Procedure
The experiment was performed with all participants under the intentional memory instruction. The participants were given standard instruction about the intentional free recall test.
Learning phase. — The task was explained with an illustration using PowerPoint slides displayed on a board at the front of a classroom. First, the participants were given the following instructions: “A Kanji word is shown in the upper part of each slide. The first task is to remember it. There is second task on each page [pointing to the example on the white board]. In this slide [assigned to the social elaboration condition], the word “person” is presented below a Kanji word. On this type of slide, your second task is to think about any person that you are reminded of by the word. In the other type of slide [assigned to the semantic elaboration condition], the word “word” is presented below a Kanji word. On this type of slide, your task is to think about any word (associated word) that you are reminded of by the word. ” Participants were given 5 seconds per slide to complete the first and second tasks.
Free recall test phase. Following the learning phase, the participants were required to recall as many of the targets (Kanji words) as possible and to write them down. Three min. were allowed for this test.
Checking phase. Following the free recall test, the participants were given a sheet with all the target words printed on it. Each participant was asked to indicate whether each target word reminded him or her of a person in the social elaboration condition or a word in the semantic elaboration condition. If a particular target word reminded the participant of a person or a word, he or she was asked to circle “generated” on the sheet; if it did not, the participant was asked to circle “not generated” on the sheet. The checking phase lasted 3 min.
Results
For the 14 targets in the social elaboration condition, the mean numbers of targets were 8.81 and 5.19 when the target reminded the participants of a parson (“generated”) or not (“not generated”), respectively. For the 14 targets in the semantic elaboration condition, the mean number of targets were 8.08 and 5.92 when the participants was reminded of a particular word (“generated”) or not (“not generated”), respectively. On the basis of these numbers, the percentages that the targets were correctly recalled was calculated. These percentages are shown in Table 2 as a function of the type of elaboration (social vs. semantic) and whether the corresponding information was generated (generated vs. not-generated). A 2 (social vs. semantic) x 2 (generated vs. not-generated) analysis of variance showed that main effects of type of elaboration (F 1,51= 17.01, p <.001, η2=.03) and whether the corresponding information was generated (F 1,51= 121.15, p < .01,η2=.39) were significant and that the interaction between elaboration type and whether the corresponding information was generated (F1,51= 3.12, p < .08, η2=.01) was marginally significant. Planned comparisons were performed for this interaction; although the simple effect of the elaboration type was not significant in the non-generated condition (F 1,102 = 1.61), it was significant in the generated condition (F1, 102 = 15.93, p< .001).
Table 2
Mean percentages of correct recall as a function of the type of elaboration (social vs. semantic) and whether the corresponding information was generated or not.