Memory codes and episodes in models of word identification: a reply to Johnston, van Santen, and Hale

We suggest that the activation model of identification benefits for repeated words and pseudowords proposed by Johnston, van Santen, and Hale (1985) is a variant of our own code/episode model (Salasoo, Shiffrin, & Feustel, 1985), used to explain the temporary and long-lasting effects of repetitions. In particular, Johnston et al.'s X and Y factors may reflect the operation of episodic memory traces and codification, respectively. Computational processing models, we believe, are useful because their precision helps clarify otherwise fuzzy theoretical distinctions.