Free associations with EPS and memory

In celebrating 50 years of EPS involvement in memory, I have taken a personal line rather than trying to achieve even coverage. Accordingly, the paper is in three parts. In the first, some early papers that have resonance for today's debates are described. Their framework was Bartlettian, with versions of schema theory as the guiding principles, and they were formative in my undergraduate education. This phase ended in the 1950s. From then, there was an explosion of work in memory in the U.K., the most important parts of which saw light at the EPS. This requires a proper history, which requires a proper space. In the second part, then, rather than show favourites or sample at random, my own involvement in EPS and memory is fallibly revisited, to give some flavour of the excitement, as well as some of the ideas. In the final part, some current work is described, and the Current State Buffer is introduced, in the hope of anticipating a little of the next 50 years.