Computer security has always been an issue, more so in recent years due to global network access. In this paper, we present a simple connectionist algorithm for testing the quality of computer passwords. A popular method of evaluating password quality is to test it against a large dictionary of words and near-words. Our algorithm is an approximate realization of this method. The large dictionary of words is stored in a network in distributed form. All stored words are stable; however, spurious memories may develop. Although there is no easy way to determine exactly which non-word strings become spurious, nor even exactly how many spurious memories form, numerical simulations reveal that the network works well in distinguishing words and near-words from structureless strings. Thus, to evaluate a password, one would present it to the network and, if the network labeled it a memory, the password would be considered bad.

For a given recurrent neural network, a discrete-time model may have asymptotic dynamics different from the one of a related continuous-time model. In this article, we consider a discrete-time model that discretizes the continuous-time leaky integrat or model and study its parallel, sequential, block-sequential, and distributed dynamics for symmetric networks. We provide sufficient (and in many cases necessary) conditions for the discretized model to have the same cycle-free dynamics of the corresponding continuous-time model in symmetric networks.