EXPERIMENT IN SHORT-TERM MEMORY RECALL
Published by K & C Research Assistance, Inc.
Copyright © K & C Research Assistance, Inc., 1971
All rights reserved. Except for the quotation of short passages this publication may not be reproduced, stored in a retrieval system, or transmitted in any form or by any means withoutthe publisher's prior consent.
If you have any questions about this report, or if you would like us to write custom research for you, please contact our customer service department at 1-800-501-6465.
The short-term memory experiment to be described in this paper is an attempt to verify the Bieder-Szafran experiment, which is a variation of the Buschke experiment of 1962. Previous to the time of Buschke's experiment, short-term memory studies had been conducted according to the classical method of simply presenting a given amount of information to a subject, who was then asked to recall or recognize, either in oral or written manner, as much of the material presented as he could. G. A. Miller used this approach in the study in which he determined the memory span for humans to be seven, plus or minus two. Miller also stated that the accuracy and amount of recall depended not on the information content of the material, but instead on the number of items presented. This was in 1956.
Four years later, Averbach and Sperling, using a method of recall of only a specified portion of the stimuli presented, showed that both the number of items presented and the information content of the items had a bearing on the accuracy and amount of recall. These people used a tachistoscope. They discovered, more specifically, that, when 12 or fewer items were presented, the accuracy of recall was 90 per cent; but, when 16 items were presented, the accuracy fell to a surprisingly low 60 per cent. The latter is almost at chance level. In 1961, Shepard and Teghtsoonian showed that "correct recognition of repetitions of...