Differences in performance on an implicit learning task--sequence learning--among normal young and aging adults and patients with Alzheimer's disease (AD) were investigated. All participants show evidence of learning, yet the patterns of learning appear to be quite dissimilar. Differences in performance are accounted for by differences in factors pertinent to resource supply such as efficiency and capacity of the participants. By varying its efficiency and capacity, a connectionist model based on the simple recurrent network architecture (Elman, 1990) is demonstrated to capture major aspects of the learning processes of different groups of participants, thus verify the resource supply account. Taken together, the results indicate that: (a) Mechanisms of implicit learning are quite resistant to aging and AD. (b) Implicit learning can be resource demanding. Effective learning will not occur if the resource demand exceeds the resource supply.
Keywords: Implicit learning, aging, Alzheimer's disease (AD), efficiency, capacity, simple recurrent network (SRN)