AI Chatbot: Against Loneliness and Grey Digital Divide in Elderly

As human beings, the elderly require social interaction just like anyone else. Studies conducted by Sanjaya and Rusdi (2012), Amalia (2013), and Nuraini et al. (2018) have shown that social interaction can reduce loneliness in older adults. Unfortunately, many older adults experience social exclusion, leading to a lack of social participation and isolation, as Balard et al. (2019) noted. This social exclusion directly impacts senior’s mental health, leading to loneliness, a global problem that can result in depression and suicide. According to the Ministry of Health (2013), only 16% of older adults in Indonesia do not experience loneliness. The remaining 69% are mildly lonely, 11% are moderately lonely, and 2% are severely lonely. It is concerning that in 2045, Indonesia is predicted to experience a surge in the number of older adults, with one in five people over 60 years old (Harsono, 2022). Furthermore, access to mental health services is costly and focuses primarily on physical needs (Ryu et al., 2020).

Given these challenges, chatbots can be a valuable innovation that facilitates social interaction for the elderly. A chatbot, chatterbot, or chat robot is a computer program that allows human interaction with robots via text or natural language, similar to commercial conversational agents such as Siri, Google Assistant, and Cortana (Mahapatra et al., 2012). While older adults have traditionally relied on broadcast media such as television and radio, which have been shown to create a sense of companionship (Garcia-Mendez et al., 2021), chatbots offer more than that with their intelligent ability to respond like humans. Chatbot’s conversational nature can reduce loneliness among the elderly by providing social interaction.

In addition to reducing loneliness, chatbots can help overcome the grey digital divide faced by the elderly when using internet technology. The grey digital divide refers to the marginalization of the elderly in the first level of the digital divide (Friemel, 2016; Huxhold et al., 2020). This digital divide is associated with the elderly’s use of internet technology, which can be doubly dangerous as it is not only a consequence of age but also other socio-demographic factors such as education, income, and gender (Huxhold et al., 2020). This digital divide can also contribute to the digital exclusion of the elderly, particularly in developing countries where they may lack the necessary skills to connect and perform basic tasks online, such as claiming old-age benefits or ordering tickets (Mubarak & Suomi, 2022).

A Combination of Health and Entertainment

There are various types of chatbots, but there are three main classifications based on their purpose: informative chatbots to obtain specific information, natural chat/conversation based like humans and task-based such as booking tickets (Adamopoulou & Moussiades, 2020). Several categories of chatbot applications are widely used today ranging from ‘education, customer service, health, and robotics’ to ‘industrial use cases’ (Adamopoulou & Moussiades, 2020). Concerning seniors, the chatbot is designed primarily to relate to the health and well-being of those who do not have a social support system for assistance. So, chatbots with 24-hour service can be helpful for older adults to carry out their daily lives, from engaging in meaningful conversations, sharing stories, and providing reminders about treatment to providing information with helpful advice on various topics.

Historically, chatbots have been strongly connected with health, especially mental health. Chatbots are considered to be able to provide companionship and emotional support to the elderly. The first chatbot appeared in 1966, ELIZA, to imitate psychologists in clinical therapy; then, in 1972, PARRY appeared, which played the role of a schizophrenia patient. Only in 1995 Dr. Richard S. Wallace created the Artificial Linguistic Internet Computer Entity (ALICE), which ELIZA inspired. ALICE is the first online chatbot that has become open source and produces Artificial Intelligence Markup Language (AIML). (Adamopoulou & Moussiades, 2020; Han, 2023)

Several developers have also created chatbot applications specifically for the elderly. One example is EBER, designated as a ‘smart radio,’ reading the news and responding empathetically. EBER, developed by Garcia-Mendez et al. (2021), combines healthcare and entertainment chatbots to reduce the digital gap for the elderly. They combine the AIML language, automatic Natural Language Generation, and Sentiment Analysis. They also tested it on the elderly and found that 80% of users gave a satisfaction score of 4 out of 5, which was proven to improve the information search abilities of the elderly.

Next is the chatbot Charlie, a virtual assistant developed by Valtolina & Marchionna to become a caregiver and virtual friend for those aged 60 years and over. Like EBER, Valtolina and Marchionna (2021) also prioritize their chatbot as an empathetic, sensitive, friendly, and sociable robot that combines healthcare and entertainment. Using DialogFlow, an NLP language from Google Corporation, they also considered Charlie’s personality by giving him the characteristics of a child. The similarities between seniors and children create a friendly conversation with Charlie. Through its features, Charlie seeks to provide more preventive mental health care with a gamification approach that also helps older adults who experience memory loss.

Apart from EBER and Charlie, there are many other chatbot applications aimed at the mental health of the elderly. Chou et al. (2023) summarize several medical chatbot applications that are available on Android or iOS with participants aged 65 years and over, such as Oiva to treat depression symptoms and improve mindfulness skills; MIND MORE for insomnia problems; Interactors for homecare and well-being; and MIT App Inventor to improve mindfulness and well-being. Based on the results of a systematic review from Chou et al. (2023), there was good acceptance of using chatbots for mental health. Seniors benefited from psychological variables: well-being, stress, and depression.

Current Challenges and Problems

Senior’s technology skills are often an issue in accessing commercial chat systems. Therefore, the developers created a virtual assistant by adjusting its components. The main components remain the same: user interface, normalizer, and knowledge base (Garcia-Mendez et al., 2021). Regarding the user interface, chatbots need to develop an interface that remains simple, language that is easy to understand, and user-friendly (Chou et al., 2023; Ryu et al., 2020). Meanwhile, the normalizer occurs after inputting the command, and the data will be normalized, namely confirming the spelling of each word, calculating the similarity with existing data (sentence similarity), and then matching (Rizaldhi et al., 2020). This second part requires clear commands from the elderly, who input data related to the next component, namely the knowledge base.

The knowledge base regarding the elderly is essential to produce more personalized services. Apart from understanding the language used, it is necessary to collect as much information as possible about their backgrounds (Valtolina & Marchionna, 2021). Valtolina and Marchionna (2021) use a machine-learning techniques model to predict and calculate what advice should be given to the elderly, then extract the knowledge collected and describe the behaviour of the elderly as users. Specifically regarding loneliness, Garcia-Mendez et al. (2021) consider the assessment of loneliness and the development of the quality of user responses with attention to age-related language differences is also influential.

Apart from chatbots being able to build companionship with the elderly, chatbots should also be able to connect the elderly with friends or family. Indeed, digital skills mean that seniors cannot connect with their peers via digital communication networks. This connection is related to task-based chatbot functions such as ElliQ, voice-enabled conversational agents that can make video calls and messages to friends and family (Ryu et al., 2020). Regarding the complexity of this problem, Ryu et al. (2020) stated the importance of involving seniors in the design and development process to meet their needs.

As we have seen, social interaction is crucial for the mental well-being of senior citizens, and chatbot technology can be a valuable innovation that facilitates social interaction for the elderly. The use of chatbots has the potential to reduce loneliness and overcome the digital divide faced by older adults. Chatbots can provide companionship, emotional support, and helpful advice to seniors, especially those who do not have a social support system. However, paying attention to data privacy and security issues while developing chatbots is essential. Developers must protect user data and ensure seniors control their information. By doing so, we can ensure that chatbots provide a safe and beneficial experience for the elderly. (Chou et al., 2023)

References

Adamopoulou, E., & Moussiades, L. (2020). Chatbots: History, technology, and applications. Machine Learning with Applications, 2(November), 100006. https://doi.org/10.1016/j.mlwa.2020.100006

Amalia, A. D. (2013). Kesepian Dan Isolasi Sosial Yang Dialami Lanjut Usia: Tinjauan Dari Perspektif Sosiologis. Sosio Informa, 18(3), 203–210. https://doi.org/10.33007/inf.v18i3.56

Balard, F. (2019). The Social Inclusion Of Older People In France: Social Participation, Loneliness And Giving. Gerontology & Geriatric Medicine, 5(4), 1–7. https://doi.org/10.24966/ggm-8662/100040

Chou, Y.-H., Lin, C., Lee, S.-H., Chang Chien, Y.-W., & Cheng, L.-C. (2023). Potential Mobile Health Applications for Improving the Mental Health of the Elderly: A Systematic Review. Clinical Interventions in Aging, Volume 18(September), 1523–1534. https://doi.org/10.2147/cia.s410396

Friemel, T. N. (2016). The digital divide has grown old: Determinants of a digital divide among seniors. New Media and Society, 18(2), 313–331. https://doi.org/10.1177/1461444814538648

Garcia-Mendez, S., De Arriba-Perez, F., Gonzalez-Castano, F. J., Regueiro-Janeiro, J. A., & Gil-Castineira, F. (2021). Entertainment Chatbot for the Digital Inclusion of Elderly People without Abstraction Capabilities. IEEE Access, 9, 75878–75891. https://doi.org/10.1109/ACCESS.2021.3080837

Han, Z. (2023). The applications of chatbot. Highlights in Science, Engineering and Technology, 57, 258–266. https://doi.org/10.54097/hset.v57i.10011

Harsono, F. H. (2022). Seperlima Penduduk RI pada 2045 adalah Lansia, Tertinggi Yogyakarta. Liputan6. https://www.liputan6.com/health/read/4973792/seperlima-penduduk-ri-pada-2045-adalah-lansia-tertinggi-yogyakarta?page=2

Huxhold, O., Hees, E., & Webster, N. J. (2020). Towards bridging the grey digital divide: changes in internet access and its predictors from 2002 to 2014 in Germany. European Journal of Ageing, 17(3), 271–280. https://doi.org/10.1007/s10433-020-00552-z

Mahapatra, R. P., Sharma, N., Trivedi, A., & Aman, C. (2012). Adding interactive interface to E-Government systems using AIML based chatterbots. 2012 CSI 6th International Conference on Software Engineering, CONSEG 2012, 1–6. https://doi.org/10.1109/CONSEG.2012.6349510

Ministry of Health. (2013). Buletin Jendela Data dan Informasi Kesehatan: Gambaran Kesehatan Lanjut Usia di Indonesia.

Mubarak, F., & Suomi, R. (2022). Elderly Forgotten? Digital Exclusion in the Information Age and the Rising Grey Digital Divide. Inquiry (United States), 59, 1–7. https://doi.org/10.1177/00469580221096272

Nuraini, Kusuma, F. H. D., & H., W. R. (2018). Hubungan interaksi sosial dengan kesepian pada lansia di kelurahan Tlogomas Kota Malang. Nursing News: Jurnal Ilmiah Keperawatan, 3(1), 603–611. https://salmandj.uswr.ac.ir/browse.php?a_id=1453&sid=1&slc_lang=en&html=1

Rizaldhi, D. A., Rosyad, G. A. K., & Hartanto, A. D. (2020). Implementasi Algoritma Sentence Similarity Terhadap Chatbot Seputar Amikom. METHOMIKA Jurnal Manajemen Informatika Dan Komputerisasi Akuntansi, 4(1), 10–14. https://doi.org/10.46880/jmika.vol4no1.pp10-14

Ryu, H., Kim, S., Kim, D., Han, S., Lee, K., & Kang, Y. (2020). Simple and Steady Interactions Win the Healthy Mentality. Proceedings of the ACM on Human-Computer Interaction, 4(CSCW2), 1–25. https://doi.org/10.1145/3415223

Sanjaya, A., & Rusdi, I. (2012). Hubungan Interaksi Sosial Dengan Kesepian Pada Lanjut Usia. Jurnal Keperawatan Bina Sehat, 14(2), 26–31.

Valtolina, S., & Marchionna, M. (2021). Design of a Chatbot to Assist the Elderly. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 12724 LNCS(March 2020), 153–168. https://doi.org/10.1007/978-3-030-79840-6_10