Social interactions are a basic human need, serving a multitude of purposes by fulfilling the need for social integration and by providing feelings of closeness, relatedness, social support, and belonging with others [1-7]. The quantity and quality of social interactions have an impact on subjective well-being and quality of life [8,9], and are associated with better health [10-13].

The digital revolution has offered more and more possibilities for individuals to connect with others and share experiences. Younger people use the internet almost universally, while older adults increasingly use digital technology to communicate, albeit at lower proportions than younger adults [14]. Although the use of digital technology comes naturally to those who have been familiar with it their entire lives, or have used the technology during their professional lives, older generations might find it more difficult to become accustomed to using these new technologies. However, these technologies could be more and more useful for older people today. Growing distances between family members as well as declining marriage and birth rates [15,16] may lead to concerns that older adults are at increasing risk of loneliness. This stands in contrast to findings that show lower levels of loneliness in older adults currently compared to previous generations [17,18]. Therefore, it is vital to investigate possible mechanisms by which older adults might compensate for weaker family ties.

One compensatory mechanism is the move from mainly family-oriented to more friend-focused social circles [19,20]. In addition, a greater variety of living situations have become more accepted in recent years; for example, being divorced does not affect social loneliness to the same extent as in previous generations [21]. Finally, the digital revolution offers older adults the opportunity to stay in close contact with family and friends, regardless of geographical distance, as well as the opportunity to connect with new social contacts based on shared interests [22].

Older adults use the internet at increasing rates [14,23]. In Switzerland, 32% of older adults report having a smartphone and 26% own a tablet computer [24]. The internet is used most commonly to communicate with others [25], and the same is true for older adults’ internet use: in Switzerland, 80% of older internet users report using the internet for social interactions [23]. In addition, older internet users agree that the internet has made it easier for them to reach people [26]. Despite the increase in older adults’ internet use, some older people still do not have access to digital technologies. Sociodemographic characteristics play an important role [27], such as education [28] and age [14,23]. Furthermore, cognitive ability is a stronger predictor of older adults’ internet use than age alone [29]. Need for cognition, a personality trait that reflects a preference for cognitively effortful activities, is also positively associated with the frequency of internet use in older adults [30].

Previous research has primarily focused on factors that predict internet use in older adults, as well as the activities older adults perform online. However, relatively little is known about how older adults integrate digital technologies into their day-to-day lives [31]. To begin to understand the role of digital technologies for social interaction and for reducing loneliness in old age, it is vital to investigate the (digital) social lives of older adults.

Information about social interactions is often gathered through retrospective self-reports; however, these responses may be biased by more salient recent events [32]. In contrast, in an event-contingent, microlongitudinal approach, participants answer a set of questions each time a prespecified event, in this case a social interaction, occurs [33]. This enables participants to report their experiences and reactions to this event immediately, or soon after it occurs, and minimizes retrospective bias. Several studies have investigated social interactions using such microlongitudinal designs [34,35]. One study reported an average of 7 interactions that lasted 5 minutes or longer [36,37], highlighting the large amount of data that can be collected using such a design. Thus, detailed information about the social lives of older adults will facilitate the investigation of a variety of research questions related to older adults’ use of digital technology.

To study the day-to-day social experiences of older internet users, this study is based on an event-contingent microlongitudinal design. Specifically, participants were asked to complete a brief protocol on every social interaction lasting 5 minutes or longer, and on every text-based interaction (eg, letter, text message, email) over a period of 3 weeks. The brief protocol assessed information about the duration, conversation partner(s), and purpose, as well as the perceived quality of the conversation.

In addition to the interaction reports, participants were asked to fill in questionnaires every evening on their daily mood, health, and feelings of loneliness. At baseline, data on global measures of technology use, cognition, personality, loneliness, and quality of life were collected. Finally, participants were asked to provide 6 saliva samples per day for 3 days, which will be used to examine daily trajectories of diurnal cortisol. Salivary cortisol is a biomarker that is commonly used to assess stress and resilience [38] throughout the lifespan and is also related to loneliness [39]. Salivary cortisol has a detectable daily rhythm, showing high levels in the morning, around 30 minutes after awakening, and decreasing throughout the day [40,41]. This daily rhythm of cortisol secretion can reliably be detected in older adults, despite increased intraindividual variability [42].

The goal of this study is to understand daily social experiences and their associations with well-being in older internet users. Specifically, the analyses will focus on individual patterns of social communication and their relationship with facets of daily and global well-being, taking into consideration the modalities through which participants communicate.

A total of 120 community-dwelling participants were recruited for this project. Inclusion criteria were being aged 65 years and above, using digital devices to communicate, sufficient vision and hearing, and being fluent in German. Inclusion criteria were aged 65 years and above, using digital devices to communicate, sufficient vision and hearing, and being fluent in German.

Participants were recruited via advertisements in local and national newspapers, and through a database of participants hosted at the University of Zurich. Initial contact took place via telephone or email, as preferred by the participant. Eligible participants were invited to a baseline session at the University of Zurich, where they were given detailed information on the study, had the opportunity to ask questions before giving informed consent, received detailed instruction on the study protocol, and received study materials to complete the protocol, including the study smartphones, and Salivettes (Sarstedt AG & Co, Nürnbrecht, Germany) to prepare saliva samples. All information was also provided in written format for participants to take away with them. In addition, they received a phone number to call in case of any questions or problems. The study took place over 21 days starting the day after this initial session. After the first day of the study, participants received a phone call from a research assistant to clarify potential questions. During the study period, a research assistant was on call to answer questions and provide technical support. After 21 days, participants visited the lab to return study materials, give feedback and share their experiences, and to complete a battery of cognitive tests. Participants received CHF 150 (~US $164) to thank them for their participation.

This study protocol was reviewed by the Ethics Committee of the Faculty of Arts and Social Sciences at the University of Zurich (Nr. 19.2.17).

This study is planned according to a microlongitudinal design with data collection occurring at different time intervals: event-based, time-based, daily, and single assessments at baseline, as summarized in Textbox 1. Participants were required to complete a brief protocol on every spoken social interaction lasting 5 minutes or longer and every text-based social interaction for a period of 21 days. The brief protocol assessed information about the duration, conversation partner(s), purpose, as well as the perceived quality of the conversation. In addition, data on participants’ mood, loneliness, health, and technological issues were collected daily. For the first 3 days of the study period, participants were asked to prepare 6 saliva samples per day. Finally, questionnaires assessing sociodemographic information, health, quality of life, personality, and technology use and attitudes, as well as cognitive ability tests were completed at baseline.

Data collection started in April 2019 and was completed in November 2019. Participants were on average 73 years old (SD 5.08, range 66-95 years). Most of the participants were highly educated: 26.7% (32/120) completed their school education with the Swiss Matura, a secondary school degree qualifying for university education. The vast majority (98.3%, 118/120) completed further training after school with 22.5% (27/120) completing university, including 4.2% (5/120) participants with a PhD.

In general, participants were very technologically versed. On the computer proficiency questionnaire [47], they scored an average of 132 out of 160 possible points; 96.7% (116/120) of participants owned a computer, 86.7% (104/120) owned a smartphone, and 65.8% (79/120) owned a tablet. Generally, 40.0% (48/120) of participants reported that using the internet to communicate had made them feel closer to other people, 50.0% (60/120) reported that nothing had changed, and 10.0% (12/120) reported that they felt less close to other people.

Participants reported an average of 96.35 interactions across the 21 days (SD 67.57, range 9-517). These data are summarized Table 1. Women reported a higher average of interactions than men. The majority of interactions were face-to-face, followed by interactions by phone, email, and text message with relatively equal frequencies. Social media, letters, and videochat were the least frequent types of interactions reported.

Most conversations took place with people the participants knew, with less than 10% of the conversations taking place with people they did not know and service providers, respectively.

These data will enable more detailed investigation of how daily social interactions can shape older adults’ daily lives, as well as how intraindividual interaction patterns are related to individual differences such as personality or cognition. Specifically, we will also be able to further examine the role of different patterns of communication through digital modalities in older adults’ daily lives and well-being. To analyze the interplay between (digital) social communication, well-being, and individual differences, we will conduct multilevel analyses to capture the individual variety in communication patterns. This includes investigating the relationship between daily communication patterns and mood in relation to personality and closeness with interaction partners, the relationship between personality (especially openness) and use of digital technologies, and the relationships among daily communication patterns, loneliness, and diurnal cortisol.