information loss is viewed as simple decay of the strength of past memories. The research on EPAM and SAL (Feigenbaum, 1963, 1970; Hintzman, 1968; Simon & Feigenbaum, 1964) has been the only attempt to produce simulation models that make contact with the basic facts of forgetting from long-term memory. However, these conceptions of memory are not adequate for the task of expressing the propositional character of memory.
