We conducted two experiments to specify the properties of the representation underlying perceptual priming. We transformed a particular property of stimuli between study and test. We used novel stimuli to eliminate contamination by semantic memory. They were rotated by 0°, 90°, or 180° in Experiment 1 and were distorted horizontally in Experiment 2. These transformations were chosen for topological structures of stimuli to be kept unchanged. In Experiment 1 the priming effect of the rotated stimuli was smaller than that of the identical ones. The rotational angles had no effect. More surprisingly, the horizontally distorted stimuli showed the same amount of priming as the identical ones in Experiment 2. In contrast, recognition performance of the rotated or horizontally distorted stimuli was lower than that of the identical ones. Statistical independence between priming and recognition was found in both experiments. Thus, recognition was statistically and functionally independent of priming. These results suggest that the priming representation encodes both orientation-independent and orientation-dependent information, and that the global topological structure is critical for priming.
Three experiments investigated the effects of identity of facial expression and person of two successively presented faces on face recognition. In each experiment, a combination of a familiar/unfamiliar person with neutral/smile expression was presented. Four different groups of 8 or 12 undergraduates were asked to judge the facial expression or the familiarity of the second face. The results showed that reaction times in both judgments were shorter when the same person repeatedly appeared than when two different people were presented. This repetition effect was not affected by facial expression of the stimulus for expression judgments, while it depended on the expression and familiarity of the first and second faces for familiarity judgments. In facial expression judgments, reaction times to smile faces were shorter than those to the neutral faces, only when subjects were familiar with those faces. This facial expression effect appeared only when the identical familiar and smile person was repeatedly presented for familiarity judgments. These results suggested the interdependency between analysis of facial expression and that of person information in face recognition processes.
The purpose of this study was to investigate developmental changes in Self-Recognition Need (SRN; Kamise, 1992), measuring the need of middle aged and elder people. One hundred and ninety eight (198) men and women in their forties, and 199 in their sixties filled a questionnaire that measured SRN. Results showed that SRN declined with advancing years, and it was higher in women than in men. Targets of SRN also changed with advancing years. Men in their forties and women in their sixties directed their SRN toward their interpersonal and future selves, women in their forties toward their current and future selves, and men in their sixties toward their past selves.
This study focused on the moderating effects of positive appraisal of family caregivers of older people on caregiving burden. In Study 1, 16 original items of the Caregiving Burden Scale and 14 original items of the Positive Appraisal Scale were administered 140 caregivers. By factor analysis, the Caregiving Burden Scale was divided into four subscales named: “Restriction of freedom (Kosoku-kan)”; “Wish to give up caregiving (Genkai-kan)”; “Family trouble (Taijin-katto)”; and “Economic restraint (Keizaiteki-futan)”. The positive appraisal scale was also divided into three subscales named “Caregiving satisfaction (Manzoku-kan)”, “Caregiver's self-growth (Jiko-seicho-kan)”, and “Will to continue-care-giving (Kaigo-keizoku-ishi)”. In Study 2, 306 caregivers completed the Caregiving Burden Scale and Positive Appraisal Scale. The results indicated that; (a) positive appraisal moderated caregiver's wish to give up caregiving (“Genkai-kan”); (b) Caregiving satisfaction (“Manzoku-kan”) was more effective than caregiver's self-growth (“Jiko-seicho-kan”); (c) moderating patterns varied by the combination of stressor, burden and positive appraisal.
We investigated the distortion that occurs during the storage of representations in visual short-term memory by using smooth closed figures with varying stimulus complexity and similarity. In similarity judgments, the results suggested that the psychological similarity between two figures was based on averaging physical similarities of matched features. In a detection experiment, the results showed that the processing of detection was limited in its capacity for the precision of shape, and that it was more difficult to detect the precise shape in the case of complex figures. In a recognition experiment, the results showed that forgetting occured during a one-second retention interval. These two results showed that internal noise distorts both detection and retention processes. These findings suggested that representations of the complex figures with many features were distorted to a larger degree because each feature was individually masked with internal noise. It is concluded that the psychological similarity between an input pattern and a distorted pattern stored in short-term memory is invariant with stimulus complexity.
In this paper, the author refutes the argument made by Dairoku (1995), by showing that his criticism of the author's theory was misguided and based on his misconception of the theory regarding the role of syllabic analysis in the acquisition of reading and writing skills for Kana syllabic letters by children. The author's definition of the acquisition is presented again, and the two models (Amano, 1988, 1989) were recited, one in writing words on code transformation processes of phonological into graphic, and another in reading words by the beginner, on code transformation processes of graphic into phonological. Based on the latter, an analysis-synthesis model on reading and understanding words in which phonological synthesis plays an important role, Dairoku's own model for reading was criticized on two points. First, it lacks a component of phonological synthesis. Second, it falls into an incoherent dualism, in which learning of Kana syllabic letters is through sound-letter associations, and that of word understanding through phonological awareness.
In this rejoinder, the author replies to the arguments made in Amano's comment (1999) on the author's paper (Dairoku, 1995). (1) Dairoku (1995) named, perhaps incorrectly, the Amano theory as “kana naming” hypothesis, because the theory was not accompanied by data showing that reading a kana syllabic letter involved more processes than kana script naming. (2) Amano divides reading a single kana letter into primitive and proper types, but it would be a tautology if the difference between them is defined in terms of syllabic analysis. (3) When Dairoku (1995) used the term “awareness of morae, ” his intention was that it included a synthetic process. (4) The study (Dairoku, 1995) investigated reading comprehension without referring to a synthetic process, because its purpose was to find out a requisite for understanding written words, and not to study the process underlying it. (5) The same study (Dairoku, 1995) included an associative learning task between sounds and letters. The task might seem incoherent from the phonological viewpoint concerning reading instructions, but could be useful in showing that instructions without regard to phonological skills indeed had their limits. Finally, the author offers an apology and correction for some misleading characterization of Amano's theory in his previous paper (Dairoku, 1995).
Models of spoken word recognition, such as logogen model, cohort model, new cohort model, trace model, shortlist model, and neighborhood activation model, were examined from the viewpoints of lexical processing. Although none of these models were good enough to explain all the results of studies on activation, competition, and composition of word candidates, the trace and shortlist models were less problematic than the other models.