The prediction of late-life loneliness is a highly important research endeavor, since loneliness has detrimental effects on older adults’ health and mental health (e.g., Blazer, Reference Blazer2020). While objective factors, such as social isolation constrictions, may increase late-life loneliness (for the effect that COVID-19 lockdown policies had on late-life loneliness, see Hoffart et al., Reference Hoffart, Johnson and Ebrahimi2020), loneliness is basically a subjective phenomenon, defined as the gap between the individual’s desired and actual levels of social relations (de Jong Gierveld et al., Reference de Jong Gierveld, van Tilburg and Dykstra2018). Subjective age and self-perceptions of aging are two forms of views on aging (Wurm et al., Reference Wurm, Diehl, Kornadt, Westerhof and Wahl2017), which share a similar characteristic with late-life loneliness, as these three concepts are ingrained in older adults’ experience of old age. In similar to the definition of loneliness, subjective age is measured by the age one feels as related to his/her actual age (Kotter-Gruhn et al., Reference Kotter-Gruhn, Kornadt and Stephan2016), and self-perceptions of aging are conceptualized as the subjective views and expectations that individuals have about their actual aging process (Wurm et al., Reference Wurm, Diehl, Kornadt, Westerhof and Wahl2017). It is therefore important to examine the role of subjective age and self-perceptions of aging in predicting late-life loneliness.
Spitzer et al.’s (Reference Spitzer, Segel-Karpas and Palgi2019) study is focused on this important niche of investigation. Their study contributes an additional empirical evidence to the role of subjective age in predicting late-life loneliness. In their work, the authors investigated if subjective age moderates the association between subjective perceptions of close social relationships and loneliness among people at the second half of life. Spitzer and her colleagues collected data between March and July 2017 using an in-region random digit dialing methodology. Their final sample consisted of 853 Israeli participants who completed most of the survey (were interviewed either in Hebrew or Arabic, according to their mother’s tongue), from which they selected 360 married or cohabiting community-dwelling participants at the age range of 50 to 97 years. Participants’ close social relationships level was measured by Schuster et al.’s (Reference Schuster, Kessler and Aseltine1990) 3-item scale, subjective age was measured by a 4-item adaptation of Kastenbaum et al.’s (Reference Kastenbaum, Derbin, Sabatini and Artt1972) scale, and loneliness was assessed by the 3-items of the Revised UCLA loneliness Scale (R-UCLA) (Hughes et al., Reference Hughes, Waite, Hawkley and Cacioppo2004). Analyzing these data, the authors reconstructed findings from previous studies. In line with the literature (Keyes and Westerhof, Reference Keyes and Westerhof2012), the mean subjective age of their participants as measured on a scale ranging from 1 (“much younger than my age”) to 5 (“much older than my age”) with 3 (“as my age”), was younger than the participants’ actual age (i.e., M = 2.42; SD = 0.66). Moreover, both close social relationships and older subjective age were positively associated with increased loneliness, although these associations were small. But more importantly, they found that participants with an older subjective age did not demonstrate a significant negative association between their level of close social relationships and their loneliness level, whereas participants with a younger subjective age did.
Before discussing these interesting findings, it is important to view the contribution of this study in the context of its limitations. First, their use of a cross-sectional study design is a lacuna which, as the authors themselves admit, prevents any conclusions about causality. Therefore, the explanations suggested by the authors for the moderating effect of subjective age (see below) can only be regarded as directions for future research. Moreover, probably due to space limitations (the article is a brief report), the authors did not provide statistical comparisons between participants who attrited from the survey and participants who completed the survey, and we therefore lack this information about the representativeness of their sample. Additionally, while the rational for selecting participants above the age of 50 years is clear, the rational for choosing married, or cohabiting participants, is missing. Furthermore, their participants ranged between 50 and 97 years. This is a very wide age range for measuring loneliness, because loneliness during the eighth and ninth decades is much stronger than in previous decades (for a review see Donovan and Blazer, Reference Donovan and Blazer2020). Finally, the use of very short unidimensional measures of loneliness and close social relationships did not allow for a more comprehensive investigation of the effect of subjective age on loneliness, as the authors also admit.
Nevertheless, the authors’ discussion on potential mechanisms that may explain the moderation effect of subjective age provides several insights. First, following the natural tendency of people to perceive themselves as younger than their chronological age (Keyes and Westerhof, Reference Keyes and Westerhof2012), the authors suggested that people with an older subjective age may feel alienated from their social peers who perceive themselves as younger. Second, in line with previous findings, showing that subjective age also serves as an indicator for people’s physical and mental health condition (Stephan et al., Reference Stephan, Sutin and Terracciano2018), they suggested that an older subjective age may be associated with worse physical and mental health and that health problems can make it harder for older adults to benefit from close social relations. These reasonable explanations can be researched by using longitudinal studies that will examine the mediating roles of feelings of alienation and physical and mental health in the connection between close social relationship and loneliness among participants with an older and a younger subjective age, using moderated mediation models.
Spitzer et al.’s (Reference Spitzer, Segel-Karpas and Palgi2019) study has several strengths. From the theoretical point of view, this study emphasizes the importance of self-perceptions of aging (in this case of subjective age) in affecting the painful experience of loneliness. In view of the bulk of studies which focus on loneliness as predicted by objective circumstances, such as social isolation (see review by Leigh-Hunt et al., Reference Leigh-Hunt2017), the focus of the authors on the subjective aspects of both loneliness, close social relations, and subjective age is important, as it leaves room for psychological interventions aimed at improving these subjective perspectives. Therefore, their study calls for future studies to construct and examine psychological interventions that will focus on people with an older subjective age and improve their subjective perceptions of aging, so they will be able to enjoy close social relations and decrease their sense of loneliness. Moreover, as the authors argue, their study adds an important tier to the socio-emotional selectivity theory (Carstensen, Reference Carstensen1992), which claims that the ability to focus on meaningful social friendships improves with age. Their findings may imply that this social ability improves among older adults with a younger subjective age, but not among those with an older subjective age.
Before closing this commentary, I would like to elaborate on another directive for future investigations. A review of the literature on self-perceptions of aging and late-life loneliness which corresponds well with their results may offer several insights regarding the mechanism that enables subjective age to moderate the connection between close social relationships and late-life loneliness. This mechanism may operate by activating self-expectations regarding old age. In this regard, Pikhartova et al., (Reference Pikhartova, Bowling and Victor2016) have demonstrated that one of the basic tenets that older adults hold regarding old age is that it is a time of loneliness. Pikhartova and colleagues (2016) examined data from several waves of the English Longitudinal Study of Ageing (ELSA). These data included the relationship between perceived stereotypes about old age and the expectation for late-life loneliness among participants who were 50 years old and over in 2004, and their actual loneliness reported within 8 years of follow-up (throughout the years 2006–2021). They found that 24% of their participants in 2004 (N = 4,465) agreed that old age is a time of loneliness and 33% expected to be lonely in old age. They also found that both stereotypes and expectations significantly predicted future loneliness eight years later. These findings are in line with the stereotype embodiment theory (Levy, Reference Levy2009), which argues that stereotypes about aging are internalized throughout the life cycle and become self-expectations. In other words, it is possible that people who feel older than their actual age also tend to believe that old age is a time of loneliness and are therefore inclined to report feeling lonelier. Moreover, their internalized negative expectation about loneliness being an integral part of their older subjective age may hamper their ability to utilize close relationships as a way to decrease their loneliness. Future studies may examine the association between this negative expectation and an older subjective age.
There are also additional studies that did not consider the specific expectation regarding the existence of loneliness in old age, but focused on the association between other self-perceptions of aging and social engagement. These studies provide support for the idea that self-perceptions of aging may have an effect on the experience of social engagement in older adults. In a study conducted by Menkin et al. (Reference Menkin, Robles, Gruenewald, Tanner and Seeman2016), data from the Baltimore Experience Corps Trial of adults aged 60 years and older were examined at baseline, 12 and 24 months later. The authors used the 12-item Expectations Regarding Aging Survey (Sarkisian et al., Reference Sarkisian, Steers, Hays and Mangione2005), and also asked their participants how many participants they had encountered since the beginning of the study, and how many of these people they considered close. Participants were also asked to indicate how many people they felt they could turn to for help and to report the number of close friends they had. Their level of received support and their desire for additional support were also measured. Similar to Pikhartova et al.’s (Reference Pikhartova, Bowling and Victor2016) findings, positive expectations of aging were predictive of making new friends and of reporting higher levels of perceived social support.
Lately, a new study (Schwartz et al., Reference Schwartz, Ayalon and Huxhold2021) examined if bidirectional associations of perceptions of aging exist between self-perceptions of aging and social involvement, using data from the 2008 and 2014 waves of the German Ageing Survey (DEAS). Social involvement was measured by assessing both informal social involvement (i.e., the frequency of providing advice and emotional support to others) and formal social involvement (i.e., the number of groups and organizations one is affiliated to and the frequency of attending them). The results showed that the relationships between these two variables were bidirectional, so that positive self-expectations regarding aging led to more social involvement and vice versa.
In light of these studies, the mechanism behind the moderating role of subjective age in the connection between close social relationships and loneliness can be described as follows: positive/negative self-expectations regarding one’s old age (as measured by subjective age) may lead to the activation of effective/ineffective social engagement strategies, which can lead to positive/negative perceptions of close social relationship, which may then lead to decreased/increased sense of loneliness. The decreased/increased sense of loneliness may then affect subjective age and thereby create an ongoing cycle.
In conclusion, the study by Spitzer and colleagues (Reference Spitzer, Segel-Karpas and Palgi2019) can be regarded as an initial yet important step toward understanding the effect that self-perceptions of aging might have on social engagement and loneliness. Their findings have both theoretical and practical implications. Moreover, their initial inquiry provides important research directions for future studies that will examine possible trajectories leading from subjective age to loneliness.
You have
Access
The prediction of late-life loneliness is a highly important research endeavor, since loneliness has detrimental effects on older adults’ health and mental health (e.g., Blazer, Reference Blazer2020). While objective factors, such as social isolation constrictions, may increase late-life loneliness (for the effect that COVID-19 lockdown policies had on late-life loneliness, see Hoffart et al., Reference Hoffart, Johnson and Ebrahimi2020), loneliness is basically a subjective phenomenon, defined as the gap between the individual’s desired and actual levels of social relations (de Jong Gierveld et al., Reference de Jong Gierveld, van Tilburg and Dykstra2018). Subjective age and self-perceptions of aging are two forms of views on aging (Wurm et al., Reference Wurm, Diehl, Kornadt, Westerhof and Wahl2017), which share a similar characteristic with late-life loneliness, as these three concepts are ingrained in older adults’ experience of old age. In similar to the definition of loneliness, subjective age is measured by the age one feels as related to his/her actual age (Kotter-Gruhn et al., Reference Kotter-Gruhn, Kornadt and Stephan2016), and self-perceptions of aging are conceptualized as the subjective views and expectations that individuals have about their actual aging process (Wurm et al., Reference Wurm, Diehl, Kornadt, Westerhof and Wahl2017). It is therefore important to examine the role of subjective age and self-perceptions of aging in predicting late-life loneliness.
Spitzer et al.’s (Reference Spitzer, Segel-Karpas and Palgi2019) study is focused on this important niche of investigation. Their study contributes an additional empirical evidence to the role of subjective age in predicting late-life loneliness. In their work, the authors investigated if subjective age moderates the association between subjective perceptions of close social relationships and loneliness among people at the second half of life. Spitzer and her colleagues collected data between March and July 2017 using an in-region random digit dialing methodology. Their final sample consisted of 853 Israeli participants who completed most of the survey (were interviewed either in Hebrew or Arabic, according to their mother’s tongue), from which they selected 360 married or cohabiting community-dwelling participants at the age range of 50 to 97 years. Participants’ close social relationships level was measured by Schuster et al.’s (Reference Schuster, Kessler and Aseltine1990) 3-item scale, subjective age was measured by a 4-item adaptation of Kastenbaum et al.’s (Reference Kastenbaum, Derbin, Sabatini and Artt1972) scale, and loneliness was assessed by the 3-items of the Revised UCLA loneliness Scale (R-UCLA) (Hughes et al., Reference Hughes, Waite, Hawkley and Cacioppo2004). Analyzing these data, the authors reconstructed findings from previous studies. In line with the literature (Keyes and Westerhof, Reference Keyes and Westerhof2012), the mean subjective age of their participants as measured on a scale ranging from 1 (“much younger than my age”) to 5 (“much older than my age”) with 3 (“as my age”), was younger than the participants’ actual age (i.e., M = 2.42; SD = 0.66). Moreover, both close social relationships and older subjective age were positively associated with increased loneliness, although these associations were small. But more importantly, they found that participants with an older subjective age did not demonstrate a significant negative association between their level of close social relationships and their loneliness level, whereas participants with a younger subjective age did.
Before discussing these interesting findings, it is important to view the contribution of this study in the context of its limitations. First, their use of a cross-sectional study design is a lacuna which, as the authors themselves admit, prevents any conclusions about causality. Therefore, the explanations suggested by the authors for the moderating effect of subjective age (see below) can only be regarded as directions for future research. Moreover, probably due to space limitations (the article is a brief report), the authors did not provide statistical comparisons between participants who attrited from the survey and participants who completed the survey, and we therefore lack this information about the representativeness of their sample. Additionally, while the rational for selecting participants above the age of 50 years is clear, the rational for choosing married, or cohabiting participants, is missing. Furthermore, their participants ranged between 50 and 97 years. This is a very wide age range for measuring loneliness, because loneliness during the eighth and ninth decades is much stronger than in previous decades (for a review see Donovan and Blazer, Reference Donovan and Blazer2020). Finally, the use of very short unidimensional measures of loneliness and close social relationships did not allow for a more comprehensive investigation of the effect of subjective age on loneliness, as the authors also admit.
Nevertheless, the authors’ discussion on potential mechanisms that may explain the moderation effect of subjective age provides several insights. First, following the natural tendency of people to perceive themselves as younger than their chronological age (Keyes and Westerhof, Reference Keyes and Westerhof2012), the authors suggested that people with an older subjective age may feel alienated from their social peers who perceive themselves as younger. Second, in line with previous findings, showing that subjective age also serves as an indicator for people’s physical and mental health condition (Stephan et al., Reference Stephan, Sutin and Terracciano2018), they suggested that an older subjective age may be associated with worse physical and mental health and that health problems can make it harder for older adults to benefit from close social relations. These reasonable explanations can be researched by using longitudinal studies that will examine the mediating roles of feelings of alienation and physical and mental health in the connection between close social relationship and loneliness among participants with an older and a younger subjective age, using moderated mediation models.
Spitzer et al.’s (Reference Spitzer, Segel-Karpas and Palgi2019) study has several strengths. From the theoretical point of view, this study emphasizes the importance of self-perceptions of aging (in this case of subjective age) in affecting the painful experience of loneliness. In view of the bulk of studies which focus on loneliness as predicted by objective circumstances, such as social isolation (see review by Leigh-Hunt et al., Reference Leigh-Hunt2017), the focus of the authors on the subjective aspects of both loneliness, close social relations, and subjective age is important, as it leaves room for psychological interventions aimed at improving these subjective perspectives. Therefore, their study calls for future studies to construct and examine psychological interventions that will focus on people with an older subjective age and improve their subjective perceptions of aging, so they will be able to enjoy close social relations and decrease their sense of loneliness. Moreover, as the authors argue, their study adds an important tier to the socio-emotional selectivity theory (Carstensen, Reference Carstensen1992), which claims that the ability to focus on meaningful social friendships improves with age. Their findings may imply that this social ability improves among older adults with a younger subjective age, but not among those with an older subjective age.
Before closing this commentary, I would like to elaborate on another directive for future investigations. A review of the literature on self-perceptions of aging and late-life loneliness which corresponds well with their results may offer several insights regarding the mechanism that enables subjective age to moderate the connection between close social relationships and late-life loneliness. This mechanism may operate by activating self-expectations regarding old age. In this regard, Pikhartova et al., (Reference Pikhartova, Bowling and Victor2016) have demonstrated that one of the basic tenets that older adults hold regarding old age is that it is a time of loneliness. Pikhartova and colleagues (2016) examined data from several waves of the English Longitudinal Study of Ageing (ELSA). These data included the relationship between perceived stereotypes about old age and the expectation for late-life loneliness among participants who were 50 years old and over in 2004, and their actual loneliness reported within 8 years of follow-up (throughout the years 2006–2021). They found that 24% of their participants in 2004 (N = 4,465) agreed that old age is a time of loneliness and 33% expected to be lonely in old age. They also found that both stereotypes and expectations significantly predicted future loneliness eight years later. These findings are in line with the stereotype embodiment theory (Levy, Reference Levy2009), which argues that stereotypes about aging are internalized throughout the life cycle and become self-expectations. In other words, it is possible that people who feel older than their actual age also tend to believe that old age is a time of loneliness and are therefore inclined to report feeling lonelier. Moreover, their internalized negative expectation about loneliness being an integral part of their older subjective age may hamper their ability to utilize close relationships as a way to decrease their loneliness. Future studies may examine the association between this negative expectation and an older subjective age.
There are also additional studies that did not consider the specific expectation regarding the existence of loneliness in old age, but focused on the association between other self-perceptions of aging and social engagement. These studies provide support for the idea that self-perceptions of aging may have an effect on the experience of social engagement in older adults. In a study conducted by Menkin et al. (Reference Menkin, Robles, Gruenewald, Tanner and Seeman2016), data from the Baltimore Experience Corps Trial of adults aged 60 years and older were examined at baseline, 12 and 24 months later. The authors used the 12-item Expectations Regarding Aging Survey (Sarkisian et al., Reference Sarkisian, Steers, Hays and Mangione2005), and also asked their participants how many participants they had encountered since the beginning of the study, and how many of these people they considered close. Participants were also asked to indicate how many people they felt they could turn to for help and to report the number of close friends they had. Their level of received support and their desire for additional support were also measured. Similar to Pikhartova et al.’s (Reference Pikhartova, Bowling and Victor2016) findings, positive expectations of aging were predictive of making new friends and of reporting higher levels of perceived social support.
Lately, a new study (Schwartz et al., Reference Schwartz, Ayalon and Huxhold2021) examined if bidirectional associations of perceptions of aging exist between self-perceptions of aging and social involvement, using data from the 2008 and 2014 waves of the German Ageing Survey (DEAS). Social involvement was measured by assessing both informal social involvement (i.e., the frequency of providing advice and emotional support to others) and formal social involvement (i.e., the number of groups and organizations one is affiliated to and the frequency of attending them). The results showed that the relationships between these two variables were bidirectional, so that positive self-expectations regarding aging led to more social involvement and vice versa.
In light of these studies, the mechanism behind the moderating role of subjective age in the connection between close social relationships and loneliness can be described as follows: positive/negative self-expectations regarding one’s old age (as measured by subjective age) may lead to the activation of effective/ineffective social engagement strategies, which can lead to positive/negative perceptions of close social relationship, which may then lead to decreased/increased sense of loneliness. The decreased/increased sense of loneliness may then affect subjective age and thereby create an ongoing cycle.
In conclusion, the study by Spitzer and colleagues (Reference Spitzer, Segel-Karpas and Palgi2019) can be regarded as an initial yet important step toward understanding the effect that self-perceptions of aging might have on social engagement and loneliness. Their findings have both theoretical and practical implications. Moreover, their initial inquiry provides important research directions for future studies that will examine possible trajectories leading from subjective age to loneliness.
Conflict of interest
None.