IntroductionResearch on collocations has become an essential issue in L2 acquisition and cognitive psychology. Previous studies have mainly focused on phonographic languages such as English, Swedish, and German, and primarily discussed the effect of semantic transparency and translational congruency. However, these studies have lacked (1) an analysis of the interactions between presentation modalities (visual vs. auditory) and the semantic transparency and translational congruency, and (2) a discussion of an ideographic language, such as Chinese and Japanese.MethodsWe conducted an experiment with 36 Chinese Japanese-as-a-Foreign-Language learners to examine the processing of Japanese collocations. In the experiment, we manipulated the presentation modality, semantic transparency, and translational congruency during a lexical judgment task.ResultsData analysis using linear mixed-effects models revealed the following. (1) In both conditions of semantic transparency and translational congruency, the auditory presentation was associated with longer reaction times than the visual presentation. (2) In the visual presentation condition, neither semantic transparency nor translational congruency showed significant effects. (3) In the auditory presentation condition, the reaction time for collocations with high semantic transparency tended to be longer than that for collocations with medium semantic transparency and significantly longer than that for collocations with low semantic transparency. The reaction time for collocations with congruent translation was longer than that for collocations with incongruent translation.DiscussionThese results support the dual-route model of Japanese collocational processing by Chinese Japanese-as-a-Foreign-Language learners. Our findings suggest that whether the analytic or holistic processing dominates is closely related to the learners’ knowledge of Chinese and Japanese Kanji words and strongly influenced by the presentation modality, semantic transparency, and translational congruency.

No abstract

The present study investigated how visuospatial working memory (VSWM) is involved in the construction of a spatial situation model for spatial passages presented auditorily. A simple spatial tapping condition, a complex tapping condition as a target-tracking task, and a control condition, were used to analyze the role of VSWM. To understand how individuals who differ in verbal working memory (VWM) capacity (determined with a listening span test) process spatial text during dual-task performance, individual differences in VWM capacity were analyzed. In two experiments, the participants listened to a spatial text at the same time as performing a spatial concurrent task or no concurrent task. The results of the free recall test in Experiment 1 showed that there were no differences between the tapping conditions in the high VWM capacity group; the low VWM capacity group had a lower performance in both spatial tapping tasks compared to the control condition. The results of the map drawing test in Experiment 2 showed that complex spatial tapping impaired performance in comparison to simple spatial tapping and the control condition in the high VWM capacity group; in the low VWM capacity group, both spatial tapping tasks impaired recall performance. In addition, the participants with high VWM capacity demonstrated better performance. Overall, the results suggest that individuals with high VWM capacity have more resources to process verbal and spatial information than those with low VWM capacity, indicating that VWM capacity is related to the degree of the involvement of VSWM.

No abstract