The Examination of Discriminative Function of Retrieval Cue in the Buildup of Proactive Interference

The purpose of this experiment was to examine the effects of discriminative function of retrieval cue on the buildup of proactive interference by using four category lists with different category similarity. A 4×3 factorial design was used, which incorporated list similarity conditions (Discriminative ・Dissimilar, Discriminative・Similar, Nondiscriminative・Dissimilar, Nondiscriminative ・Similar) and number of trials (1,2,3). The subjects were 60 students with a mean age of 19yr. and 6 mo. who were assingned to one of the four list conditions (see Table 1). Each list had three words. Under the Discriminative・Dissimilar(D-D) condition, the list had the words from different categories. Under the Discriminative・Similar(D-S) condition, the list had words from different subcategories in one higher level category. Under the Nondiscriminative ・Dissimilar(N-D) condition, the list had words from different subcategories in one category. Under the Nondiscriminative・Similar(N-S) condition, the list had the words from the same categories in the three trials. Using the Brown-Peterson paradigm, three words were presented simultaneously for 2 sec. rate, which was followed by the distractor task (20 sec). Then recall test was given 10 sec. After the 3rd trial was over, final free recall test was required to recall all items presented from 1st to 4th trials. The main results were as follows : on the performance of Brown-Peterson recall test (see Fig. 1), Under D-D condition, PI was not built up as a function of trials, and the performances of recall were at a high level and kept constant. Under N-S condition, PI was built up as a function of trials. Under D-S and N-D conditions, PI was built up as a function of trials and the performances of both conditions were about the same, but the degree of PI formation was smaller than N-S condition. Response time of first recall word increased as a function of trials under all list conditions. On the performance of final free recall test (see Fig. 2), the performances of recall under three conditions which PI was built up did not decrease as a function of trials. On the bases of above results, we concluded that the buildup of PI was due to the discriminative function of retrieval cues and this factor was infulenced by the degree of list similarity