Participants were 20 elderly people who attended a community college for the elderly.

When we hypothesized that the effect size was 0.63, power analysis was 0.8, alpha was 0.05, and a paired t-test, the sample size was 23. However, the period of recruitment of the community college was limited, and a total of 20 elderly people participated. The number of males was 8 and the number of females was 12. They lived in Kyushu in Western Japan. The mean age was 77 years old. The education level was varied, and they were all totally healthy.

We made the Nostalgia Reminiscence application (app) in consultation with an information technology company. Nostalgia was induced by videos of nostalgic photos or illustrations with music.

Firstly, a participant answered the questionnaires using Google Forms. The app consists of the following ( Table 1 ). The app asked each question, and the participant answered it orally. The voice was translated into words.

We used the following questionnaires.

1) Nostalgia States Scale ( Sedikides et al., 2015 ): It consisted of 3 questions measured from 1 point (not at all) to 6 (very true).

2) Nostalgia Function Scale ( Kusumi , 2021 ): It is based on Cheung et al. (2013) or Sedikides et al. (2016) . It consists of 4 factors: Social Connection, temporal Self-Continuity (Self-Continuity), Meaning of Life, and Self-Clarification. There are 16 question items measured from 1 point (not at all) to 5 (well true).

3) Japanese version of the trait Gratitude scale ( Shiraki & Igarashi, 2014 ). It is translated from the original by McCullough, Emmons, & Tsang (2002) . There are 5 question items measured from 1 (not at all) to 7 (all).

4) Subjective Happiness Scale ( Matsubayashi et al., 1992 ). We created a visual analog scale measured from 0 to 10 for elders to answer easily.

We recruited participants through a community college for elders, which was set beside NishiKyusyu University in Western Japan. Participants received documents explaining this research. The researcher explained this research and obtained informed consent from those who were interested in the research. The reminiscence therapy consisted of a session. The duration of the session is about 60 minutes. A participant met the researcher in a private room. The researcher explained how to use the app, and the participants began the Nostalgia Reminiscence. The researcher attended the intervention and helped when the participant had trouble.

We conducted 1) basic statistics and t-test between pre and post, and 2) correlation analysis among each questionnaire. To conduct t-test, we tested normality and we confirmed that t-test was useful.

This study was approved by the ethical board of Nisikyusyu University.

The significant increase in the Nostalgia State scale after intervention shows that this method could introduce nostalgic feelings in elders and suggests that the Nostalgia Reminiscence Therapy was effective in introducing these feelings. Many people use smartphones, and this method will be available to many people in the future. Using ICT for reminiscence therapy was suggested by Haron et al. (2017) .

Moreover, the completion rate was 100%, and all participants completed. The reason for this rate is that a participant found it easy to participate in this therapy because there was only one session, and they could narrate their lives easily and freely to a smartphone. We will investigate the effects of session times, including the completion rate.

Regarding the feeling of nostalgia, since the Nostalgia State Scale score significantly increased, Nostalgia Reminiscence Therapy via smartphone improved nostalgia. We will investigate the levels of nostalgia induced by various video content.

Regarding the subscales of Nostalgia Function, Social-Connection, Self-Continuity, Meaning of Life, and Self-Clarification, all significantly increased after the intervention. When a person remembers significant events’ memories, they also remember significant people related to the events. Then they recognize social connection. This supports a previous study that Nostalgia can facilitate social connection ( Polletta & Callahan, 2017 ).

Moreover, as participants review their lives from the past to the present time, they may feel self-continuity. It supports Sedikides et al. (2008) , in which nostalgia is associated with self-continuity.

About gratitude, nostalgia improved gratitude. In the session, many participants expressed their gratitude or thanks to their families or people around them. They seemed to recognize that a person could not live alone, and they were helped by others. Thus, nostalgia improved gratitude. When individuals recall past events, they probably think of the assistance and support received from significant others, which leads them to a sense of gratitude for what they have ( Puente-Diaz & Cavanzos-Arroyo, 2021 ).

We referred to characteristic points. Nostalgia correlated with Gratitude and Nostalgia Function. Particularly, the correlation score between Nostalgia and Meaning of Life (r = 0.68) is high in this study, and it suggests that Nostalgia contributes to making meaning of their lives. This result supports Routledge et al. (2012) , in which nostalgia was associated with existential meaning, or Sedikides & Wildschut (2018) , in which they showed that nostalgia increased meaning of life.

Significant relation between Nostalgia and Self-Continuity (r = .64) in the present study supports Sedikides et al. (2008) . Also, there were significant correlations between Gratitude and Subjective Happiness (r = .53), and Self-Continuity and Subjective Happiness (r = .39). From these results, we suggest that Nostalgia affects Gratitude, Meaning of Life, and Self-Continuity. And Gratitude and Self-Continuity affect Subjective Happiness. Thus, we can present a model like Figure 1 . This idea can add the finding ( Li et al., 2023 ) such that gratitude affects relationships between nostalgia and subjective well-being. The reason for this is that elderly people feel not only gratitude to others but also meaning of life or self-continuity and feel subjective happiness. We need to analyze this model statistically in the future.

About Subjective Happiness, there was no significant correlation directly between nostalgia and subjective happiness. The reason for no relation is as follows:

Diener et al. (1999) showed that subjective well-being includes two components: one is emotional experience and life satisfaction, which is the judgment of one’s personal life overall ( Li et al., 2016 ). Since the subjective happiness as well-being in this study is to measure emotional experience and does not measure life satisfaction, there was not a significant relation between nostalgia and well-being. If we include life satisfaction as a measure of well-being, there may be significant relations.