Computational modelling of serial memory

Current research work is directed towards developing our model of immediate serial recall (Primacy Model, Page & Norris, 1998).

This involves a good deal of experimental and theoretical work investigating aspects such as:

  • The effect of time on STM for verbal materials, and the adequacy of decay-based models to explain the same any such effect.
  • The relationship between verbal immediate serial recall and everyday speech production, with particular emphasis on the effects of phonological confusability (eg, repeated rhyme) on both.
  • The extension of our model to simulate long-term memory for order, in particular that seen in the so-called Hebb Effect (Hebb, 1961).
  • Theoretical aspects of the relationship between the Hebb effect and the long-term serial memory implicit in the acquisition of word-forms. Like all our theoretical work, this is complemented by an extensive experimental program that has recently been supported by an award of £278,000 to Mike Page and Graham Hitch.
  • The potential unification of models of verbal serial recall with models of word recognition from continuous speech. This work is conducted in collaboration with Dr Dennis Norris and our colleagues at the Max Planck Institute in Nijmegen (Dr James McQueen, Prof Anne Cutler).

Other current research interests include: errorless learning in clinical populations - a large experimental study has recently been completed and a project involving technological intervention for people with Alzheimer's disease is under development; and theoretical considerations relating to the use of brain scanning in cognitive psychology.