Feeling lonely when not socially isolated

Abstract

Although it is widely acknowledged that people can feel lonely when not socially isolated, to our knowledge, no study has examined whether loneliness would be differently associated with daily social relations for people who are not socially isolated. The present study examined the role of social isolation (i.e., small social network size) in moderating the association between loneliness and characteristics of daily social interactions—interaction type and qualitative characteristics—using the 7-day social interaction diary data of 118 individuals. The moderating effect of social isolation on self-informant agreement in loneliness ratings was additionally examined, using three informant ratings of loneliness. Greater loneliness was more related to less frequency of strong tie interactions for people who are less socially isolated, while loneliness was more associated with greater self-focus during interactions for more socially isolated people. In addition, for those who are less socially isolated, the self-informant agreement in loneliness ratings was lower, suggesting that their loneliness might be underrecognized, even by those close to them. These results indicate that the relationships between loneliness and such social variables may operate differently depending upon the personal level of social isolation. In particular, our findings underscore the significant role of frequent interactions with close ties in alleviating the loneliness of people who are not socially isolated, suggesting that human social needs cannot be satisfied merely by a large number of social contacts.

Keywords

Informant ratings loneliness self-absorption social interaction social isolation social network strong ties

When people perceive their social relationships as inadequate, they suffer from a tormenting feeling which is called loneliness (Perlman & Peplau, 1981). Feeling lonely is not just emotionally painful but literally deleterious to human health. There is growing evidence that loneliness is a significant risk factor for a wide range of physical health problems encompassing fragmented sleep (Cacioppo et al., 2002) and heightened mortality (Holt-Lunstad, Smith, Baker, Harris, & Stephenson, 2015; Luo, Hawkley, Waite, & Cacioppo, 2012) as well as mental health problems such as depressive symptoms (VanderWeele, Hawkley, Thisted, & Cacioppo, 2011) and cognitive decline (Holwerda et al., 2014). In addition to the harmful effects on health, loneliness is not an unusual experience that only a few suffer from. For example, 33% of adults reported they had experienced loneliness in the past month in a New Zealand survey (Statistics New Zealand, 2013), and 35% of middle- and old-aged adults were classified as lonely among Americans (Wilson & Moulton, 2010). All things considered, loneliness is an urgent health issue that commands clinical attention (Heinrich & Gullone, 2006).

Prior studies have conceptually distinguished loneliness from social isolation, suggesting that loneliness is not the exclusive experience of socially isolated people (de Jong Gierveld, Van Tilburg, & Dykstra, 2006; Perlman & Peplau, 1984). While social isolation refers to an objective state of lack of social contacts, manifested by quantifiable features such as small social network size or infrequent social interactions (Cornwell & Waite, 2009; Steptoe, Shankar, Demakakos, & Wardle, 2013), loneliness refers to an emotional state arising from the subjective perception that one’s social relationships are deficient either in quantitative or in qualitative ways (Perlman & Peplau, 1981).

This conceptual distinction has been supported by empirical data. The correlation between loneliness and social isolation was found to be weak to moderate (Cornwell & Waite, 2009; Coyle & Dugan, 2012), which suggests that they are related yet independent constructs. More specifically, Matthews and his colleagues (2016) showed that the 49% of individuals ranked in the top quartile for social isolation were not in the top quartile for loneliness, while the 53% of individuals ranked in the top quartile for loneliness were not in the top quartile for isolation. These findings suggest that social isolation and loneliness do not always co-occur. In other words, some may feel lonely even with abundant social connections (Perlman & Peplau, 1984).

Why then do some individuals feel lonely when they are not socially isolated? Deficiency in quality of social relationships was often demonstrated to induce a feeling of loneliness in the presence of others (Cacioppo, Grippo, London, Goossens, & Cacioppo, 2015; Gierveld, 1998). In particular, daily diary studies have found that loneliness is significantly linked to qualitative aspects of daily social interactions such as a lack of interactions with intimate others (Wheeler, Reis, & Nezlek, 1983) and poor emotional quality of interaction (Hawkley, Burleson, Berntson, & Cacioppo, 2003). This suggests that the way a person forms relations with others in his or her daily life is a significant determinant of feeling lonely. If a person forms mainly superficial interactions with distant others, he or she may feel isolated even when possessing numerous social contacts. What is lacking in the daily interactions of such a person may be distinct from the quantitative deficiency a socially isolated person may have. Thus, to gain a better understanding of the differential aspects of daily social interactions associated with the loneliness of people not socially isolated, the present study investigated the role of social isolation in moderating the association between loneliness and the characteristics of daily social interactions.

Interaction type: Who to interact with?

Individuals interact in their daily lives with social ties who compose personal social networks. Weiss (1973) posited that different types of social ties meet specific relational needs, and therefore, the absence or loss of specialized relationships might lead to distinct loneliness. Based on this assumption, he drew a distinction between two types of loneliness: loneliness of social isolation (i.e., social loneliness), resulting from the lack of an engaging social network, and loneliness of emotional isolation (i.e., emotional loneliness), resulting from the absence of close attachment relationships. According to his theory, specific social ties may have different roles in the experience of loneliness.

Broadly, a personal social network can be divided into strong and weak ties by the level of intimacy or the frequency of social contact (Granovetter, 1983). Strong ties refer to intimate and close relationships, which occupy the core layer of one’s social network. Although they require more social resources (e.g., time and emotional commitment) to maintain, they can provide more stable and broader social support (Roberts, Dunbar, Pollet, & Kuppens, 2009). In contrast, weak ties, peripheral relationships with less frequent contact and limited intimacy, require only a minimal amount of social resources to form and maintain, but they can deteriorate easily with time (Burt, 2002) and provide less emotional support and companionship. People can use their limited social resources to maintain a few strong ties or form multiple weak ties (Sutcliffe, Dunbar, Binder, & Arrow, 2012).

Weak ties may play a temporary role as an alternative to strong ties, particularly when one’s belongingness is thwarted (Cacioppo & Patrick, 2008). Weak tie interaction was found to contribute to a greater sense of belonging and less loneliness, especially when strong tie interactions were fewer, suggesting weak ties can be a buffer to social fluctuations by diversifying one’s social network (Sandstrom & Dunn, 2014). Still, given weak ties interact infrequently and exchange limited intimacy, they may not replace strong ties which can provide stable and sturdy support.

Several studies show that loneliness is not consistently related to the total number of social ties but rather to the number of people one feels close to (Stokes, 1985) and to the average intimacy felt through social ties (Green, Richardson, Lago, & Schatten-Jones, 2001). Specifically, Jones (1981) showed that a lower proportion of interactions with close friends and family was related to loneliness while a higher proportion of interactions with acquaintances and strangers was not, implying that interactions with nonclose ties may be insufficient to satisfy one’s social needs.

Viewed thus, merely a large number of ties may not be a sufficient buffer against loneliness, and people with a large-scale social network may feel lonely when their daily relations are weighted toward superficial and temporary relationships with a lack of close ones. Therefore, a lack of interactions with strong ties may be particularly more associated with loneliness for people who have larger social networks.

Qualitative characteristics of interactions: How to interact with others

The extant literature indicates that the qualitative characteristics of social interaction are closely linked to loneliness. The emotional experience during interactions, such as greater positive emotion (Hawkley et al., 2003) and meaningfulness (Wheeler et al., 1983), were consistently found to be related to a lower degree of loneliness. The extent of self-disclosure in interactions was also found to be related to loneliness. Loneliness was found to be associated with the tendency of disclosing less of oneself to others (Stokes, 1987; Wei, Russell, & Zakalik, 2005) or showing different patterns of disclosure, either disclosing too much or very little (Solano, Batten, & Parish, 1982). Moreover, reciprocal self-disclosure produced a better quality of interaction than nonreciprocal disclosure (Sprecher, Treger, Wondra, Hilaire, & Wallpe, 2013), which may be relevant to a decreased level of loneliness.

It was also suggested that loneliness was associated with the tendency to be excessively self-focused (i.e., self-absorbed) when interacting with others (see the review, Heinrich & Gullone, 2006). For instance, in Jones’s study (1982), lonely people interacted with others in a more self-absorbed manner, and when instructed to increase partner attention, their loneliness decreased. Based on the findings, he argued that the lack of partner attention was the core social skill deficit of lonely individuals. Neuroimaging data also supported their self-focused attention to social stimuli, showing that they attend less to others’ perspectives (Cacioppo, Norris, Decety, Monteleone, & Nusbaum, 2009).

Prior studies have shown that such qualitative aspects of daily interactions with others are closely linked to feeling of loneliness. In particular, people who feel lonely in spite of more quantitative social connections may interact with others in ways that are ineffective against loneliness. Accordingly, their heightened loneliness might be more strongly associated with those qualitative aspects.

The present study

The current study examined the moderating role of personal level of social isolation on the associations between loneliness and interaction type and qualitative characteristics in one’s daily life. By studying social interactions of everyday life, we attempt to acquire information about participants’ actual patterns of relations with others within their social capacity and eventually to understand their association with loneliness.

In order to obtain measures of both the quantity and quality of individuals’ daily social encounters, we asked participants to complete a 7-day interaction diary. The advantage of this daily event self-recording is that it can provide more accurate and diverse information about naturally occurring social experiences (Wheeler & Reis, 1991). For measuring the frequency of each interaction type, participants were asked to distinguish whether each interaction was either with strong ties or weak ties (i.e., interaction type) and to keep track of the number of each interaction type employing two methods: the Rochester Interaction Record (RIR; Wheeer & Nezlek, 1977), only for interactions that last 10 min or longer and mechanical tally counters (clickers; Sandstrom & Dunn, 2014) to capture every momentary interaction. Using clickers is particularly useful for capturing short interactions with others, while using the RIR is practical for assessing the qualitative aspects of each interaction. Using the RIR, participants assessed the qualitative characteristics of each interaction in three dimensions: emotional experience, reciprocal self-disclosure, and self-absorbed states during the interaction.

Additionally, we tested the moderating effect of social isolation on the self-informant agreement in loneliness ratings. Lonely persons who are less socially isolated may seemingly have more adequate quantities of social relations, and therefore, their perceived social isolation may be less conspicuous to others than that of more socially isolated individuals. For testing this, we collected informant ratings of loneliness from people close to each participant.

The present study seeks to extend previous work on the associations between loneliness and daily social interactions by examining whether the associations differ depending on one’s level of social isolation. By identifying the unique associations between daily social experience and loneliness for less socially isolated individuals, the present study can help understand how social needs remain unfulfilled, even when one possesses a sufficient amount of social relations. In addition, it can help inform interventions targeting individuals who suffer from loneliness in a large social world.

Method

Participants

One hundred eighteen participants (80 females) completed all the procedures for this study. Participants ranged from 19 to 64 years of age (M = 33.66, SD = 9.58), and all were Koreans. This study was described as an investigation about social networks and advertised via flyers at local community centers and a university, along with personal contacts and referrals. In exchange for their participation, they received psychological assessment services after completing the study.

Procedures

Participants attended an orientation session, where they were informed about the procedures and rewards of this study and were instructed on how to keep a social interaction diary. Only participants who provided informed consent were given tally counters to count the number of interactions. Participants could choose to do the surveys either online or on paper.

In the first stage, participants responded to measures of loneliness and social network size. Concurrently, to collect informant ratings of social isolation, each participant was asked to name three informants who were well acquainted with the participant’s recent life. The majority of participants were rated confidentially by three informants such as family members and friends (M = 2.97, SD = 0.16).

Next, the participants were asked to complete a daily social interaction record at the end of each day for 7 consecutive days, using two methods: One was mechanical tally counters (clickers) to capture every momentary interaction, and the other was the Rochester Interaction Record (RIR) to describe interactions which lasted 10 or more minutes. They were contacted twice a day and were reminded in the morning to carry clickers with them and were reminded at night to complete the RIR. The study protocol was approved by the institutional review board at Korea University (KU-IRB-15-96-P-2).

Materials

Loneliness was assessed with the Revised UCLA Loneliness Scale (R-UCLA Scale; Peplau & Cutrona, 1980). This 20-item measure assesses one’s subjective feelings of loneliness. Responses range from 1 (never) to 4 (often); α = .91. For informant ratings, a modified form of this scale was used, adapted for informant ratings (α = .93).

Social network size was assessed with the Social Network Index (SNI; Cohen, Doyle, Skoner, Rabin, & Gwaltney, 1997) designed to assess participation in 12 types of social relationships, including spouse, parent, child, child-in-law, close relative, close friend, church/temple member, student, employee, neighbor, volunteer, and group member. Participants were asked to count the number of people they spoke to on a regular basis in each type of social relationship (e.g., close friends: “How many of these friends do you see or talk to at least once every 2 weeks?”). Social network size was computed by summing the number of people in the social network across every type of social relationship, indicating the total number of people who one regularly contacts at least once every 2 weeks.

Social interaction diary

Social interaction was defined as “any encounter with another person (or people) in which participants attended to one another and adjusted their behavior in response to one another” (Wheeler & Nezlek, 1977, p. 743), and examples of what was considered an interaction (e.g., a conversation) and what was not (e.g., listening to a lecture) were provided in detail.

To identify interaction type, participants were asked to categorize all social ties with whom they interacted as either strong or weak ties. Participants were also given detailed instructions on the criteria of strong ties and weak ties, respectively, with several examples, following the guidelines of Sandstrom and Dunn (2014). For instance, strong ties were “who you are very close to,” “who you know (and knows you) very well,” and “who you confide your problems in and ask for help,” while weak ties were the opposite. Each diary consisted of the following:

Number of interactions recorded with tally counter (clickers)

Using the same method as Sandstrom and Dunn (2014) to count the number of daily interactions, participants were asked to carry two colored clickers (red for strong ties and blue for weak ties) to count each time they initiated a new interaction with someone in their social network. This was used to capture even short interactions at the moment they happened, since they can easily be forgotten by the end of the day. Participants were then asked to write down the number of both types of social interactions before going to bed.

In the original data collected from 118 participants, a few outliers outside the general scope of the others’ observations were detected. Considering a possible misunderstanding on how to count new interactions with the same partner, five outliers who scored more than 1.5 IQR (interquartile ranges) below Q1 or 1.5 IQR above Q3 on total numbers of interactions were excluded from our data analysis. Accordingly, 113 participants’ records were analyzed for interactions recorded with clickers.

Rochester Interaction Record

A modified version of the RIR was used. We revised questionnaires from Berry and Hansen (1996) for our research questions. Participants recorded details about each interaction, involving date, time, the length of the interaction, and information about partner(s) (e.g., strong or weak ties). On an 8-point Likert-type scale, they responded to two qualitative descriptions which asked to what extent they and their partner(s) disclosed themselves (i.e., reciprocal self-disclosure) and to four descriptions which asked to what extent they felt pleasant, comfortable, meaningful, and connected (i.e., positive emotions).

We also added 4 items to measure excessive, sustained, and rigid self-focused attention (i.e., self-absorption) during interactions, including 2 items with the highest loadings on two factors (i.e., public self-absorption and private self-absorption), derived from the Self-Absorption Scale (McKenzie & Hoyle, 2008). Participants were asked to complete a distinct record of each interaction; consequently, the number of daily records per day varied across participants depending on the number of interactions which lasted 10 min or longer (M = 3.33, SD = 1.54).

Results

Correlation analyses

Table 1 shows correlations of loneliness and social isolation with all outcome variables included in this analysis. Loneliness and social network size were weakly negatively correlated, r(116) = −.23, p = .011. Greater loneliness was correlated with a smaller number of strong tie interactions and lower levels of positive emotional experience and reciprocal self-disclosure during interactions but greater levels of self-absorption and informant-rated loneliness scores. Loneliness was unrelated to the number of weak tie interactions. Smaller social network size was correlated with a smaller number of both strong and weak tie interactions counted with clickers and strong tie interactions recorded with the RIR. Social network size was not significantly associated with the qualitative characteristics of social interactions and informant ratings of loneliness.

Table 1.
Correlation analyses and descriptive statistics (N = 118).

M (SD) Loneliness Social network size

Loneliness 38.56 (9.60) – –

Social network size 24.94 (13.33) −. 23* –

Clickers-strong tie interactions^a,b 4.95 (3.33) −.41 .27

Clickers-weak tie interactions^a,b 6.87 (5.23) −.18 .29

RIR-strong tie interactions^a 2.37 (1.36) −.26 .37*

RIR-weak tie interactions^a 0.96 (0.79) −.02 .11

RIR-Positive emotional experience 5.87 (0.91) −.43* .08

RIR-Reciprocal self-disclosure 6.11 (0.88) −.38* .03

RIR-Self-absorption 2.31 (0.83) .34* −.14

Informant ratings of loneliness 32.61 (6.39) .55*** −.16

^aThe average number of interactions per day.

^b n = 113.

p < .05, p < .01, p< .001.

Moderation analyses

To test the moderating effect of social isolation (i.e., small social network size), we conducted separate multiple regression analyses for different social variables (interaction type and qualitative characteristics and informant ratings of loneliness) predicted by loneliness, social network size, and their interactions. The main effects with each mean-centered variable and the multiplicative interaction of both variables were entered simultaneously. Tables 2 to 4 present results from eight multiple linear regressions. The significant interaction effects were found in predicting the number of strong tie interactions counted both by clickers and the RIR, self-absorption during interactions, and informant ratings of loneliness.

Table 2.
Standardized regression weight (β) of loneliness, social network size, and their interaction predicting the average number of each type of interactions per day (N = 118).

Outcome variables

RIR_Strong tie interactions RIR_Weak tie interactions clickers_Strong tie interactions^a clickers_Weak tie interactions^a

Loneliness −.222* .010 −.400* −.106

Social network size .255 .117 .104 .287**

Loneliness × Social network size −.215* .008 −.223* .058

R ² .213 .013 .240 .101

F 10.307* .490 11.464* 4.085**

^a n = 113.

p < .05, p < .01, p < .001.

Table 3.
Standardized regression weight (β) of loneliness, social network size, and their interaction predicting qualitative characteristics of interactions (N = 118).

Outcome variables

Positive emotional experience Reciprocal self-disclosure Self-absorption

Loneliness −.446* −.404* .299**

Social network size −.033 −.083 −.130

Loneliness × Social network size −.037 −.060 −.184*

R ² .191 .151 .153

F 8.954* 6.758* 6.845***

p < .05, p < .01, p < .001.

Table 4.
Standardized regression weight (β) of loneliness, social network size, and their interaction predicting informant ratings of loneliness (N = 118).

Outcome variable

Informant ratings of loneliness

Loneliness .515***

Social network size −.090

Loneliness × Social network size −.167*

R ² .327

F 18.489***

p < .05, p < .01, p < .001.

To understand the nature of the interaction effect, we conducted simple slope analyses (Aiken & West, 1991; see Figure 1). Simple slope tests revealed that loneliness was significantly associated with the smaller number of strong tie interactions for participants having a large social network (+1 SD) but not for those having a small social network (−1 SD). In contrast, the positive association between loneliness and self-absorption was significant only for those having a small social network (−1 SD) and not for those having a large social network (+1 SD). Furthermore, self-reported loneliness scores were less associated with informant reported loneliness scores for participants having a large social network (+1 SD) compared with for participants having a small social network (−1 SD).

Figure 1.
Plots of loneliness × Social network size interactions predicting the number of strong tie interactions, self-absorption, and informant ratings of loneliness.

Discussion

The present study investigated how loneliness is associated with characteristics of daily social interactions within the contexts of different personal levels of social isolation. Using daily records of social interactions and informant ratings of loneliness, we employed a multiple regression approach to test the effects of interaction between loneliness and social network size on several measures of the characteristics of daily social interactions—interaction type and qualitative characteristics—and informants’ evaluation of their loneliness. Our findings indicate that the relationships between loneliness and such social variables may operate differently depending upon the personal level of social isolation.

Consistent with prior studies (e.g., Coyle & Dugan, 2012; Matthews et al., 2016), the association between loneliness and social network size (opp. social isolation) was significant but weak (r = −.23), supporting that loneliness and social isolation do not necessarily co-occur. The relationships of these two constructs with daily social interactions were distinct as well. Social network size was related to the quantitative features of daily interactions (frequent interactions both with weak and strong ties), whereas loneliness was related to interaction type and quality; these associations accord with the conceptual distinction between loneliness and social isolation (e.g., de Jong Gierveld et al., 2006). More specifically, participants who reported a greater extent of loneliness had less frequent interactions with strong ties and rated the overall quality of their daily interactions poorly: They reported feeling less enjoyable, comfortable, meaningful, and connected, perceiving themselves and their interaction partners as more reluctant to reveal their true feelings and having a tendency to focus more on oneself while interacting with others. In our unpublished manuscript¹ (Lee & Ko, 2017), people were found to feel lonelier on days they interacted less with strong ties, felt less positive emotions, disclosed themselves less, and were more self-focused in interactions. Considered together, this suggests that the lack of interactions with close ties and overall deficiency in quality of interactions among lonely people may contribute to their daily experience of loneliness.

Our moderation analyses extend prior work on the link between loneliness and characteristics of social relationships by noting that these associations can be moderated by the personal level of social isolation. For participants who were less socially isolated (i.e., having a larger social network), a greater level of loneliness was more associated with a smaller number of strong tie interactions; this has been confirmed in the records kept with both clickers and the RIR. Specifically, this association between loneliness and the smaller number of strong tie interactions was only significant in individuals with a large social network. Our unpublished manuscript¹ (Lee & Ko, 2017) found a negative association of the daily experience of loneliness with the frequency of strong tie interactions, in contrast to the insignificant or positive association with that of weak tie interactions at the daily level, suggesting that social meetings in which a close other is not included may not be an effective buffer against loneliness. Especially for people interacting with a large number of ties, the frequency of interactions with close others might result in greater changes in feeling lonely.

Moderation analyses also noted that for participants who were less socially isolated, the self-informant agreement in loneliness ratings was lower. This suggests that the perceived isolation of individuals not socially isolated may be unrecognized and less understood, even by those close to them. The inclination of lonely people to hold in their feelings can further mask their loneliness and make it more difficult for them to be understood. The gap between what they experience inwardly and how they appear outwardly may disturb their recovery from social pain, given that feeling understood is crucial for satisfying relatedness needs (Reis, Sheldon, Gable, Roscoe, & Ryan, 2000).

Meanwhile, moderation analyses found that in the context of greater social isolation, loneliness was more associated with a greater tendency to be excessively self-focused during interactions. Specifically, the association was only significant in socially isolated people but not in people having a large social network. Considering the propensity to be self-focused was viewed as a core social skills deficit of lonely individuals (Jones, 1982), highly self-focused states during interactions of socially isolated people may point to a social ability deficit. Given that personal network size can vary due to individual differences in social capacity such as social–cognitive abilities or personalities (Dunbar, 2008), socially isolated people may have a limited social capacity to broaden their network, and this limited social ability may be associated with the loneliness of socially isolated people but not with that of those having a large social network.

The present findings shed light on the loneliness of people having a large personal social network. Our findings exhibited that although some people can suffer from loneliness while being surrounded by a large number of social ties, their emotional distress from unfulfilled social needs tends to be underrecognized by others. Given that loneliness contributes to poor mental health outcomes independent of social isolation (See the review, Cacioppo & Cacioppo, 2014), they might be a group whose vulnerable mental health is prone to being overlooked. Our moderation analyses highlighted in particular the important role of interactions with core ties in the loneliness of such socially connected people. A person has finite personal resources for social relationships (e.g., time for socializing, emotional commitment), which can cause a trade-off between personal network size and intimacy with each tie (Roberts et al., 2009). As weak tie connections require less personal resources to build than strong tie connections, which require a greater extent of emotional commitment and time (Sutcliffe et al., 2012), some lonely people may mistakenly use high amounts of their resources to establish multiple weak ties, instead of a fewer number of strong ties, to appear to be socially connected with a larger personal network in an effort to avoid the social stigma of loneliness (Lau & Gruen, 1992). However, as our findings suggest, even when connected to a large social network, if they do not interact sufficiently with the close ties with whom they could share their life, the quantitative scale of their social network may not compensate for its qualitative costs. Frequent and stable interaction with significant ties may be a prerequisite to fulfill one’s social needs and navigate a broader social world (e.g., Baumeister & Leary, 1995).

Furthermore, as the first study testing the effect of interaction between social isolation and loneliness, our findings noted the differential aspects of daily social interactions that are more related to a person’s loneliness according to the personal level of social isolation. While a lack of strong tie interactions is more related to the loneliness of less socially isolated people, a self-absorbed state in interactions is more associated with the loneliness of more socially isolated people. This suggests that underlying factors and mechanisms triggering loneliness may differ depending upon the personal level of social isolation. Therefore, different intervention targets may be required for addressing the loneliness of socially isolated and socially connected people. For socially isolated people, providing social skills training to expand personal resources for social relationships could be more effective. While, for people who feel lonely but not socially isolated, the aim could be to recompose existing bonds by utilizing their resources for social relationships, which they may already have, to strengthen their close ties.

These findings may accord with Weiss’s typology of loneliness (1973), which differentiated between emotional loneliness and social loneliness. Particularly, the loneliness of people connected to a larger social network, on which the present study focused, may correspond to emotional loneliness that, he postulated, must be the response to an absence of satisfying close relationships and, therefore, can only be alleviated by the installment or replacement of close relationships. However, using a one-dimensional loneliness measure, the present study cannot test his assumption that the two types of loneliness may have different symptomatological patterns such that emotional loneliness produces anxiety and oversensitivity whereas social loneliness results in depression and boredom. Further study is needed to understand whether lonely people with a larger social network would have different loneliness symptoms than those with a smaller social network.

Limitations

First, our analyses to examine the relationships of each characteristic of social interaction with loneliness were correlational and so cannot determine causality. Although we tentatively interpreted the results in a way that the features of daily interaction may have contributed to the level of loneliness, loneliness could conversely have affected interactions. For example, people can be unwilling to initiate more interactions with strong ties because they feel lonely. Therefore, this study cannot confirm a causal relationship between the qualitative features of daily interactions and loneliness.

Furthermore, the representativeness of this study population can be limited by its relatively small sample size (N = 118) and particularly a fewer number of men (n = 38). In addition, it should be noted that the participants may underrepresent severely socially isolated people for two reasons. First, we required every participant to identify three informants who were well acquainted with the participant’s recent conditions to judge their loneliness. Therefore, severely isolated people who could not name three people close to them could have been excluded from the study’s recruitment. In addition, we advertised for participants at community centers and a university as well as via personal contacts and referrals. Therefore, it is possible that we could not reach severely socially isolated people with little to no social participation and/or social contacts.

	M (SD)	Loneliness	Social network size
Loneliness	38.56 (9.60)	–	–
Social network size	24.94 (13.33)	−. 23*	–
Clickers-strong tie interactions^a,b	4.95 (3.33)	−.41**	.27**
Clickers-weak tie interactions^a,b	6.87 (5.23)	−.18	.29**
RIR-strong tie interactions^a	2.37 (1.36)	−.26**	.37***
RIR-weak tie interactions^a	0.96 (0.79)	−.02	.11
RIR-Positive emotional experience	5.87 (0.91)	−.43***	.08
RIR-Reciprocal self-disclosure	6.11 (0.88)	−.38***	.03
RIR-Self-absorption	2.31 (0.83)	.34***	−.14
Informant ratings of loneliness	32.61 (6.39)	.55***	−.16

	Outcome variables
Loneliness	−.222*	.010	−.400***	−.106
Social network size	.255**	.117	.104	.287**
Loneliness × Social network size	−.215*	.008	−.223*	.058
R ²	.213	.013	.240	.101
F	10.307***	.490	11.464***	4.085**

	Outcome variables
Loneliness	−.446***	−.404***	.299**
Social network size	−.033	−.083	−.130
Loneliness × Social network size	−.037	−.060	−.184*
R ²	.191	.151	.153
F	8.954***	6.758***	6.845***

	Outcome variable
Loneliness	.515***
Social network size	−.090
Loneliness × Social network size	−.167*
R ²	.327
F	18.489***

Footnotes

Authors’ note

Portions of this article were submitted as the first author’s Master’s thesis to Korea University. Part of the data of this article was presented at the 28th Annual Convention of the Association for Psychological Science, Chicago, IL, USA.

Acknowledgment

The authors thank the members of Happiness Psychology Laboratory of Korea University, including Youngmi Na and Subin Cho, for their contribution to the participant management and data collection.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Note

References

Aiken

L. S.

West

S. G.

(1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage.

Baumeister

R. F.

Leary

M. R.

(1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117, 497–529. doi:10.1037/0033-2909.117.3.497

Berry

D. S.

Hansen

J. S.

(1996). Positive affect, negative affect, and social interaction. Journal of Personality and Social Psychology, 71, 796–809. doi:10.1037/0022-3514.71.4.796

Burt

R. S.

(2002). Bridge decay. Social Networks, 24, 333–363. doi:10.1016/S0378-8733(02)00017-5

Cacioppo

J. T.

Cacioppo

(2014). Social relationships and health: The toxic effects of perceived social isolation. Social and Personality Psychology Compass, 8, 58–72. doi:10.1111/spc3.12087

Cacioppo

J. T.

Hawkley

L. C.

Berntson

G. G.

Ernst

J. M.

Gibbs

A. C.

Stickgold

Hobson

J. A.

(2002). Do lonely days invade the nights? Potential social modulation of sleep efficiency. Psychological Science, 13, 384–387. doi:10.1111/1467-9280.00469

Cacioppo

J. T.

Norris

C. J.

Decety

Monteleone

Nusbaum

(2009). In the eye of the beholder: Individual differences in perceived social isolation predict regional brain activation to social stimuli. Journal of Cognitive Neuroscience, 21, 83–92. doi:10.1162/jocn.2009.21007

Cacioppo

J. T.

Patrick

(2008). Loneliness: Human nature and the need for social connection. New York, NY: Norton.

Cacioppo

Grippo

A. J.

London

Goossens

Cacioppo

J. T.

(2015). Loneliness: Clinical import and interventions. Perspectives on Psychological Science, 10, 238–249. doi:10.1177/1745691615570616

10.

Cohen

Doyle

W. J.

Skoner

D. P.

Rabin

B. S.

Gwaltney

J. M.

(1997). Social ties and susceptibility to the common cold. Journal of the American Medical Association, 277, 1940–1944. doi:10.1001/jama.1997.03540480040036

11.

Cornwell

E. Y.

Waite

L. J.

(2009). Social disconnectedness, perceived isolation, and health among older adults. Journal of Health and Social Behavior, 50, 31–48. doi:10.1177/002214650905000103

12.

Coyle

C. E.

Dugan

(2012). Social isolation, loneliness and health among older adults. Journal of Aging and Health, 24, 1346–1363. doi:10.1177/0898264312460275

13.

de Jong Gierveld

Van Tilburg

Dykstra

P. A.

(2006). Loneliness and social isolation. In Vangelisti

A. L.

Perlman

(Eds.), The Cambridge handbook of personal relationships (pp. 485–499). New York, NY: Cambridge University Press.

14.

Dawson

(2017). Interpreting interaction effects. Retrieved March 3, 2017, from http://www.jeremydawson.com/slopes.htm

15.

Dunbar

(2008). Cognitive constraints on the structure and dynamics of social networks. Group Dynamics: Theory, Research, and Practice, 12, 7–16. doi:10.1037/1089-2699.12.1.7

16.

Gierveld

J. D. J.

(1998). A review of loneliness: Concept and definitions, determinants and consequences. Reviews in Clinical Gerontology, 8, 73–80.

17.

Granovetter

(1983). The strength of weak ties: A network theory revisited. Sociological Theory, 1, 201–233. doi:10.2307/202051

18.

Green

L. R.

Richardson

D. S.

Lago

Schatten-Jones

E. C.

(2001). Network correlates of social and emotional loneliness in young and older adults. Personality and Social Psychology Bulletin, 27, 281–288. doi:10.1177/0146167201273002

19.

Hawkley

L. C.

Burleson

M. H.

Berntson

G. G.

Cacioppo

J. T.

(2003). Loneliness in everyday life: Cardiovascular activity, psychosocial context, and health behaviors. Journal of Personality and Social Psychology, 85, 105–120. doi:10.1037/0022-3514.85.1.105

20.

Heinrich

L. M.

Gullone

(2006). The clinical significance of loneliness: A literature review. Clinical Psychology Review, 26, 695–718. doi:10.1016/j.cpr.2006.04.002

21.

Holt-Lunstad

Smith

T. B.

Baker

Harris

Stephenson

(2015). Loneliness and social isolation as risk factors for mortality: A meta-analytic review. Perspectives on Psychological Science, 10, 227–237. doi:10.1177/1745691614568352

22.

Holwerda

T. J.

Deeg

D. J.

Beekman

A. T.

van Tilburg

T. G.

Stek

M. L.

Jonker

Schoevers

R. A.

(2014). Feelings of loneliness, but not social isolation, predict dementia onset: Results from the Amsterdam Study of the Elderly (AMSTEL). Journal of Neurology, Neurosurgery, & Psychiatry, 85, 135–142. doi:10.1136/jnnp-2012-302755

23.

Jones

W. H.

(1981). Loneliness and social contact. The Journal of Social Psychology, 113, 295–296. doi:10.1080/00224545.1981.9924386

24.

Jones

W. H.

(1982). Loneliness and social behavior. In Peplau

L. A.

Perlman

(Eds.), Loneliness: A sourcebook of current theory, research and therapy (pp. 238–252). New York, NY: Wiley.

25.

Lee

Y. G.

(2017). Association between characteristics of social interactions and daily experience of loneliness: A daily diary study. Unpublished manuscript, Department of Psychology, Korea University, Seoul, South Korea.

26.

Lau

Gruen

G. E.

(1992). The social stigma of loneliness: Effect of target person’s and perceiver’s sex. Personality and Social Psychology Bulletin, 18, 182–189. doi:10.1177/0146167292182009

27.

Luo

Hawkley

L. C.

Waite

L. J.

Cacioppo

J. T.

(2012). Loneliness, health, and mortality in old age: A national longitudinal study. Social Science & Medicine, 74, 907–914. doi:10.1016/j.socscimed.2011.11.028

28.

Matthews

Danese

Wertz

Odgers

C. L.

Ambler

Moffitt

T. E.

Arseneault

(2016). Social isolation, loneliness and depression in young adulthood: A behavioural genetic analysis. Social Psychiatry and Psychiatric Epidemiology, 51, 339–348. doi:10.1007/s00127-016-1178-7

29.

McKenzie

K. S.

Hoyle

R. H.

(2008). The self-absorption scale: Reliability and validity in non-clinical samples. Personality and Individual Differences, 45, 726–731. doi:10.1016/j.paid.2008.07.020

30.

Peplau

L. A.

Cutrona

C. E.

(1980). The revised UCLA Loneliness Scale: Concurrent and discriminant validity evidence. Journal of Personality and Social Psychology, 39, 472–480.

31.

Perlman

Peplau

L. A.

(1981). Toward a social psychology of loneliness. In Duck

Gilmour

(Eds.), Personal relationships 3: Personal relationships in disorder (pp. 31–56). London, England: Academic Press.

32.

Perlman

Peplau

L. A.

(1984). Loneliness research: A survey of empirical findings. In Peplau

L. A.

Goldston

S. E.

(Eds.), Preventing the harmful consequences of severe and persistent loneliness (pp. 13–46). Rockville, MD: National Institute of Mental Health.

33.

Reis

H. T.

Sheldon

K. M.

Gable

S. L.

Roscoe

Ryan

R. M.

(2000). Daily well-being: The role of autonomy, competence, and relatedness. Personality and Social Psychology Bulletin, 26, 419–435. doi:10.1177/0146167200266002

34.

Roberts

S. G.

Dunbar

R. I.

Pollet

T. V.

Kuppens

(2009). Exploring variation in active network size: Constraints and ego characteristics. Social Networks, 31, 138–146. doi:10.1016/j.socnet.2008.12.002

35.

Sandstrom

G. M.

Dunn

E. W.

(2014). Social interactions and well-being: The surprising power of weak ties. Personality and Social Psychology Bulletin, 40, 910–922. doi:10.1177/0146167214529799

36.

Solano

C. H.

Batten

P. G.

Parish

E. A.

(1982). Loneliness and patterns of self-disclosure. Journal of Personality and Social Psychology, 43, 524–531. doi:10.1037/0022-3514.43.3.524

37.

Sprecher

Treger

Wondra

J. D.

Hilaire

Wallpe

(2013). Taking turns: Reciprocal self-disclosure promotes liking in initial interactions. Journal of Experimental Social Psychology, 49, 860–866. doi:10.1016/j.jesp.2013.03.017

38.

Steptoe

Shankar

Demakakos

Wardle

(2013). Social isolation, loneliness, and all-cause mortality in older men and women. Proceedings of the National Academy of Sciences, 110, 5797–5801. doi:10.1073/pnas.1219686110

39.

Stokes

J. P.

(1985). The relation of social network and individual difference variables to loneliness. Journal of Personality and Social Psychology, 48, 981–990. doi:10.1037/0022-3514.48.4.981

40.

Stokes

J. P.

(1987). The relation of loneliness and self-disclosure. In Derlega

V. J.

Berg

J. H.

(Eds.), Self-disclosure (pp. 175–201). Boston, MA: Springer.

41.

Sutcliffe

Dunbar

Binder

Arrow

(2012). Relationships and the social brain: Integrating psychological and evolutionary perspectives. British Journal of Psychology, 103, 149–168. doi:10.1111/j.2044-8295.2011.02061.x

42.

VanderWeele

T. J.

Hawkley

L. C.

Thisted

R. A.

Cacioppo

J. T.

(2011). A marginal structural model analysis for loneliness: Implications for intervention trials and clinical practice. Journal of Consulting and Clinical Psychology, 79, 225–235. doi:10.1037/a0022610

43.

Wei

Russell

D. W.

Zakalik

R. A.

(2005). Adult attachment, social self-efficacy, self-disclosure, loneliness, and subsequent depression for freshman college students: A longitudinal study. Journal of Counseling Psychology, 52, 602–614. doi:10.1037/0022-0167.52.4.602

44.

Weiss

R. S.

(1973). Loneliness: The experience of emotional and social isolation. Cambridge, MA: MIT Press.

45.

Wheeler

Nezlek

(1977). Sex differences in social participation. Journal of Personality and Social Psychology, 35, 742–754. doi:10.1037/0022-3514.35.10.742

46.

Wheeler

Reis

Nezlek

J. B.

(1983). Loneliness, social interaction, and sex roles. Journal of Personality and Social Psychology, 45, 943–953. doi:10.1037/0022-3514.45.4.943

47.

Wheeler

Reis

H. T.

(1991). Self-Recording of Everyday Life Events: Origins, Types, and Uses. Journal of Personality, 59, 339–354. doi:10.1111/j.1467-6494.1991.tb00252.x

48.

Wilson

Moulton

(2010). Loneliness among older adults: A national survey of adults 45+. Washington, DC: AARP Inc.

49.

Statistics New Zealand. (2013). Loneliness in New Zealand: Findings from the 2010 NZ General Social Survey. Retrieved from www.stats.govt.nz/∼/media/Statistics/browse-categories/people-and-communities/older-people/loneliness-in-nz/loneliness-in-nz-2010.pdf

	Outcome variables
	RIR_Strong tie interactions	RIR_Weak tie interactions	clickers_Strong tie interactions^a	clickers_Weak tie interactions^a
Loneliness	−.222*	.010	−.400***	−.106
Social network size	.255**	.117	.104	.287**
Loneliness × Social network size	−.215*	.008	−.223*	.058
R ²	.213	.013	.240	.101
F	10.307***	.490	11.464***	4.085**

	Outcome variables
	Positive emotional experience	Reciprocal self-disclosure	Self-absorption
Loneliness	−.446***	−.404***	.299**
Social network size	−.033	−.083	−.130
Loneliness × Social network size	−.037	−.060	−.184*
R ²	.191	.151	.153
F	8.954***	6.758***	6.845***