TIC3 Abstracts

(41) Learning complete versus partial linear orders

Eriko Kawasaki (Kawamura Gakuen Woman’s University)

My hypothesis was that partially-ordered lattices (A>B, A>C, B>D) would be more difficult to learn than the comparable fully-ordered linear structures (A>B, B>C, C>D). The full order should be easier to learn because it permits subjects to use transitive inferences to fully order all elements along a unified line. In contrast, the partial order blocks a number of inferences involving the indeterminate case, and it thus prevents unification of the objects into one dimension. However, the result showed that the partial order was easy to learn probably because subjects neglected the indeterminate cases.

(42) The Relationship between Category Learning and Verbal Coding

Daisuke Tanaka (Dept. of Psychology, Grad. Sch. of Humanities and Sociologies, Uni. of Tokyo)

This study examined the relationship between category learning and verbal coding. Previous researches have shown that objects in the same category become more similar to each other and objects in different categories become increasingly dissimilar through category learning. Using letter matrices as examplars of two arbitrary categories, similarity between two matrices was rated at three conditions, before category learning and after category learning with or without articulatory suppression. The result suggests subjects store learned categories using verbal coding.

(43) Prototype abstraction in memory and the course categorization process

Takashi Ueda (Waseda University, Japan)

The exemplar view of categorization assumes we calculate the similarities between a target item and retrieved exemplars in memory, rather than category prototypes. Though numbers of evidences support the theory, prototype abstraction per se is not ruled out, and other evidences also suggest the use of prototypes. Then how do we think of the relationships between two types of representations in our memory system? In this presentation, the author proposed a “moving average prototype model” in which the use of exemplars is limited and updated through the learning process. The model was evaluated with ordinary prototype and exemplar models.

(44) Why does the tie effect occur ?: Structural interference approach to single-digit addition and multiplication

Hideaki Shimada (University of Tsukuba)

Arithmetic tie problems like 3＋3=? or 6×6=? can be solved faster than non-ties. The present experiment, in which participants solved a simple verification task, showed that while tie-mul-add condition problems (tie multiplication problems with addition answers, e.g., 3×3=6) were rejected faster than non-tie-mul-adds (e.g., 3×4=7), tie-add-muls (e.g., 3＋3=9) were rejected as fast as nontie-add-muls (e.g., 3＋4=12). This results suggest that automatic activation for tie problems is stronger than for nontie problems in addition, but not in multiplication. According to structural interference model, in the case of addition, the tie effect occur because of the difference of strength of automatic activation. Whereas, in multiplication, the tie effect occur because of the difference of the number of false-candidate answers, which interfere with true answer.

(45) The effect of articulatory suppression in the Tower of London task

Jun Ishikawa (University of Tokyo)
Yohtaro Takano (University of Tokyo)

The Tower of London task was used to investigate the interference of articulatory suppression. The dual task condition and the difficulty of the phase were manipulated as independent variable. As was observed in the previous study, the interference of articulatory suppression was not observed in the EASY phase condition. In the DIFFICULT phase condition, although the interference of articulatory suppression was observed in the Experiment 1, this pattern of result was not observed in the subsequent experiments.

(46) The Enactment Effect in Alzheimer's Disease Patients

Kouhei Masumoto (Osaka University, Japan)
Tsuneo Takai (Yodogawa Christian Hospital)
Tetsuo Kashiwagi (Osaka University , Japan)
Satoru Tsuneto (Osaka University , Japan)

We examined four tasks: verbal tasks (VTs), verbal tasks/object (VTs/O), subject-performed tasks (SPTs) and experimenter-performed tasks (EPTs), in 15 young adults, 13 normal older adults and 10 the mild to moderately impaired Alzheimer's disease (AD) patients. Consequently, in every group, enactment effect and object presentation enhanced recall performance. Moreover, it also became clear that the enactment effect reduced the error in recall for every group. These results indicate AD patients can benefit from enactment at encoding no less than young adults and normal older adults. (This study was the Sasakawa Scientific Research Grant from The Japan Science Society.)

(47) The effects of recency on the short-term serial memory for subject-performed tasks

Hiroyuki Kanashiki (Research Fellow of the Japan Society for the Promotion of Science (Kansai University), Japan)

In subject-performed tasks (SPTs) the participants are usually presented with verbal action phrases, such as “raise your arm”, and then they are required to perform the denoted actions. Compared to verbal tasks (VTs), with standard verbal learning conditions, SPTs show a greatly enhanced memory performance. Moreover, Zimmer, Helstrup, and Engelkamp (2000) suggested that the recency effects was more extended in SPTs than VTs in the free recall task. The aim of this paper was to examine such recency effects on SPTs in the serial recall task. In two experiments, each participant learned nine action phrases, and recalled forward order (Experiment 1), or recalled backward order (Experiment 2). These results indicated that SPTs enhanced the memory of recency items than VTs in the forward recall task.

(48) Working memory for speech pitch

Akihiro Tanaka (University of Tokyo, Japan)

In Baddeley’s working memory model, auditory-verbal information is considered to be held in the phonological loop. However, previous studies have focused mainly on segmental information of verbal memory. In this study, working memory for suprasegmental information, especially for pitch, in speech was examined. A dual-task experiment was conducted, in which subject was required to articulate irrelevant verbal material during short-term retention of auditory-verbal stimuli which include phonological or non-phonological pitch change. The results suggest that phonological and non-phonological pitch informations are retained separately.

(49) Role of working memory in text comprehension: Individual differences in storage and retrieval

Yukiko Nishizaki (Osaka University of Foreign Studies, Japan)
Mariko Osaka (Osaka University of Foreign Studies, Japan)

This study addressed the relationship between the reading span test (RST) and the text comprehension performances from the view of storage and retrieval systems. The effects of the serial recall task (Experiment 1) and the word fluency task (Experiment 2) on the performances of the text comprehension between high and low RST groups were investigated with a dual-task method. The results showed that the text comprehension of low RST group was influenced by the serial recall task, in contrast that of high RST group by the word fluency task.

(50) What Kind of Visuo-Spacial Task is Affected by Affective Valence in Working Memory?

Fumiko Gotoh (Doctoral Program in Psychology, University of Tsukuba, Japan) Nobuo Ohta (University of Tsukuba, Japan)

The effect of affective valenced words for visuo-spacial tasks was investigated. In the present study two kinds of visuo-spacial tasks were examined: a locational relation task between 2 letters, and a one-letter rotation task. Participants were asked to perform these tasks and to process affective valeneced kanji words at the same time. Affective valence was manipulated by affective kanji words that were negative, positive or neutral. Results showed that there were more errors for locational tasks with negative valenced words than for neutral words of marginal significance. In the rotation task, however, no differences were shown for errors of task in terms of affective valenced words. These results suggest that tasks that involved more work were most affected by affective valence in working memory.

[RETURN to TIC3 PROGRAM INDEX] | [NEXT]