The human face and gaze direction elicit explicit attention in terms of the fixations that guide selective attention to visual information. We examine this possibility by tracking eye movements while participants twice viewed a video of a magic trick and demonstrate that participants fixate on the magician's face. During the first viewing, participant eye movements largely followed the objects that the magician had previously attended to. These results indicate that one's own fixations substantially guide the fixations of others. Even during the second viewing, when participants knew what to expect, they continued to be strongly influenced by the magician's gaze direction. This tendency was also observed even in the absence of explicit information concerning the magician's gaze direction due to wearing a mask. However, this bias gradually declined as the video continued. Accordingly, we argue that, when individuals must accurately infer the other person's intention to comprehend some video content, prior viewings are largely ineffectual in attenuating the other people's impact on controlling own fixation.

View full abstract

Pictures acquired through successive views are remembered briefly, but most are forgotten within a few seconds, suggesting that the scene detail is removed from the memory representation according to the time dependent function. However, previous researchers mimicked this situation by using the stimulus set of the pictures, which were in no context to each other. Thus, it remained unclear whether or not the scene context was critical for the memory decay. The present study examined whether context between scenes affects the memory decay by presenting the multiple different parts of a single picture in a rapid sequence. Furthermore, we assessed the recognition performance from two independent measure (i.e., sensitivity and criterion) by using signal detection theory. The results indicated that the lack of context between visual scenes affects the overall sensitivity but does not affect the time dependent function of the sensitivity. Furthermore, it affects the time dependent function of the criterion for judgment whether the scene was glimpsed previously or not. This suggested that the context between scenes do not affect the memory decay in the recognition sequence but affects the memory retrieval at decision-level stage in recall processing.

View full abstract

This study examines the development of drawing composition within young children, with particular focus on the compositions of noncanonical drawings in terms of spatial cognition and inhibition of canonicality as drawing-related cognitive factors. Examining the relationship between drawing and cognition, Experiment 1 presented four- to six-year-old participants with three tasks; a drawing task, a spatial cognitive task, and an inhibition task. The results indicated that spatial cognition and inhibition scores for a noncanonical drawing group were higher than those of a canonical drawing group. To explore the causal relationship between cognition and drawing, Experiment 2 provided cognitive training to the participants of the canonical drawing group in Experiment 1. The results indicated that children showing improved spatial cognition and inhibition scores were able to compose noncanonical drawings. These findings suggest that both spatial cognition and inhibition contribute to the development of drawing composition within young children.

View full abstract

The present study develops group-format reading span tests (RSTs) of Japanese and English for Japanese high-school students and investigates whether it is feasible to administer the tests to groups within reasonably short periods. Reliability coefficients for the newly-developed RSTs were high; for Japanese, α=.864 and for English, α=.875. Moreover, test scores were found to be normally distributed. These findings indicate that the RSTs are effective for group administration. Further analyses of correlation differences between Japanese-English RST scores found correlation coefficients of .677 and .531 for higher English-proficiency and lower English-proficiency students, respectively, while a prior study has reported a correlation coefficient of .84 for English-major students. These findings suggest that working memory in L2 reading functions more similarly to L1 reading when L2 proficiency is higher.

View full abstract

The present study examined the word length effects in order reconstruction tasks conducted immediately after presentation of a 6-word list and 14 seconds later. The results indicated that reconstruction was more accurate for short words than for long words in the immediate condition, whereas no word length effect was found in the delay condition. These results suggest that the phonological representations in order memory decayed in the delay condition. These findings are discussed in terms of a functional distinction between short-term and long-term memory systems.

View full abstract