Abstract
Two sets of simulations were carried out in order to bring forward evidence on the two basic theoretical approaches about the well-known implicit/explicit memory dissociations: different memory systems (see e.g. Schacter et al., 1993: Squire et al, 1993) versus one memory store or "processing view" (see e.g. Graf and Ryan, 1990; Roediger, 1990). In a multi-layer back-propagation models we simulated: 1) the anterograde amnesia effect (see Warrington and Weiskrantz, 1970, or why patients with anterograde amnesia show similar results as normal patients in implicit tasks, whereas they show poorer performance in recall/recognition tasks); and 2) the type of processing made over the stimuli (semantic vs perceptual; see Graf and Mandler, 1984) that seems to affect only to explicit tasks, not to the implicit ones. Results support the processing view in the sense that these experimental dissociations can be explained as result of two different ways of processing the contextual information. Morever this latter theoretical approach seems to be more parsimonious than the different memory systems approach.
