A similarity between amnesic memory and normal forgetting

Abstract

Warrington and Weiskrantz reported a significant interaction between method of testing and subject group when amnesics' and controls' memory was tested after a 1 min retention interval. An attempt was made to duplicate this finding without using any amnesic subjects but substituting for them a group of normal subjects with a presumably weak memory trace and comparing them with a group of normal subjects with a presumably strong memory trace. The attempt succeeded. Each subject learned a list of 100 words and was then tested on half the words after 1 min and on the other half after one week. A separate group of six subjects was used for each of the following methods of testing: yes-no recognition; cueing with initial letters; cueing with a fragmented form of the words. The statistic corresponding to the interaction between method of testing and retention interval was significant. Also normal performance on yes-no recognition at one week was (like amnesic performance significantly inferior to normal performance at 1 min; and normal performance on each of the methods of partial information at one week (like amnesic performance) did not differ significantly from normal performance at 1 min. Certain reservations and implications are discussed but it is concluded that Warrington and Weiskrantz's experiment can no longer be regarded as providing support for the theory that the amnesic syndrome is limited to a defect of retrieval.