Constrained prose recall and the assessment of long-term forgetting: The case of ageing and the Crimes Test

Alan Baddeley, Bruce Rawlings, Amie Hayes

Research output: Contribution to journalArticlepeer-review


It has become increasingly clear that some patients with apparently normal memory may subsequently show accelerated long-term forgetting (ALF), with dramatic loss when retested. We describe a constrained prose recall task that attempts to lay the foundations for a test suitable for detecting ALF sensitively and economically. Instead of the usual narrative structure of prose recall tests, it employs a matrix structure involving four episodes, each describing a minor crime, with each crime involving the binding into a coherent episode of a specified range of features, involving the victim, the crime, the criminal and the location, allowing a total of 80 different probed recall questions to be generated. These are used to create four equivalent 20-item tests, three of which are used in the study. After a single verbal presentation, young and elderly participants were tested on three occasions, immediately, and by telephone after a delay of 6 weeks, and at one of a varied range of intermediate points. The groups were approximately matched on immediate test; both showed systematic forgetting which was particularly marked in the elderly. We suggest that constrained prose recall has considerable potential for the study of long-term forgetting.
Original languageEnglish
Pages (from-to)1052-1059
Number of pages8
Issue number8
Early online date3 Dec 2013
Publication statusPublished - 2014

Bibliographical note

© 2013 Taylor & Francis.

Cite this