In epistemological terms, most of our cognition is shaped in a mediated way rather than by personal, sensory experience. Scientific technology represents one of the most significant modern media to arbitrate human cognition, as may be exemplified by Galileo's use of the telescope to expand the knowledge of astronomy. Contemporary digital devices can be positioned in the genealogy of this optical instrument.

This paper, first, analyzes human cognition and behavior through the concept of media, and positions the digital revolution in an epistemological context. It then explores the implications of the digital revolution for the human apprehension of the morality that drives the actions of our vulnerable species. Finally, I suggest a desirable prospect for digital media.

With advances in digitalization, it has become possible to obtain information from digital devices without actually experiencing them and use that knowledge in actual experiences. In this study, we examined whether preschoolers can learn biological concepts by interacting with pet robots. In Study 1, for two conditions—a static robot and a moving robot—it was observed that preschoolers considered the robots as inanimate when these were stationary and as animate when the robots moved. In Study 2, when preschoolers reared a robot instead of a living creature for one month, many utterances and instances of interaction similar to as with an organism were observed during the early days. However, after two weeks, the robots were perceived as boring and irrelevant. Thus, the effect of the educational instruction was ineffective after two weeks. From the two studies, it was observed that when the contact time between the robot and the child was short and the presence/absence of “movement” was switched in an instant, the preschoolers perceived the presence of movement in the robot as the movements of a living creature. However, the preschoolers did not perceive monotonous movement of the robot similar to a living creature when the robot took a long time to move. Thus, it can be concluded that “movement” does not always contribute to the acquisition of biological concepts and it is important to study the quality of robot movement.

We examined how new computer technologies influence the methodologies of developmental psychology. Focusing on sensor-based activity-recognition technology, we divided previous studies into two categories: those that focused on “places” where sensors were placed (on the human body or in the environment) and those that focused on a “research goal.” The research goal category was further divided into two subcategories: (i) constructing a developmental theory or model for explaining the developmental process and (ii) supporting child rearing and caring. Using this categorization, we reviewed previous studies that focused on children's behavioral development processes using sensors. The problems of adopting sensing technologies included the following: (i) recognition inaccuracies, (ii) temptation to use all available sensors without deep consideration, (iii) lack of participants'privacy, and (iv) risk of sensor data affecting the direction of educational practice. The advantages of using these technologies include the following: (i) enhanced understanding of the impact of children's everyday experiences on their developmental process, (ii) new possibilities for research on children's activities that are difficult to observe, and (iii) prompt feedback to the participants.

Today's rapidly changing society places higher significance on collaborative learning, which integrates one's life experiences through interaction with others and fosters the meaning of life in interpersonal relationships. This research develops a collaborative system involving community senior citizens and college students who participated in programming education for elementary school students (compulsory since 2020) to explore the core concept of a community practice connecting multiple generations through learning. The “DANCE Method” programming workshop was developed using a robot dance programming system and designed based on three components: accepting diversity, utilizing technology, and creating an environment that engages many generations. Focusing on changes in multigenerational relationships through activities, we validated the community formation process and found three factors that support a multigenerational community: (1) a shared altruistic purpose that lets everyone play the role of a teacher, (2) a multilayer project structure that promotes diverse roles, and (3) equal opportunities to use technology.

This study examined the two most potentially important factors involved in the control of facial expressions: emotional understanding and facial expression-making skills. 101 child participants aged 4–6 years old were given three tasks. First, two children played a card game, and the change in their facial expressions was measured. Second, the children listened to two stories and were asked about “the positive/negative effects of negative/positive expressions.” Third, the children were asked to make eight types of faces expressing four emotions: happy, sad, angry, and surprised, with two intensities: very and little. The main results were as follows: (1) The first facial expression control task score correlated with age and (2) the scores for the second emotional understanding task and the third facial expression-making task correlated with the facial expression control task scores. These results suggest that the development of facial expression control in preschool and elementary school children progresses with the development of emotional understanding and facial expression-making skills.

We examined factors that influence error patterns in kana letter naming. In study 1, using data on young children's naming errors (NINJAL, 1972), we investigated (1) effects of phonological and formal similarity on frequency of pairs of the target letter and the response (hereafter right/wrong pairs) and (2) causes of the phenomenon in which the frequency of right/wrong pairs is asymmetric (e.g., 240 for ぬ–ね，30 for ね–ぬ). For (1), ordinal logistic regression analysis revealed that factors of consonant, vowel, and formal similarity contributed independently to the frequency level of right/wrong pairs, but the extent of contribution of formal similarity was much greater than that of the other factors. As for (2), we found that in highly frequent pairs in asymmetric cases, the appearance frequency of wrong letters was significantly higher than that of right letters. Using recent data, Study 2 examined naming errors' developmental change. Results showed that beginners made more formal errors than advanced learners, whereas the reverse was true for formal-phonological errors.