Despite the fact that mental arithmetic is based on only a few hundred basic facts and some simple algorithms, humans have a difficult time mastering the subject, and even experienced individuals make mistakes. Associative multiplication, the process of doing multiplication by memory without the use of rules or algorithms, is especially problematic. Humans exhibit certain characteristic phenomena in performing associative multiplications, both in the type of error and in the error frequency. We propose a model for the process of associative multiplication, and compare its performance in both these phenomena with data from normal humans and from the model proposed by Anderson et al (1994).