Our brains have an extraordinarily large capacity to store and recognize complex patterns after only one or a very few exposures to each item. Existing computational learning algorithms fall short of accounting for these properties of human memory; they either require a great many learning iterations, or they can do one-shot learning but suffer from very poor capacity. In this paper, we explore one approach to improving the capacity of simple Hebbian pattern associators: adding hidden units. We propose a deterministic algorithm for choosing good target states for the hidden layer. In assessing performance of the model, we argue that it is critical to examine both increased stability and increased basin size of the attractor around each stored pattern. Our algorithm achieves both, thereby improving the network's capacity to recall noisy patterns. Further, the hidden layer helps to cushion the network from interference effects as the memory is overloaded. Another technique, almost as effective, is to "soft-clamp" the input layer during retrieval. Finally, we discuss other approaches to improving memory capacity, as well the relation between our model and extant models of the hippocampal system.

We propose a new view of word priming in attractor networks, which involves deepening the basins of attraction for primed words. In a network that maps from orthographic to phonological word representations via semantics, this view of priming leads to novel predictions about the interactions between orthographically and/or semantically similar primes and targets, when compared on an orthographic versus a semantic retrieval task. W e confirm these predictions in computer simulations of long-term priming in a word recognition network. Connectionist models have strongly influenced current thinking about the nature of human memory storage and retrieval processes. One reason for their appeal is that they can account for a wide range of human performance on tasks such as word recognition (McClelland and Rumelhart, 1981), reading (Seidenberg and McClelland, 1989), and repetition priming (McClelland and Rumelhart, 1986). Further, because connectionist models make relatively specific assumptions about the mechanisms of cognitive processes, they can lead to novel predictions about human performance. One of the most exciting developments in the last decade of human memory research is the characterization of implicit memory (Graf &; Schacter, 1985; Schacter, 1985), a form of automatic, unconscious retrieval of previously encountered material. A widely used experimental method for testing implicit m e m ory is repetition priming, in which the accuracy or speed of processing is measured on successive presentations of a target stimulus. Evidence of implicit memory is observed when subjects are more accurate or efficient in responding to previously studied targets than to new targets. The priming literature is highly relevant to connectionist models of learning and memory for two reasons: 1) Priming effects can be extremely long-lasting, ranging from minutes to many hours, or even months, and apparently reflect fundamental automatic ("unsupervised") learning processes employed by the brain. 2) W h e n the prime and target are not identical, but have similar input and/or semantic features, the priming effects may range from facilitation to inhibition; these effects can shed light on the nature of h u m a n memory organization, and provide constraints on the representations employed in connectionist models. In this paper, w e first review the previous connectionist accounts of priming. W e then propose a new view of word priming in attractor networks with orthographic and semantic levels of representation, which involves deepening the basins of attraction for primed words. This leads to some novel predictions about the interactions between primes and targets, which we explore in computer simulations.