再生と再認の記憶における検索属性

多鹿 秀継

doi:10.4992/jjpsy.51.259

RETRIEVAL ATTRIBUTES IN RECALL AND RECOGNITION MEMORY

HIDETSUGU TAJIKA

Author information

JOURNAL FREE ACCESS

1980 Volume 51 Issue 5 Pages 259-266

DOI https://doi.org/10.4992/jjpsy.51.259

Details

Published: January 20, 1981 Received: February 01, 1980 Available on J-STAGE: July 16, 2010 Accepted: - Advance online publication: - Revised: -
Correction information
Date of correction: July 16, 2010 Reason for correction: - Correction: ABSTRACT Details: Wrong : Memory processes in recall and recognition have been interpreted in two different theories. One type of the theory, that is, the two stage theory of recall, thought of recall and recognition memory to be basically different. The other type of the theory suggested that memory processes in recall and recognition were not fundamentally different as to retrieving stored information. A number of recent experiments revealed that recall and recognition represented similar retrieval processes. The data reported in the present study have focussed on the retrieval attributes, which were generally assumed to influence performances on recall and recognition memory.
The 142 undergraduate students were tested for two sessions over a week. Two kinds of lists were used in each task. Eleven different tasks were employed as retrieval measures. These tasks were as follows: (1) Subjects learned serial lists and were tested with free-recall method (SL-FR). (2) Subjects learned paired-associate lists and were tested with free recall method (PAL-FR). (3) Subjects learned serial lists and were given categorical cued recall tests (SL-CA). (4) Subjects learned serial lists and were tested with one initial letter of each item serving as cue (SL-1LT). (5) Subjects learned serial lists and were tested with two initial letters of each item (SL-2LT). (6) Subjects learned serial lists and were tested with all three initial letters of each item (SL-3LT). (7) Subjects learned pairedassociate lists and were tested with stimulus items in a usual way (PAL-CO). (8) Subjects learned paired-associate lists and were tested with extralist cues (PAL-EX). (9) Subjects learned serial lists and were given two-forced-choice recognition tests (SL-2FC). (10) Subjects learned serial lists and were given three-forced-choice recognition tests (SL-3FC). (11) Subjects learned serial lists and were given four-forced-choice recognition tests (SL-4FC).
The reliability estimates of the 11 variables were very high and most of them were intercorrelated. The correlations among the 11 variables for 142 subjects were summarized in Table 2. Initial letter cue tasks were more strongly related to recognition tasks. The principal factor analysis of the correlation matrix of the 11 variables and the Varimax rotation produced two factors. The results were shown in Table 3. Variables having high factor loadings in Factor I were (4) SL-1LT, (5) SL-2LT, (6) SL-3LT, (9) SL-2FC, (10) SL-3FC, and (11) SL-4FC. This factor was associated with the retrieval attribute relating to performance on tests of the explicit targets. Factor II showed high estimates of factor loadings in (1) SL-FR, (2) PAL-FR, (3) SL-CA, (4) SL-1LT, and (7) PAL-CO. This was associated with the retrieval attribute relevant to performance of generating the targets implicitly. These results cast doubts on the earlier theories referring to recall and recognition memory.
Right : The purpose of the present study was to examine the retrieval attributes which were generally assumed to influence performances on recall and recognition memory. The 142 undergraduate students were tested for two sessions over a week and 11 different tests were employed as retrieval measures. The results showed that the reliability estimates of the 11 variables were high and most of them were intercor-related. When the principal factor analysis and the Varimax method were computed upon the correlation matrix of the 11 variables, two factors emerged. One factor was associated with the retrieval attribute relating to performance on tests of the explicit targets and the other factor to performance in generating the targets implicitly. These results cast doubts on the earlier theories of recall and recognition memory.

Abstract

The purpose of the present study was to examine the retrieval attributes which were generally assumed to influence performances on recall and recognition memory. The 142 undergraduate students were tested for two sessions over a week and 11 different tests were employed as retrieval measures. The results showed that the reliability estimates of the 11 variables were high and most of them were intercor-related. When the principal factor analysis and the Varimax method were computed upon the correlation matrix of the 11 variables, two factors emerged. One factor was associated with the retrieval attribute relating to performance on tests of the explicit targets and the other factor to performance in generating the targets implicitly. These results cast doubts on the earlier theories of recall and recognition memory.

Corresponding author

Correction information

Register with J-STAGE for free!