To realize a society in which older adults can live independently in their homes and familiar environments for as long as possible, their lives can be supported by providing appropriate technology. In this case, a new intervention for older people using socially assistive robots (SARs) is proposed; however, previous research has demonstrated that individual differences exist in the use and response to SAR interventions, and it has also been reported that SARs are not used by users in some cases. Therefore, in this study, we developed a self-disclosure function to promote continuous interaction with robots, using a Japanese corpus and self-disclosure items. In this study, we defined the specific requirements and functions of self-disclosure in SARs and developed ten non-arbitrary speech scripts from the field of social psychology using a Japanese corpus and self-disclosure items. To evaluate the effect of self-disclosure in SARs, an SAR was introduced to each household for 20 days, with the consent of seven community-dwelling older adults. Based on the recorded voice interaction data, we analyzed how the number, total time, and quality of verbal interactions changed with the SAR’s self-disclosure. Furthermore, we conducted group interviews with the participants and received positive comments regarding the robot’s self-disclosure. Some participants considered the specific personality of the SAR by accumulating its behavioral characteristics. As a consequence, these results indicate that the robot’s self-disclosure feature is effective in significantly increasing the quantity and quality of verbal interactions with older adults.

In recent years, an increasing number of studies have begun to use conversational data in spontaneous speech to estimate cognitive function in older people. The targets of spontaneous speech with older people used to be physicians and licensed psychologists, but it is now possible to have conversations with fully automatic AI agents. However, it has not yet been clarified what difference there is in conversational communication with older people when the examiner is a human or an AI agent. This study explored the psychological burden experienced by elderly participants during cognitive function assessments, comparing interactions with human and AI conversational partners. Thirty-four participants, averaging 78.71 years of age, were evaluated using the Mini-Mental State Examination (MMSE), the Visual Analogue Scale (VAS), and the State-Trait Anxiety Inventory (STAI). The objective was to assess the psychological impact of different conversational formats on the participants. The results indicated that the mental strain, as measured by VAS and STAI scores, was significantly higher during the MMSE sessions compared to other conversational interactions (p < 0.01). Notably, there was no significant difference in the mental burden between conversations with humans and AI agents, suggesting that AI-based systems could be as effective as human interaction in cognitive assessments.

Current medical science has not yet found a cure for dementia. The most important measures to combat dementia are to detect the tendency toward cognitive decline as early as possible and to intervene at an early stage. For this reason, screening for dementia based on language ability has attracted much attention in recent years. However, in most of the previous studies, the cohort of people with dementia has been smaller than the control cohort. In this paper, we use a pre-trained Japanese language model for text analysis and evaluate the effectiveness of text augmentation on a dataset consisting of Japanese-speaking healthy older adults and those with mild cognitive impairment (MCI). We also examined what tasks contributed to the results. This experimental setting can also be used to detect other diseases that may affect the language areas of the brain outside of the hospital.

Every research participant has their own personality characteristics. For example, older adults assisted by socially assistive robots (SAR) may have their own unique characteristics and may not be representative of the general population of older adults. In this research, we compared the average personality characteristics of participants in a workshop on robotics recruited directly through posting with those of older Japanese adults to examine participant selection bias and group representativeness for future study of SARs. After a one-week recruitment period, the workshop was attended by 20 older participants (nine males and 11 females) aged between 62 and 86 years. Extroversion among workshop participants was 4.38, 0.40 higher than the average for older adults in Japan. The workshop participants’ openness was 4.55, 1.09 higher than the average for the Japanese elderly. Thus, the results indicate a slight selection bias in the personal characteristics of the participants depending on the recruitment method when compared to the Japan national average for older adults. In addition, only one of 20 participants was below the cutoff on the LSNS-6 score and considered to have a tendency toward social isolation. The development and introduction of socially assistive robots is often being considered to support people in social isolation in their daily lives; however, the results of this study showed that it is difficult to recruit people who tend to be socially isolated when gathering research participants by methods such as posting. Therefore, the effectiveness of the method of recruiting participants should be carefully verified in research regarding socially assistive robots.

In this article, we developed an interview framework and natural language processing model for estimating cognitive function, based on an intake interview with psychologists in a hospital setting. The questionnaire consisted of 30 questions in five categories. To evaluate the developed interview items and the accuracy of the natural language processing model, we recruited participants with the approval of the University of Tokyo Hospital and obtained the cooperation of 29 participants (7 men and 22 women) aged 72–91 years. Based on the MMSE results, a multilevel classification model was created to classify the three groups, and a binary classification model to sort the two groups. For each of these models, we tested whether the accuracy would improve when text augmentation was performed. The accuracy in the multi-level classification results for the test data was 0.405 without augmentation and 0.991 with augmentation. The accuracy of the test data in the results of the binary classification without augmentation was 0.488 for the moderate dementia and mild dementia groups, 0.767 for the moderate dementia and MCI groups, and 0.700 for the mild dementia and MCI groups. In contrast, the accuracy of the test data in the augmented binary classification results was 0.972 for moderate dementia and mild dementia groups, 0.996 for moderate dementia and MCI groups, and 0.985 for mild dementia and MCI groups.

In recent years, an increasing number of studies have begun to use conversational data in spontaneous speech to estimate cognitive function in older people. The providers of spontaneous speech with older people used to be physicians and licensed psychologists, but it is now possible to have conversations with fully automatic AI agents. However, it has not yet been clarified what differences exist in conversational communication with older people when the examiner is either a human or an AI agent. In this study, elderly people living in the community and attending a silver human resource center and a day service center were the subjects. Dialogues were conducted using generic interview items for estimating cognitive function through daily conversation, which were developed through research on estimation methods for cognitive function. From the data obtained from the dialogues, we compared the effects of human–AI interaction on the number of utterances, speaking time, and silence time. This study was conducted at a facility in Japan and included 32 subjects (12 males and 20 females). The results showed significant differences between human and AI dialogue in the number of utterances and silent time. This study suggests the effectiveness of AI in communication with older people and explores the possibility of using AI in social welfare.