The Necessary Conditions for Implicit Memory of New Associations
Date Issued
2004
Date
2004
Author(s)
Han, Hui-Ya
DOI
zh-TW
Abstract
This study examined the conditions necessary for obtaining associative repetition priming effect (ARP) for unrelated pairs with an implicit word identification task. On the basis of the findings of a previous study (Uttl, Graf, & Cosentino, 2003), we proposed that stable representations of the new associations and pre-activation of the target’s representation are both necessary for implicit memory of new associations. Through a pilot study, we selected two sets of unrelated pairs by the ease of relating items in each pair. The difficult set was used as stimuli in Experiment 1 A and the easy set was used in Experiment 1B for the participants to rate their likings of the context item in each pair for 2 s and also generate a sentence that meaningfully related the two items in each pair in 15 s. In the test phase, the participants rated likings of the context item again and then identified the target item that was presented with threshold duration. The results showed ARP only with the easy set for both picture-word and word-word pairs. Recognition rates to intact pairs that contained the newly learned associations were higher than to recombined pairs, which in turns were higher than to new pairs reflecting the item repetition priming effect (IRP). Each context item was presented for 30 ms in the test phase of Experiments 2 and 3 with the easy set to change the test context, with a delay of 1970 ms to pre-activate the target item in the former and a delay of 30 ms to prevent pre-activation in the latter experiment. The results showed IRP in both experiments, and ARP manifested only with picture-word pairs in Experiment 2. With a brief presentation of the context item, 1970 ms was sufficient to pre-activate the target item in picture-word pairs perhaps because both visual and verbal links of a picture connected to the target word. However, the existence of two links became detrimental when the delay was 30 ms as the overall performance with picture-word pairs was worse than word-word pairs in Experiment 3. To further confirm that a stable representation of the newly learned associations is necessary, a matching task was used in Experiment 4 with the easy set of picture-word pairs while the test phase remained the same as that of Experiment 2. The results showed IRP only. In Experiment 5, we replicated the previous study to exclude the possibility that the results of Experiment 1 arose from a slight difference in methodology. The results are the same as in Experiment 1. The results from these five experiments suggested that sentence generation that meaningfully relates two unrelated items in the study phase is important for forming sable representations, but not sufficient for the manifestation of implicit memory of new associations. Consistent with the hypothesis, the stable representations of the newly acquired associations and the pre-activation of target items indeed are necessary for implicit memory of new associations between unrelated pairs after a single study trial.
Subjects
隱示記憶
新聯結
Implicit memory
New associations
Type
other
File(s)![Thumbnail Image]()
Loading...
Name
ntu-93-R90227014-1.pdf
Size
23.53 KB
Format
Adobe PDF
Checksum
(MD5):8aa215a03fd941452060ec5e524ba0d0