Downloads: 333

Files in This Item:
File Description SizeFormat 
a0038825.pdf494.56 kBAdobe PDFView/Open
Title: Long-term phonological knowledge supports serial ordering in working memory.
Authors: Nakayama, Masataka  kyouindb  KAKEN_id  orcid (unconfirmed)
Tanida, Yuki
Saito, Satoru  kyouindb  KAKEN_id  orcid (unconfirmed)
Author's alias: 中山, 真孝
Issue Date: Sep-2015
Publisher: American Psychological Association
Journal title: Journal of experimental psychology. Learning, memory, and cognition
Volume: 41
Issue: 5
Start page: 1570
End page: 1578
Abstract: Serial ordering mechanisms have been investigated extensively in psychology and psycholinguistics. It has also been demonstrated repeatedly that long-term phonological knowledge contributes to serial ordering. However, the mechanisms that contribute to serial ordering have yet to be fully understood. To understand these mechanisms, we demonstrate 2 effects using triples of Japanese nonwords in immediate serial recall. One, a type of bielement frequency effect, is a retrograde compatibility effect. Bielement frequency effects are well-established phenomena whereby a 2-element sequence (e.g., "ka-re") that frequently appears in a language instantiates better recall of any sequence that includes this element (e.g., "ka-re-su-mo"). We demonstrate that bielement frequency affected both the first (e.g., "ka" for "ka-re"; retrograde compatibility effect) and second part of a sequence, indicating the existence of minicontext representations of 2-element sequences. The other effects are the position-element(s) frequency effects, whereby an element (e.g., the mora "ka") that more frequently appears in 1 position of a sequence (e.g., in the first mora of a word) than in other positions facilitates better recall of that element (i.e., the first mora). The effects demonstrated in this article indicate the long-term associations of position representations and elements. These effects are discussed in terms of the extensive learning hypothesis, which assumes that phonological structures are learned gradually. Implications for computational models are also discussed.
Rights: This article may not exactly replicate the final version published in the APA journal. It is not the copy of record.
This is not the published version. Please cite only the published version.
DOI(Published Version): 10.1037/a0038825
PubMed ID: 25730304
Appears in Collections:Journal Articles

Show full item record

Export to RefWorks

Export Format: 

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.