Beyond Social Deficits: Personal Agency and Social Connection Shape Loneliness Over Time

Beyond Social Deficits: Personal Agency and Social Connection Shape Loneliness Over Time | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Beyond Social Deficits: Personal Agency and Social Connection Shape Loneliness Over Time Oscar Ybarra This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6099787/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 29 Oct, 2025 Read the published version in Communications Psychology → Version 1 posted You are reading this latest preprint version Abstract This longitudinal study examined how social connections and personal agency influenced loneliness over time. Utilizing four waves of data from the Irish Longitudinal Study on Ageing, analyses consistently revealed four distinct prototypes within each data wave that reflect combinations of social connection and personal agency. Subsequent analyses showed that one prototype (low agency, low connection) experienced the highest levels of loneliness, while another (high agency, high connection) reported the lowest. The remaining two prototypes exhibited equivalent, intermediate loneliness despite the stark differences in social connection levels between them. Tracking transitions between prototypes across waves revealed that shifts toward less favorable prototypes predicted increased loneliness, whereas transitions toward more favorable prototypes predicted reduced loneliness. These findings significantly enhance our understanding of loneliness, an experience generally thought to be driven by relational deficits, highlighting the crucial role of personal agency. Biological sciences/Psychology Biological sciences/Psychology/Human behaviour Loneliness Personal Agency Social Connection Longitudinal Data Cluster Analysis Prototypes Figures Figure 1 Summary Loneliness is a global public health crisis that affects individuals across diverse populations 1, 2, 3 . Although research has shown that loneliness predicts various health issues, including depression, cognitive decline, and weakened immunity 4, 5 , our understanding of its root causes remains limited, as theoretical frameworks have mainly concentrated on deficits in social connection as the primary driver. This study demonstrates that loneliness arises from distinct combinations of levels of social connection and personal agency, identifying four stable prototypes: “separated” (high agency, low connection), “neglected” (low in both), “muted” (low agency, higher connection), and “empowered” (high in both). Across four waves of data spanning nine years, individuals in the “neglected” prototype consistently reported the highest levels of loneliness, while those in the “empowered” prototype experienced the least. Importantly, transitions between prototypes predicted subsequent changes in loneliness, with shifts toward the “neglected” prototype increasing loneliness and shifts toward the “empowered” prototype decreasing it. The findings illustrate that loneliness emerges from predictable configurations of both social and personal resources. These results suggest that interventions for loneliness should focus not only on enhancing social connections but also on fostering personal agency. Introduction In an article in the New York Times 1 , U.S. Surgeon General Vivek Murthy highlighted the critical issue of loneliness in America, calling it an epidemic. The World Health Organization (WHO) echoed this message 2 , stating that loneliness is a pressing global public health crisis that transcends borders and social groups. In response, the WHO launched a dedicated commission on social connection to raise awareness about the harmful effects of loneliness 3 . Loneliness merits recognition as it predicts psychological and physical health issues, including symptoms of depression, social anxiety, suicidal thoughts, and cognitive decline. On a physical level, loneliness can lead to problems such as high blood pressure, increased activity of the hypothalamic-pituitary-adrenal (HPA) axis, weakened immunity, and obesity 4, 5 . The urgency of addressing loneliness has led researchers to inquire about its fundamental nature. As Nature Portfolio asks, "What makes people lonely, how do the experiences of lonely people differ from those of less lonely people, and what predicts whether loneliness is detrimental to health?" 6 While loneliness is in large part about feeling disconnected or isolated from others 4 , it is believed to manifest when individuals perceive their level of social connection does not meet their needs 7 . This subjective nature means that people can experience loneliness regardless of their actual number of social connections 8, 9 . Over the past two decades, our understanding of loneliness has evolved significantly. It is an experience that can affect anyone at various life stages 10 , although it can vary by culture, age, and sex 11 . Additionally, loneliness is influenced by current social preferences and past experiences, which shape how one interprets others' warmth and care 5 . Furthermore, loneliness differs in terms of emotional intimacy; some individuals may feel lonely because of a lack of close relations, while others may feel disconnected from their broader community 12 . Recent research has also indicated that whether a person feels lonely depends on their beliefs about what being alone means 13 , and that the emotion regulation strategies used by lonely individuals often include several maladaptive approaches 14 . Despite these advancements, most theoretical frameworks and interventions for loneliness focus primarily on the relational dimension—addressing social connection needs and social skills 8, 9, 14 . The present research proposes that personal agency—encompassing self-direction, choice, and personal control—deserves equal attention in understanding loneliness. The dual needs for supportive social connections and the ability to direct one’s life reflect a fundamental aspect of human relationships: the tension between "we" (collective) and "me" (individual) needs 15, 16 . Meaningful relationships require balancing personal needs with those of others. When relationships fail to promote understanding and respect for individual needs, or when people engage in interactions with a limited sense of agency, the risk of loneliness can increase as well 17 . Research supports this perspective. For example, research establishes agency (autonomy) as a fundamental psychological need 18 . Studies on self-efficacy demonstrate that people's beliefs about their control over life outcomes directly affect their social relationships 19 . When individuals lack agency, they may suppress their desires to maintain relationships, miss opportunities to develop more supportive connections, or fail to develop competencies that make them valuable social partners 17 . Building on these insights, the present research examines how supportive social connections and personal agency combine to create distinct prototypes that predict different patterns of loneliness. Preliminary research has validated the existence of four prototypes and their relationship to loneliness 17 . However, that work was limited by its cross-sectional design. The present research extends this framework by using four waves of longitudinal data, allowing us to replicate the prototype patterns and examine both the stability of these patterns and how transitions between prototypes affect loneliness over time. Theoretical Framework The squad framework identifies four distinct patterns in which social connection and personal agency influence psychological outcomes and how individuals experience their social world (see Figure 1) 17 . Each prototype represents a unique configuration of the communion or relational dimension and the agency dimension. Q1 (Separated) describes individuals who prioritize independence and actively protect their agency and autonomy. This pattern can emerge through relationships that reinforce self-sufficiency norms, personal goals taking precedence over relationships, or negative relational experiences that lead to protecting oneself. While these individuals maintain their sense of agency, their limited interactions likely result in declining relationship quality 20, 21, 22 . Q2 (Neglected) represents individuals struggling with both social connection and personal agency. Often feeling ignored or uncertain about others' support, these individuals may suppress their desires in order to accommodate others, or they avoid taking opportunities to develop new relationships out of fear of leaving their current social prospects. Their reduced agency may also stem from underdeveloped competencies, which can make them less appealing as social partners 23, 24 . Q3 (Muted) characterizes individuals with moderately supportive social relationships but low agency. While some of these individuals may prioritize relational goals over personal ones, external factors often constrain their agency. Parents caring for a child, or a husband attending to their wife with dementia, for instance, will have limited time for personal pursuits 25 . Different social connections also can exert significant normative influences on their individual choices 26 . Q4 (Empowered) represents the optimal integration of care and support from social relations and personal agency. These individuals maintain warm, supportive relations that actively encourage their agency and personal pursuits 27, 28 . Their relationships provide a secure base for exploration while remaining strong through active engagement and nurturing. Preliminary research has validated these prototypes and their ability to predict loneliness 17 . Using clustering methods, the findings showed that Q2 or Prototype 2 individuals reported the highest loneliness scores, followed by those in Q1 (Prototype 1) and Q3 (Prototype 3), while Q4 (Prototype 4) individuals experienced the least loneliness. These findings make three important points: First, they are consistent with research showing that relational deficits predict loneliness. Second, they indicate different forms loneliness can take, as evidenced by similar loneliness scores for Prototype 1 (separated) and Prototype 3 (muted) individuals, despite their markedly different levels of social support. Third, they illustrate the role of agency, particularly evident in Prototype 2 (neglected) individuals who exhibit the greatest deficits in both relationships and personal agency. Despite these insights, the Prototypes need to be replicated along with their effects on loneliness. Furthermore, cross-sectional data cannot reveal how membership in a particular Prototype and loneliness change over time. This research aims to advance our understanding of who becomes lonely and why by utilizing four waves of data to track the stability of Prototype configurations and examine how transitions between Prototypes affect loneliness. Methods Participants This study utilized data from TILDA, the Irish Longitudinal Study on Ageing, which consists of five waves of data collection spanning nine years starting in 2009 29 . Few samples provide the richness of data on people’s social relationships, and this study does so with data collected across time (see supplemental materials for more information on samples). The previously reviewed research on the squad framework relied solely on data from Wave 1, 17 . The current study focuses on Waves 2 through 5. Measures Loneliness To assess loneliness , the TILDA study used a 5-item measure adapted from the UCLA loneliness scale 30 . Sample items included: “How often do you feel lonely?” and “How often do you feel isolated from others?” and these were answered on 3-point scales (1 = often, 2 = some of the time, and 3 = hardly ever or never). One item was positively worded, so it was reverse-scored before averaging it with the other items. Higher scores reflect greater feelings of loneliness . Table 1 presents the descriptive statistics for all the variables in the study across the four data waves. The number of participants ( n ) listed for each variable represents valid listwise observations after selecting participants who are married and reported having children. Communion The communion variable reflects the relational dimension, that is, the support and care individuals report receiving from their relationships, along with any strain present in those relationships 31 . The composite communion variable is derived from the averaged responses to twenty-eight questions. The participants answered the same seven questions separately about their spouse, children, other family members, and friends. Examples of these questions include: “How much can you rely on him/her if you have a serious problem?” and “How much does he/she make too many demands on you?” Responses were made on 4-point scales that ranged from 1=a lot to 4=not at all. Relevant items were reverse-scored, so that higher scores mean greater support and care from social relationships. Agency Agency assessed the level of control and autonomy individuals feel in their lives. Aside from Wave 2, the measure consisted of seven items (Wave 2 had nine items). Additionally, the Wave 2 data set only provided a composite variable, which means it was not possible to calculate Cronbach’s alpha for that wave. Examples of the seven common items across the four data waves included “I feel that what happens to me is out of my control,” “I feel free to plan for the future,” and “My health stops me from doing the things I want to do.” Responses to the items were given on 4-point scales (1=often to 4=never) and were reverse-scored so that higher averages indicate greater agency . Covariates When examining how communion and agency impact loneliness, it is important to account for factors related to these variables. Research shows that women typically receive more support from diverse social connections 32, 33 , and age influences the structure of social networks and relationship quality 34, 35 , so the analyses controlled for participant gender and age . Socioeconomic status is linked to the cultivation of diverse opportunities in life 36, 37, 38 . Therefore, I also controlled for income (gross total quintiles) and educational attainment. Education was assessed using a 7-point scale in Wave 2 and a 3-point scale in subsequent waves, with higher scores indicating a higher education level. In addition to these variables, I controlled for health, given that negative health conditions are associated with declines in social engagement 35 and greater dependency on existing relationships. Health was assessed with a self-report measure that used a 5-point scale ranging from 1 “excellent” to 5 “poor.” The response was reversed-scored so that higher ratings indicate better health. Similarly to health, the ability to fulfill activities of independent living should influence social engagement 39 . Some of the activities included shopping for groceries, preparing meals, and managing finances. Hence, I controlled for instrumental activities of daily living ( IADLs ), which were assessed on a scale running from 0 = none to 3 = 4+ IADLs. Depression can coincide with loneliness 40 , so it is important to account for it. Depression was evaluated with two items from the C-ESD 41 , which asked participants about their experiences in the week preceding the survey. The two items included: “I felt depressed” and “I felt that everything I did was an effort,” scored on 4-point scales ranging from 1 “rarely or none of the time (less than 1 day)” to 4 “all of the time (5-7 days).” The two items were averaged, with higher scores indicating greater depression. Finally, I accounted for the number of close relatives and friends, as this predicts loneliness 42 . This question was answered on a scale running from 1 – 6 (1 = “0-4 relations,” 6=”25 + relations”). The responses across the waves indicate an average of approximately 12 close relations. Table 1: Descriptives Statistics for Wave 2 – Wave 5 Variables Wave 2 (overall n = 7206) Wave 3 (overall n = 6397) Wave 4 (overall n = 5713) Wave 5 (overall n = 4978) Loneliness 1.32 (sd=.39) Cronbach’s = .78 (n=4117) 1.28 (sd=.38) Cronbach’s = .79 (n=3508) 1.27 (sd=.37) Cronbach’s = .79 (n=3131) 1.26 (sd=.37) Cronbach’s = .80 (n=2711) Communion 3.37 (sd=.36) Cronbach’s = .87 (n=3458) 3.38 (sd=.36) Cronbach’s =.87 (n=2841) 3.39 (sd=.36) Cronbach’s = .87 (n=2566) 3.40 (sd=.36) Cronbach’s = .87 (n=2184) Agency 2.11 (sd=.49) Cronbach's = n/a (n=3858) 2.04 (sd=.54) Cronbach’s = .70 (n=3419) 2.10 (sd=.54) Cronbach’s = .71 (n=2935) 2.09 (sd=.52) Cronbach’s = .70 (n=2524) Age 63.07 (sd=8.46) (n=4710) 65.04 (sd=8.21) (n=4055) 66.70 (sd=8.01) (n=3577) 68.16 (sd=7.64) (n=3119) Sex Male = 2226 Female = 2484 (n=4710) Male = 1919 Female = 2137 (n=4056) Male = 1695 Female = 1882 (n=3577) Male = 1484 Female = 1635 (n=3119) Education 3.84 (sd=1.54) (n=4693) 2.14 (sd=.75) (n=4055) 2.18 (sd=.74) (n=3577) 2.21 (sd=.74) (n=3119) Finances 3.26 (sd=1.37) (n=3627) 3.28 (sd=1.34) (n=2929) 3.29 (sd=1.34) (n=2569) 3.25 (sd=1.35) (n=2136) Health status 3.49 (sd=.99) (n=4710) 3.48 (sd=.99) (n=4055) 3.50 (sd=.94) (n=3576) 3.49 (sd=.95) (n=3119) IADL’s .05 (sd=.30) (n=4710) .05 (sd=.31) (n=4054) .05 (sd=.30) (n=3577) .05 (sd=.30) (n=3118) Depression 1.24 (sd=.33) r = .53 (n=4707) 1.25 (sd=.52) r = .53 (n=4052) 1.24 (sd=.50) r = .48 (n=3573) 1.24 (sd=.49) r = .42 (n=3113) Number of relations 2.63 (sd=1.18) (n=4708) 2.56 (sd=1.07) (n=4054) 2.37 (sd=.92) (n=3577) 2.42 (sd=.94) (n=3116) Note: The statistics are derived from individuals who were married at each time point and reported having children at Wave 1, the only time this variable was assessed. Results Analytic approach Attrition can affect longitudinal data, as those who remain across data waves may differ from individuals no longer in the study. Additionally, people undergo changes in their social relationships 32, 33, 43 and their ability to exert control over their lives 44, 45 . Therefore, the relative standing of individuals within a sample can shift. Consequently, it is crucial to re-standardize the criterion variables for each wave of data, which also allows us to capture the dynamics of change over time. For each data wave, our analytical approach included the following steps: 1) standardize the criterion variables (communion and agency) and conduct a cluster analysis. 2) subject the loneliness scores to ANCOVA to test the hypothesized effects of Prototype on levels of loneliness. 3) conduct analyses to determine how prototype membership changed over time and how those changes influenced loneliness scores. Main cluster analysis Drawing on previous research that identified four distinct patterns in how individuals experience their social relationships and personal agency 17 , we employed k-means clustering to classify participants into four mutually exclusive clusters. This theoretical constraint, along with the requirement that individuals can belong to only one cluster, directed our analytical approach. K-means clustering was chosen based on comparative analyses that demonstrate its superior performance compared to other clustering methods 46 . To maximize data retention while ensuring analytical rigor, we applied pairwise deletion during the clustering analysis. To examine the consistency of the clustering solutions, analyses using listwise deletion were also conducted, which yielded similar clustering results (see supplemental materials). Subsequently, the elbow method suggested a four-cluster solution and possibly a two-cluster solution. However, the four-cluster solution demonstrated superior differentiation of the data. In contrast, the two-cluster solution merely dichotomized observations into favorable (positive scores on both dimensions) and unfavorable clusters, reducing explanatory power (see supplementary materials for the two-cluster solutions). We also generated ASW metrics (average silhouette width) for the four-cluster solutions. The ASW metric ranges from −1 to 1, where −1 indicates the worst clustering solution, and values close to 0 suggest overlap among clusters. Positive values signify a good match to the assigned cluster. The ASW scores were as follows: W2 = .36; W3 = .36; W4 = .35; and W5 = 0.35. Therefore, four clusters (prototypes) were kept based on theory and analyses of relevant metrics. The results shown in Panel 1 and Table 2 indicate the four clusters align with the four predicted prototypes. - Prototype 1 (Separated) includes individuals who have high agency but low communion. - Prototype 2 (Neglected) consists of individuals who score low on both dimensions. - Prototype 3 (Muted) features individuals with lower agency but higher communion. - Prototype 4 (Empowered) is characterized by individuals who score high on both dimensions. The larger number of individuals in Prototype 4 (empowered) likely reflects the positive skew observed in quality measures of social relationships 33 . Nevertheless, each prototype had a sufficient number of individuals. Table 2: The data represent four clusters across Waves 2-5, showing sample sizes (n) and standardized scores (z) for the Communion and Agency dimensions. Cluster Wave n Communion (z) Agency (z) Prototype 1 (Separated) 2 1085 -0.64 0.33 3 860 -0.58 0.54 4 809 -0.63 0.29 5 548 -0.91 0.26 Prototype 2 (Neglected) 2 582 -1.51 -1.47 3 475 -1.63 -1.43 4 470 -1.48 -1.41 5 461 -1.19 -1.44 Prototype 3 (Muted) 2 996 0.36 -0.69 3 1019 0.11 -0.68 4 749 0.34 -0.69 5 795 0.44 -0.41 Prototype 4 (Empowered) 2 1353 0.81 0.93 3 1082 0.95 0.33 4 1030 0.86 0.92 5 840 0.75 1.04 Loneliness results for the different prototypes Among the prototypes, ANCOVAs revealed consistent patterns of loneliness across all waves. In Wave 2 (F(3, 3057) = 181.53, p < .001), Prototype 2 (neglected) showed the highest loneliness (M = 1.64, SE = .02), followed by Prototype 1 (separated) (M = 1.32, SE = .01) and Prototype 3 (muted; M = 1.32, SE = .01), while the empowered Prototype showed the lowest loneliness scores (M = 1.16, SE = .01). Prototype 1 and Prototype 3 did not differ (F < 1), but all other comparisons were significant (ps < .001; Bonferroni adjusted significance value per wave for nine comparisons = .005). This pattern replicated across subsequent waves, with analyses controlling for prior loneliness. Wave 3 showed a significant prototype effect (F(3, 2315)=65.80, p <.001), replicating Wave 2's pattern. Prototype 2 (neglected) had the highest loneliness (M = 1.47; SE = .02) versus Prototype 1 (separated, M = 1.26; SE = .01), Prototype 3 (muted, M = 1.27; SE = .01), and Prototype 4 (empowered, M = 1.20; SE = .01). Prototype 4 differed from all others (all ps < 0.001). Prototypes 1 and 3 did not differ (F<1). Wave 4 showed similar patterns (F(3, 2063)=67.14, p <.001). Prototype 2 (neglected) had the highest loneliness scores (M = 1.47, SE = .02) versus all others (Prototype 1: M = 1.27, SE = .01; Prototype 3: M = 1.29, SE = .01; Prototype 4: M = 1.17, SE = .01; all ps < 0.001). Prototype 4 had the lowest loneliness scores and differed from all others (ps < 0.001), while Prototypes 1 and 3 did not differ. Wave 5 maintained these findings (F(3, 1685)=50.58, p <.001). Prototype 2 (M = 1.44, SE = .02) differed from all others (Prototype 1: M = 1.25, SE = .01; Prototype 3: M = 1.22, SE = .01; Prototype 4: M = 1.18, SE = .01; all ps < 0.001). Prototype 4 differed from all others (ps < 0.001), with Prototypes 1 and 3 showing no difference. Change in Prototypes Across Time The analyses revealed consistent prototypes across data waves that predict loneliness, even though the clusters are based on re-standardized criterion variables. While the same prototypes appeared consistently, this does not mean the same individuals ended up in the same Prototype at the next time interval. Subsequent analyses assessed change over time and, more importantly, whether these changes resulted in differences in loneliness. Panel 2 presents change percentages among prototypes over time (see supplemental materials for Sankey diagrams). The prototype an individual belongs to in the previous wave has a strong chance of being the same one they will be in subsequently. Another observation is that individuals in Prototype 4 (empowered) are highly unlikely to move to Prototype 2 (neglected), and the reverse is also true. Prototype Transitions and Loneliness I examined how cluster transitions influenced levels of loneliness by categorizing these transitions into favorable and unfavorable groups. Based on earlier findings indicating that Prototype 2 (neglected) is associated with the highest levels of loneliness, Prototype 4 (empowered) with the lowest, and Prototypes 1 and 3 are in between, we classified transitions to or stays in Prototype 2 as unfavorable, while transitions to or stays in Prototype 4 were deemed favorable. Transitions from Prototype 2 to Prototype 1 or 3 were categorized as favorable, whereas transitions from Prototype 4 to Prototype 1 or 3 were categorized as unfavorable. Finally, we did not analyze transitions that simply swapped memberships between Prototype 1 (separated) and Prototype 3 (muted), given the findings that they did not differ in loneliness. We conducted analyses using both this binning approach and a simplified version that focused solely on transitions or stays that occurred for Prototype 2 and Prototype 4. The Bonferroni-adjusted significance value per wave pair for four comparisons is .012. For change in Prototypes from Wave 2 to Wave 3, ANCOVA results (controlling for Wave 3 covariates and previous loneliness) indicated a significant effect of bin (F(1, 1452)=77.60, p <.001). Individuals in unfavorable Prototypes (M = 1.35, SE = .01) had higher loneliness scores than those in favorable Prototypes (M = 1.22, SE = .01). With the simpler assumptions (analyzing only Prototype 2 and 4 transitions/stays), results remained similar (F(1, 1024)=124.97, p <.001). Prototype 2 individuals (neglected, M = 1.47, SE = .02) had higher loneliness scores than Prototype 4 individuals (empowered, M = 1.20, SE = .01). For Waves 3-4, the effect persisted under both sets of assumptions. The primary analysis showed a significant bin effect (F(1, 1293)=116.33, p <.001), with unfavorable Prototype individuals (M = 1.37, SE = .01) reporting higher loneliness than those in more favorable Prototypes (M = 1.20, SE = .01). The restricted analysis revealed similar patterns (F(1, 984)=160.93, p <.001), with Prototype 2 individuals (M = 1.46, SE = .02) showing higher loneliness than Prototype 4 (M = 1.17, SE = .01). For Waves 4-5, the pattern was similar. The primary analysis showed a significant bin effect (F(1, 1127) = 60.02, p < .001), with unfavorable Prototype individuals (M = 1.32, SE = .01) reporting higher loneliness than favorable Prototype individuals (M = 1.19, SE = .01). The restricted analysis confirmed these findings (F(1, 814) = 79.94, p < .001), with Prototype 2 individuals (M = 1.41, SE = .02) showing higher loneliness than Prototype 4 (M = 1.19, SE = .01). Analyses without covariates (available in supplemental materials) showed similar, though stronger, patterns. Discussion For each data wave, the results indicated the same four Prototypes based on individuals' reports of the support from their social relations and their personal agency. Further, loneliness levels varied significantly across these Prototypes, with the neglected Prototype reporting the highest loneliness levels and the empowered Prototype the lowest. The muted and separated Prototypes showed intermediate loneliness levels, similar to each other but distinct from both Prototype 2 (neglected) and Prototype 4 (empowered). This pattern remained stable across Waves 3, 4, and 5, despite re-standardizing the criterion variables. While Prototype membership showed stability between waves, considerable movement occurred among them. Analyses of these transitions across three consecutive wave pairs revealed that individuals who moved to or remained in the unfavorable Prototype experienced greater loneliness compared to those who moved to or remained in the more favorable Prototype. The results challenge the dominant view that loneliness primarily arises from social or relational deficits. The findings indicate that personal agency plays an equally important role. The interplay between social connection and agency, as demonstrated in our prototypes, illustrates how loneliness arises from complex patterns in how individuals configure their social relations and their positions within them. The emergence of these patterns across data waves, even as individuals transition to different prototypes, suggests that these configurations reflect meaningful social-psychological states for individuals. Furthermore, the sequencing between Prototype transitions and the subsequent experience of loneliness offers compelling evidence that relationship-agency configurations affect loneliness levels. Although longitudinal designs cannot definitively determine causation due to potential unmeasured confounding factors, the observed shifts in Prototype membership occurring before changes in loneliness deepen our theoretical understanding beyond cross-sectional correlations. If these findings are confirmed by future research, they imply that interventions could be tailored to assist individuals not only in improving their social relationships but also in recognizing the need for personal development so they can be more independent, socially attractive, and assert themselves within their relationships. Limitations and Future Directions The current research has limitations. The measurement intervals between waves may have overlooked shorter-term fluctuations in social support, strain, and changes in personal agency. Additionally, the findings could be influenced by cultural context due to variations in norms and traditions. Future research could benefit from designs that use more frequent measurements, and that also examine factors thought to promote or impede positive relationships, along with the development of personal agency. Additionally, cross-cultural studies might reveal how different social norms and values influence the relationship between agency, social connection, and loneliness. However, considering global trends toward achieving greater agency in life, along with the positive wellbeing effects that accompany such changes 47 , we may also discover more consistency across different populations 16 . Finally, it would be useful to examine individual differences, such as in attachment styles, to determine whether different ways of relating to others predispose certain individuals to end up in one prototype over another. Conclusion This research examined the nature of people’s social relationships and their sense of personal agency as key factors predicting loneliness. The findings suggest that meaningful relationships not only serve as sources of social support but also allow individuals to express their agency and to be themselves. These results should contribute to the discussion on loneliness and its causes, helping to shift the perspective away from a narrow view of loneliness as primarily arising from relational deficits. Declarations Competing Interest Declaration I declare no competing financial or non-financial interests as defined by Nature Portfolio. Author Contribution I am solely responsible for conceptualizing the research, processing and analyzing the data, and writing the manuscript. However, Tanfu Shi assisted with generating the following figures (Panels 1 and 2 in the manuscript and Panel 1 in the supplemental materials). Authorship I have read the Nature Portfolio Authorship Policy and confirm that the manuscript complies with it. I am the sole author and not an LLM. Data Availability Access to the data can be obtained by applying to the owners and curators of the Irish Longitudinal Study on Ageing (TILDA) at Trinity College Dublin (https://tilda.tcd.ie). Code Availability The code for processing and analyzing the data and for generating the figures is available from the author. Human Research Participants This study is based on secondary analyses of existing data collected by the TILDA researchers. As they note, “Ethical approval for each wave was obtained from the Faculty of Health Sciences Research Ethics Committee at Trinity College Dublin.” (tilda.tcd.ie/publications/reports/pdf/Report_CohortMaintenance.pdf). I received IRB approval from the University of Illinois to conduct secondary analyses of the data. References The New York Times. The loneliness epidemic in America [Opinion]. Retrieved from https://www.nytimes.com/2023/04/30/opinion/loneliness-epidemic-america.html (2023). World Health Organization. Social connection for mental health and well-being: Addressing isolation and loneliness (2023). World Health Organization. Commission on Social Connection. Retrieved January 30, 2025, from https://www.who.int/groups/commission-on-social-connection (n.d.). Hawkley, L. C., & Cacioppo, J. T. Loneliness matters: A theoretical and empirical review of consequences and mechanisms. Annals of Behavioral Medicine , 40, 218–227 (2010). Cacioppo, S., Grippo, A. J., London, S., Goossens, L., & Cacioppo, J. T. Loneliness: Clinical import and interventions. Perspectives on Psychological Science , 10, 238–249 (2015). Nature Portfolio. How to submit to Nature collections. Retrieved February 18, 2025, from https://www.nature.com/collections/eefieehehi/how-to-submit (n.d.). Peplau, L. A., & Perlman, D. Perspectives on loneliness. In L. A. Peplau & D. Perlman (Eds.), Loneliness: A sourcebook of current theory, research, and therapy (pp. 1–8). New York, NY: Wiley (1982). Perlman, D., & Peplau, L. A. Toward a social psychology of loneliness. Personal Relationships , 3, 31–56 (1981). Masi, C. M., Chen, H. Y., Hawkley, L. C., & Cacioppo, J. T. A meta-analysis of interventions to reduce loneliness. Personality and Social Psychology Review , 15, 219–266 (2011). Andersson, L. Loneliness research and interventions: A review of the literature. Aging & Mental Health , 4, 264–274 (1998). Barreto, M., Victor, C., Hammond, C., Eccles, A., Richins, M. T., & Qualter, P. Loneliness around the world: Age, gender, and cultural differences in loneliness. Personality and Individual Differences , 169, 110066 (2021). Hawkley, L. C., Browne, M. W., & Cacioppo, J. T. How can I connect with thee? Let me count the ways. Psychological Science , 16, 798–804 (2005). Rodriguez, M., Schertz, K. E., & Kross, E. How people think about being alone shapes their experience of loneliness. Nature Communications , 16, 1594 (2025). Tan, A. J., Mancini, V., Gross, J. J., Goldenberg, A., Badcock, J. C., Lim, M. H., ... & Preece, D. A. Loneliness versus distress: A comparison of emotion regulation profiles. Behaviour Change , 39, 180-190 (2022). Brown, D. E. Human universals and their implications. In N. Roughley (Ed.), Being humans: Anthropological universality and particularity in transdisciplinary perspectives (pp. 156–174). Berlin, Germany: Walter de Gruyter (2000). Ybarra, O., Chan, E., Park, H., Burnstein, E., Monin, B., & Stanik, C. Life's recurring challenges and the fundamental dimensions: An integration and its implications for cultural differences and similarities. European Journal of Social Psychology , 38, 1083–1092 (2008). Ybarra, O., & Chan, T. The S(quad) model, a pattern approach for understanding the individual and their social network relations: Application to loneliness. Frontiers in Social Psychology , 1, 1278671 (2023). Ryan, R. M., & Deci, E. L. Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist , 55, 68–78 (2000). Bandura, A. Self-efficacy. In V. S. Ramachaudran (Ed.), Encyclopedia of Human B ehavior (Vol. 4, pp. 71–81). San Diego, CA: Academic Press (1994). Burt, R. S. Decay functions. Social Networks , 22, 1–28 (2000). Oswald, D. L., Clark, E. M., & Kelly, C. M. Friendship maintenance: An analysis of individual and dyad behaviors. Journal of Social and Clinical Psychology , 23, 413–441 (2004). Ogolsky, B. G., & Bowers, J. R. A meta-analytic review of relationship maintenance and its correlates. Journal of Social and Personal Relationships , 30, 343–367 (2013). Berkowitz, L. Resistance to improper dependency relationships. Journal of Experimental Social Psychology , 5, 283–294 (1969). Leary, M. R., Jongman-Sereno, K. P., & Diebels, K. J. The pursuit of status: A self-presentational perspective on the quest for social value. In J. T. Cheng, J. L. Tracy, & C. Anderson (Eds.), The Psychology of Social Status (pp. 159–178). New York, NY: Springer (2014). Perry-Jenkins, M., & Gerstel, N. Work and family in the second decade of the 21st century. Journal of Marriage and Family , 82, 420–453 (2020). Cohen, S., & Lemay, E. P. Why would social networks be linked to affect and health practices? Health Psychology , 26, 410–417 (2007). Bowlby, J. Attachment . New York, NY: Basic Books (1969). Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. N. Patterns of attachment: A psychological study of the strange situation . New York, NY: Psychology Press (2015). The Irish Longitudinal Study on Ageing. TILDA: The Irish longitudinal study on ageing. Retrieved February 18, 2025, from https://tilda.tcd.ie (n.d.). Russell, D. W. UCLA Loneliness Scale (Version 3): Reliability, validity, and factor structure. Journal of Personality Assessment , 66, 20–40 (1996). Schuster, T. L., Kessler, R. C., & Aseltine, R. H., Jr. Supportive interactions, negative interactions, and depressed mood. American Journal of Community Psychology , 18, 423–438 (1990). Antonucci, T. C. Social relations: An examination of social networks, social support, and sense of control. In J. E. Birren & K. W. Schaie (Eds.), Handbook of the Psychology of Aging (pp. 427–453). San Diego, CA: Academic Press (2001). Fiori, K. L., Antonucci, T. C., & Akiyama, H. Profiles of social relations among older adults: A cross-cultural approach. Ageing & Society , 28, 203–231 (2008). Carstensen, L. L., Isaacowitz, D. M., & Charles, S. T. Taking time seriously: A theory of socioemotional selectivity. American Psychologist , 54, 165–181 (1999). Huxhold, O., Fiori, K. L., & Windsor, T. D. The dynamic interplay of social network characteristics, subjective well-being and health: The costs and benefits of socio-emotional selectivity. Psychology and Aging , 28, 3–16 (2013). Lachman, M. E., & Weaver, S. L. The sense of control as a moderator of social class differences in health and well-being. Journal of Personality and Social Psychology , 74, 763–773 (1998). Taylor, S. E., & Seeman, T. E. Psychosocial resources and the SES‐health relationship. Annals of the New York Academy of Sciences , 896, 210–225 (1999). Kraus, M. W., Piff, P. K., & Keltner, D. Social class, sense of control, and social explanation. Journal of Personality and Social Psychology , 97, 992–1004 (2009). Curl, A. L., Stowe, J. D., Cooney, T. M., & Proulx, C. M. Giving up the keys: How driving cessation affects engagement in later life. The Gerontologist , 54, 423–433 (2014). Cacioppo, J. T., Hughes, M. E., Waite, L. J., Hawkley, L. C., & Thisted, R. A. Loneliness as a specific risk factor for depressive symptoms: Cross-sectional and longitudinal analyses. Psychology and Aging , 21, 140–151 (2006). Radloff, L. S. The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement , 1, 385–401 (1977). Russell, D., Peplau, L. A., & Cutrona, C. E. The revised UCLA Loneliness Scale: Concurrent and discriminant validity evidence. Journal of Personality and Social Psychology , 39, 472–480 (1980). Carstensen, L. L. Selectivity theory: Social activity in life-span context. Annual Review of Gerontology and Geriatrics , 11, 195–217 (1991). Lachman, M. E., Neupert, S. D., & Agrigoroaei, S. The relevance of control beliefs for health and aging. In K. W. Schaie & S. L. Willis (Eds.), Handbook of the Psychology of Aging (pp. 175–190). San Diego, CA: Academic Press (2011). Heckhausen, J., Wrosch, C., & Schulz, R. A motivational theory of life-span development. Psychological Review , 117, 32–60 (2010). Preud'homme, G., Duarte, K., Dalleau, K., Lacomblez, C., Bresso, E., Smaïl-Tabbone, M., ... & Girerd, N. Head-to-head comparison of clustering methods for heterogeneous data: A simulation-driven benchmark. Scientific Reports , 11, 4202 (2021). Welzel, C., & Inglehart, R. Agency, values, and well-being: A human development model. Social Indicators Research , 97, 43–63 (2010). Panels Panels 1 and 2 are available in the Supplementary Files section Additional Declarations There is NO Competing Interest. Supplementary Files YbarraSupplementalMaterialsFinal.docx Supplementary Information Panels.docx Cite Share Download PDF Status: Published Journal Publication published 29 Oct, 2025 Read the published version in Communications Psychology → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6099787","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":421892002,"identity":"19189394-1acc-4377-b221-67bea1e3c3c2","order_by":0,"name":"Oscar Ybarra","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2UlEQVRIiWNgGAWjYFACHjApx8bO3HCAgUECxDEgSosxGzMjiVoSG4BaYEL4tfBL5B78XLnHLr2PmbHxcEGFhT0De/M2CXxaJGfkJUueeZac2wa05fCMMxKJDTzHyvBqMbiRYyDZcIAZooW3TSKBQSLHjJAW458NB+rT2aBa7Bnk3xDUYga05XACTAtjgwQPfi2SPe/SLBsOHDcEO4wH6Jc2nrRiC3xa+NlzD99sOFAtL9/efPgzT0WdPT/74Y038GlhEEhAE2DDqxxszQGCSkbBKBgFo2CkAwCToEL0xRM9IgAAAABJRU5ErkJggg==","orcid":"","institution":"University of Illinois Urbana-Champaign","correspondingAuthor":true,"prefix":"","firstName":"Oscar","middleName":"","lastName":"Ybarra","suffix":""}],"badges":[],"createdAt":"2025-02-24 20:55:06","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6099787/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6099787/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s44271-025-00329-z","type":"published","date":"2025-10-29T04:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":79240696,"identity":"dea95819-3865-49ce-938a-31ebf1e4ffec","added_by":"auto","created_at":"2025-03-26 05:43:58","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":170919,"visible":true,"origin":"","legend":"\u003cp\u003eSquad framework illustration.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6099787/v1/1afd9c3391fe72357cd3a108.png"},{"id":94733207,"identity":"60f66f4b-436f-4259-b66b-71a486472255","added_by":"auto","created_at":"2025-10-30 07:10:22","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":822680,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6099787/v1/f5610591-8e3e-4ac0-ad91-a96d6a200014.pdf"},{"id":79241858,"identity":"5b210b63-80ca-443c-b20f-455d23d7e167","added_by":"auto","created_at":"2025-03-26 06:07:58","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":8361185,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"YbarraSupplementalMaterialsFinal.docx","url":"https://assets-eu.researchsquare.com/files/rs-6099787/v1/b9eb842b4bf3a1d6c4c7c7fe.docx"},{"id":79241855,"identity":"5c370426-c518-43a4-8f8a-e6c8ff65990f","added_by":"auto","created_at":"2025-03-26 06:07:58","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":212799,"visible":true,"origin":"","legend":"","description":"","filename":"Panels.docx","url":"https://assets-eu.researchsquare.com/files/rs-6099787/v1/29025c441a2d4a47740bd520.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Beyond Social Deficits:\nPersonal Agency and Social Connection Shape Loneliness Over Time","fulltext":[{"header":"Summary","content":"\u003cp\u003eLoneliness is a global public health crisis that affects individuals across diverse populations \u003cu\u003e\u003csup\u003e1, 2, 3\u003c/sup\u003e\u003c/u\u003e. Although research has shown that loneliness predicts various health issues, including depression, cognitive decline, and weakened immunity \u003cu\u003e\u003csup\u003e4, 5\u003c/sup\u003e\u003c/u\u003e, our understanding of its root causes remains limited, as theoretical frameworks have mainly concentrated on deficits in social connection as the primary driver. This study demonstrates that loneliness arises from distinct combinations of levels of social connection and personal agency, identifying four stable prototypes: “separated” (high agency, low connection), “neglected” (low in both), “muted” (low agency, higher connection), and “empowered” (high in both). Across four waves of data spanning nine years, individuals in the “neglected” prototype consistently reported the highest levels of loneliness, while those in the “empowered” prototype experienced the least. Importantly, transitions between prototypes predicted subsequent changes in loneliness, with shifts toward the “neglected” prototype increasing loneliness and shifts toward the “empowered” prototype decreasing it. The findings illustrate that loneliness emerges from predictable configurations of both social and personal resources. These results suggest that interventions for loneliness should focus not only on enhancing social connections but also on fostering personal agency.\u0026nbsp;\u003c/p\u003e"},{"header":"Introduction","content":"\u003cp\u003eIn an article in the New York Times \u003csup\u003e1\u003c/sup\u003e, U.S. Surgeon General Vivek Murthy highlighted the critical issue of loneliness in America, calling it an epidemic. The World Health Organization (WHO) echoed this message \u003csup\u003e2\u003c/sup\u003e, stating that loneliness is a pressing global public health crisis that transcends borders and social groups. In response, the WHO launched a dedicated commission on social connection to raise awareness about the harmful effects of loneliness \u003csup\u003e3\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eLoneliness merits recognition as it predicts psychological and physical health issues, including symptoms of depression, social anxiety, suicidal thoughts, and cognitive decline. On a physical level, loneliness can lead to problems such as high blood pressure, increased activity of the hypothalamic-pituitary-adrenal (HPA) axis, weakened immunity, and obesity \u003csup\u003e4, 5\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThe urgency of addressing loneliness has led researchers to inquire about its fundamental nature. As Nature Portfolio asks, \u0026quot;What makes people lonely, how do the experiences of lonely people differ from those of less lonely people, and what predicts whether loneliness is detrimental to health?\u0026quot; \u003csup\u003e6\u003c/sup\u003e While loneliness is in large part about feeling disconnected or isolated from others \u003csup\u003e4\u003c/sup\u003e, it is believed to manifest when individuals perceive their level of social connection does not meet their needs \u003csup\u003e7\u003c/sup\u003e. This subjective nature means that people can experience loneliness regardless of their actual number of social connections \u003csup\u003e8, 9\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eOver the past two decades, our understanding of loneliness has evolved significantly. It is an experience that can affect anyone at various life stages \u003csup\u003e10\u003c/sup\u003e, although it can vary by culture, age, and sex \u003csup\u003e11\u003c/sup\u003e. Additionally, loneliness is influenced by current social preferences and past experiences, which shape how one interprets others\u0026apos; warmth and care \u003csup\u003e5\u003c/sup\u003e. Furthermore, loneliness differs in terms of emotional intimacy; some individuals may feel lonely because of a lack of close relations, while others may feel disconnected from their broader community \u003csup\u003e12\u003c/sup\u003e. Recent research has also indicated that whether a person feels lonely depends on their beliefs about what being alone means \u003csup\u003e13\u003c/sup\u003e, and that the emotion regulation strategies used by lonely individuals often include several maladaptive approaches \u003csup\u003e14\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eDespite these advancements, most theoretical frameworks and interventions for loneliness focus primarily on the relational dimension\u0026mdash;addressing\u0026nbsp;social connection needs and social skills \u003csup\u003e8, 9, 14\u003c/sup\u003e. The present research proposes that personal agency\u0026mdash;encompassing self-direction, choice, and personal control\u0026mdash;deserves equal attention in understanding loneliness.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe dual needs for supportive social connections and the ability to direct one\u0026rsquo;s life reflect a fundamental aspect of human relationships: the tension between \u0026quot;we\u0026quot; (collective) and \u0026quot;me\u0026quot; (individual) needs \u003csup\u003e15,\u003c/sup\u003e \u003csup\u003e16\u003c/sup\u003e. Meaningful relationships require balancing personal needs with those of others. When relationships fail to promote understanding and respect for individual needs, or when people engage in interactions with a limited sense of agency, the risk of loneliness can increase as well \u003csup\u003e17\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eResearch supports this perspective. For example, research establishes agency (autonomy) as a fundamental psychological need \u003csup\u003e18\u003c/sup\u003e. Studies on self-efficacy demonstrate that people\u0026apos;s beliefs about their control over life outcomes directly affect their social relationships \u003csup\u003e19\u003c/sup\u003e.\u0026nbsp;When individuals lack agency, they may suppress their desires to maintain relationships, miss opportunities to develop more supportive connections, or fail to develop competencies that make them valuable social partners \u003csup\u003e17\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eBuilding on these insights, the present research examines how supportive social connections and personal agency combine to create distinct prototypes that predict different patterns of loneliness. Preliminary research has validated the existence of four prototypes and their relationship to loneliness \u003csup\u003e17\u003c/sup\u003e. However, that work was limited by its cross-sectional design. The present research extends this framework by using four waves of longitudinal data, allowing us to replicate the prototype patterns and examine both the stability of these patterns and how transitions between prototypes affect loneliness over time.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTheoretical Framework\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe squad framework identifies four distinct patterns in which social connection and personal agency influence psychological outcomes and how individuals experience their social world (see Figure 1) \u003csup\u003e17\u003c/sup\u003e. Each prototype represents a unique configuration of the communion or relational dimension and the agency dimension.\u003c/p\u003e\n\u003cp\u003eQ1 (Separated) describes individuals who prioritize independence and actively protect their agency and autonomy. This pattern can emerge through relationships that reinforce self-sufficiency norms, personal goals taking precedence over relationships, or negative relational experiences that lead to protecting oneself. While these individuals maintain their sense of agency, their limited interactions likely result in declining relationship quality \u003csup\u003e20, 21, 22\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eQ2 (Neglected) represents individuals struggling with both social connection and personal agency. Often feeling ignored or uncertain about others\u0026apos; support, these individuals may suppress their desires in order to accommodate others, or they avoid taking opportunities to develop new relationships out of fear of leaving their current social prospects. Their reduced agency may also stem from underdeveloped competencies, which can make them less appealing as social partners \u003csup\u003e23, 24\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eQ3 (Muted) characterizes individuals with moderately supportive social relationships but low agency. While some of these individuals may prioritize relational goals over personal ones, external factors often constrain their agency. Parents caring for a child, or a husband attending to their wife with dementia, for instance, will have limited time for personal pursuits \u003csup\u003e25\u003c/sup\u003e. Different social connections also can exert significant normative influences on their individual choices \u003csup\u003e26\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eQ4 (Empowered) represents the optimal integration of care and support from social relations and personal agency. These individuals maintain warm, supportive relations that actively encourage their agency and personal pursuits \u003csup\u003e27, 28\u003c/sup\u003e. Their relationships provide a secure base for exploration while remaining strong through active engagement and nurturing.\u003c/p\u003e\n\u003cp\u003ePreliminary research has validated these prototypes and their ability to predict loneliness \u003csup\u003e17\u003c/sup\u003e. Using clustering methods, the findings showed that Q2 or Prototype 2 individuals reported the highest loneliness scores, followed by those in Q1 (Prototype 1) and Q3 (Prototype 3), while Q4 (Prototype 4) individuals experienced the least loneliness. These findings make three important points: First, they are consistent with research showing that\u0026nbsp;relational deficits predict loneliness. Second, they indicate different forms loneliness can take, as evidenced by similar loneliness scores for Prototype 1 (separated) and Prototype 3 (muted) individuals, despite their markedly different levels\u0026nbsp;of social support. Third, they illustrate the role of agency, particularly evident in Prototype 2 (neglected) individuals who exhibit the greatest deficits in both relationships and personal agency.\u003c/p\u003e\n\u003cp\u003eDespite these insights, the Prototypes need to be replicated along with their effects on loneliness. Furthermore, cross-sectional data cannot reveal how membership in a particular Prototype and loneliness change over time. This research aims to advance our understanding of who becomes lonely and why by utilizing four waves of data to track the stability of Prototype configurations and examine how transitions between Prototypes affect loneliness.\u003c/p\u003e"},{"header":"Methods","content":"\u003ch2\u003eParticipants\u003c/h2\u003e\n\u003cp\u003eThis study utilized data from TILDA, the Irish Longitudinal Study on Ageing, which consists of five waves of data collection spanning\u0026nbsp;nine years starting in 2009 \u003csup\u003e29\u003c/sup\u003e.\u0026nbsp;Few samples provide the richness of data on people\u0026rsquo;s social relationships, and this study does so with data collected across time (see supplemental materials for more information on samples). The previously reviewed research on the squad framework relied solely on data from Wave 1, \u003csup\u003e17\u003c/sup\u003e. The current study focuses on Waves 2 through 5.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eMeasures\u003c/h2\u003e\n\u003ch3\u003e\u003cu\u003eLoneliness\u003c/u\u003e\u003c/h3\u003e\n\u003cp\u003eTo assess \u003cem\u003eloneliness\u003c/em\u003e, the TILDA study used a 5-item measure adapted from the UCLA loneliness scale \u003csup\u003e30\u003c/sup\u003e. Sample items included: \u0026ldquo;How often do you feel lonely?\u0026rdquo; and \u0026ldquo;How often do you feel isolated from others?\u0026rdquo; and these were answered on 3-point scales (1 = often, 2 = some of the time, and 3 = hardly ever or never). One item was positively worded, so it was reverse-scored before averaging it with the other items. Higher scores reflect greater feelings of \u003cem\u003eloneliness\u003c/em\u003e.\u0026nbsp;Table 1 presents the descriptive statistics for all the variables in the study across the four data waves. The number of participants (\u003cem\u003en\u003c/em\u003e) listed for each variable represents valid listwise observations after selecting participants who are married and reported having children.\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003e\u003cu\u003eCommunion\u003c/u\u003e\u003c/h3\u003e\n\u003cp\u003eThe \u003cem\u003ecommunion\u003c/em\u003e variable reflects the relational dimension, that is, the support and care individuals report receiving from their relationships, along with any strain present in those relationships \u003csup\u003e31\u003c/sup\u003e. The composite \u003cem\u003ecommunion\u003c/em\u003e variable is derived from the averaged responses to twenty-eight questions. The participants answered the same seven questions separately about their spouse, children, other family members, and friends. Examples of these questions include: \u0026ldquo;How much can you rely on him/her if you have a serious problem?\u0026rdquo; and \u0026ldquo;How much does he/she make too many demands on you?\u0026rdquo; Responses were made on 4-point scales that ranged from 1=a lot to 4=not at all. Relevant items were reverse-scored, so that higher scores mean greater support and care from social relationships.\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003e\u003cu\u003eAgency\u003c/u\u003e\u003c/h3\u003e\n\u003cp\u003e\u003cem\u003eAgency\u003c/em\u003e assessed the level of control and autonomy individuals feel in their lives. Aside from Wave 2, the measure consisted of seven items (Wave 2 had nine items). Additionally, the Wave 2 data set only provided a composite variable, which means it was not possible to calculate Cronbach\u0026rsquo;s alpha for that wave. Examples of the seven common items across the four data waves included \u0026ldquo;I feel that what happens to me is out of my control,\u0026rdquo; \u0026ldquo;I feel free to plan for the future,\u0026rdquo; and \u0026ldquo;My health stops me from doing the things I want to do.\u0026rdquo; Responses to the items were given on 4-point scales (1=often to 4=never) and were reverse-scored so that higher averages indicate greater \u003cem\u003eagency\u003c/em\u003e.\u003c/p\u003e\n\u003ch3\u003e\u003cu\u003eCovariates\u003c/u\u003e\u003c/h3\u003e\n\u003cp\u003eWhen examining how communion and agency impact loneliness, it is important to account for factors related to these variables. Research shows that women typically receive more support from diverse social connections \u003csup\u003e32, 33\u003c/sup\u003e, and age influences the structure of social networks and relationship quality \u003csup\u003e34, 35\u003c/sup\u003e, so the analyses controlled for participant \u003cem\u003egender\u003c/em\u003e and \u003cem\u003eage\u003c/em\u003e. Socioeconomic status is linked to the cultivation of diverse opportunities in life \u003csup\u003e36, 37, 38\u003c/sup\u003e. Therefore, I also controlled for \u003cem\u003eincome\u003c/em\u003e (gross total quintiles) and educational attainment. \u003cem\u003eEducation\u003c/em\u003e was assessed using a 7-point scale in Wave 2 and a 3-point scale in subsequent waves, with higher scores indicating a higher education level. In addition to these variables, I controlled for health, given that negative health conditions are associated with declines in social engagement \u003csup\u003e35\u003c/sup\u003e and greater dependency on existing relationships. \u003cem\u003eHealth\u003c/em\u003e was assessed with a self-report measure that used a 5-point scale ranging from 1 \u0026ldquo;excellent\u0026rdquo; to 5 \u0026ldquo;poor.\u0026rdquo; The response was reversed-scored so that higher ratings indicate better health. Similarly to health, the ability to fulfill activities of independent living should influence social engagement \u003csup\u003e39\u003c/sup\u003e. Some of the activities included shopping for groceries, preparing meals, and managing finances. Hence, I controlled for instrumental activities of daily living (\u003cem\u003eIADLs\u003c/em\u003e), which were assessed on a scale running from 0 = none to 3 = 4+ IADLs.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDepression\u003c/em\u003e can coincide with loneliness \u003csup\u003e40\u003c/sup\u003e, so it is important to account for it. \u003cem\u003eDepression\u003c/em\u003e was evaluated with two items from the C-ESD \u003csup\u003e41\u003c/sup\u003e, which asked participants about their experiences in the week preceding the survey. The two items included: \u0026ldquo;I felt depressed\u0026rdquo; and \u0026ldquo;I felt that everything I did was an effort,\u0026rdquo; scored on 4-point scales ranging from 1 \u0026ldquo;rarely or none of the time (less than 1 day)\u0026rdquo; to 4 \u0026ldquo;all of the time (5-7 days).\u0026rdquo; The two items were averaged, with higher scores indicating greater depression. Finally, I accounted for the number of close relatives and friends, as this predicts loneliness \u003csup\u003e42\u003c/sup\u003e. This question was answered on a scale running from 1 \u0026ndash; 6 (1 = \u0026ldquo;0-4 relations,\u0026rdquo; 6=\u0026rdquo;25 + relations\u0026rdquo;). The responses across the waves indicate an average of approximately 12 close relations.\u003c/p\u003e\n\u003cp\u003eTable 1: Descriptives Statistics for Wave 2 \u0026ndash; Wave 5\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"629\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 15.1033%;\"\u003e\n \u003cp\u003eVariables\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003eWave 2\u003c/p\u003e\n \u003cp\u003e(overall n = 7206)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003eWave 3\u003c/p\u003e\n \u003cp\u003e(overall n = 6397)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.7806%;\"\u003e\n \u003cp\u003eWave 4\u003c/p\u003e\n \u003cp\u003e(overall n = 5713)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.1447%;\"\u003e\n \u003cp\u003eWave 5\u003c/p\u003e\n \u003cp\u003e(overall n = 4978)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 15.1033%;\"\u003e\n \u003cp\u003eLoneliness\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e1.32 (sd=.39)\u003c/p\u003e\n \u003cp\u003eCronbach\u0026rsquo;s = .78\u003c/p\u003e\n \u003cp\u003e(n=4117)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e1.28 (sd=.38)\u003c/p\u003e\n \u003cp\u003eCronbach\u0026rsquo;s = .79\u003c/p\u003e\n \u003cp\u003e(n=3508)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.7806%;\"\u003e\n \u003cp\u003e1.27 (sd=.37)\u003c/p\u003e\n \u003cp\u003eCronbach\u0026rsquo;s = .79\u003c/p\u003e\n \u003cp\u003e(n=3131)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.1447%;\"\u003e\n \u003cp\u003e1.26 (sd=.37)\u003c/p\u003e\n \u003cp\u003eCronbach\u0026rsquo;s = .80\u003c/p\u003e\n \u003cp\u003e(n=2711)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 15.1033%;\"\u003e\n \u003cp\u003eCommunion\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e3.37 (sd=.36)\u003c/p\u003e\n \u003cp\u003eCronbach\u0026rsquo;s = .87\u003c/p\u003e\n \u003cp\u003e(n=3458)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e3.38 (sd=.36)\u003c/p\u003e\n \u003cp\u003eCronbach\u0026rsquo;s =.87\u003c/p\u003e\n \u003cp\u003e(n=2841)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.7806%;\"\u003e\n \u003cp\u003e3.39 (sd=.36)\u003c/p\u003e\n \u003cp\u003eCronbach\u0026rsquo;s = .87\u003c/p\u003e\n \u003cp\u003e(n=2566)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.1447%;\"\u003e\n \u003cp\u003e3.40 (sd=.36)\u003c/p\u003e\n \u003cp\u003eCronbach\u0026rsquo;s = .87\u003c/p\u003e\n \u003cp\u003e(n=2184)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 15.1033%;\"\u003e\n \u003cp\u003eAgency\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e2.11 (sd=.49)\u003c/p\u003e\n \u003cp\u003eCronbach\u0026apos;s = n/a\u003c/p\u003e\n \u003cp\u003e(n=3858)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e2.04 (sd=.54)\u003c/p\u003e\n \u003cp\u003eCronbach\u0026rsquo;s = .70\u003c/p\u003e\n \u003cp\u003e(n=3419)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.7806%;\"\u003e\n \u003cp\u003e2.10 (sd=.54)\u003c/p\u003e\n \u003cp\u003eCronbach\u0026rsquo;s = .71\u003c/p\u003e\n \u003cp\u003e(n=2935)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.1447%;\"\u003e\n \u003cp\u003e2.09 (sd=.52)\u003c/p\u003e\n \u003cp\u003eCronbach\u0026rsquo;s = .70\u003c/p\u003e\n \u003cp\u003e(n=2524)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 15.1033%;\"\u003e\n \u003cp\u003eAge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e63.07 (sd=8.46)\u003c/p\u003e\n \u003cp\u003e(n=4710)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e65.04 (sd=8.21)\u003c/p\u003e\n \u003cp\u003e(n=4055)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.7806%;\"\u003e\n \u003cp\u003e66.70 (sd=8.01)\u003c/p\u003e\n \u003cp\u003e(n=3577)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.1447%;\"\u003e\n \u003cp\u003e68.16 (sd=7.64)\u003c/p\u003e\n \u003cp\u003e(n=3119)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 15.1033%;\"\u003e\n \u003cp\u003eSex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003eMale = 2226\u003c/p\u003e\n \u003cp\u003eFemale = 2484\u003c/p\u003e\n \u003cp\u003e(n=4710)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003eMale = 1919\u003c/p\u003e\n \u003cp\u003eFemale = 2137\u003c/p\u003e\n \u003cp\u003e(n=4056)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.7806%;\"\u003e\n \u003cp\u003eMale = 1695\u003c/p\u003e\n \u003cp\u003eFemale = 1882\u003c/p\u003e\n \u003cp\u003e(n=3577)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.1447%;\"\u003e\n \u003cp\u003eMale = 1484\u003c/p\u003e\n \u003cp\u003eFemale = 1635\u003c/p\u003e\n \u003cp\u003e(n=3119)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 15.1033%;\"\u003e\n \u003cp\u003eEducation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e3.84 (sd=1.54)\u003c/p\u003e\n \u003cp\u003e(n=4693)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e2.14 (sd=.75)\u003c/p\u003e\n \u003cp\u003e(n=4055)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.7806%;\"\u003e\n \u003cp\u003e2.18 (sd=.74)\u003c/p\u003e\n \u003cp\u003e(n=3577)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.1447%;\"\u003e\n \u003cp\u003e2.21 (sd=.74)\u003c/p\u003e\n \u003cp\u003e(n=3119)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 15.1033%;\"\u003e\n \u003cp\u003eFinances\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e3.26 (sd=1.37)\u003c/p\u003e\n \u003cp\u003e(n=3627)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e3.28 (sd=1.34)\u003c/p\u003e\n \u003cp\u003e(n=2929)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.7806%;\"\u003e\n \u003cp\u003e3.29 (sd=1.34)\u003c/p\u003e\n \u003cp\u003e(n=2569)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.1447%;\"\u003e\n \u003cp\u003e3.25 (sd=1.35)\u003c/p\u003e\n \u003cp\u003e(n=2136)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 15.1033%;\"\u003e\n \u003cp\u003eHealth status\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e3.49 (sd=.99)\u003c/p\u003e\n \u003cp\u003e(n=4710)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e3.48 (sd=.99)\u003c/p\u003e\n \u003cp\u003e(n=4055)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.7806%;\"\u003e\n \u003cp\u003e3.50 (sd=.94)\u003c/p\u003e\n \u003cp\u003e(n=3576)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.1447%;\"\u003e\n \u003cp\u003e3.49 (sd=.95)\u003c/p\u003e\n \u003cp\u003e(n=3119)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 15.1033%;\"\u003e\n \u003cp\u003eIADL\u0026rsquo;s\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e.05 (sd=.30)\u003c/p\u003e\n \u003cp\u003e(n=4710)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e.05 (sd=.31)\u003c/p\u003e\n \u003cp\u003e(n=4054)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.7806%;\"\u003e\n \u003cp\u003e.05 (sd=.30)\u003c/p\u003e\n \u003cp\u003e(n=3577)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.1447%;\"\u003e\n \u003cp\u003e.05 (sd=.30)\u003c/p\u003e\n \u003cp\u003e(n=3118)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 15.1033%;\"\u003e\n \u003cp\u003eDepression\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e1.24 (sd=.33)\u003c/p\u003e\n \u003cp\u003er = .53\u003c/p\u003e\n \u003cp\u003e(n=4707)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e1.25 (sd=.52)\u003c/p\u003e\n \u003cp\u003er = .53\u003c/p\u003e\n \u003cp\u003e(n=4052)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.7806%;\"\u003e\n \u003cp\u003e1.24 (sd=.50)\u003c/p\u003e\n \u003cp\u003er = .48\u003c/p\u003e\n \u003cp\u003e(n=3573)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.1447%;\"\u003e\n \u003cp\u003e1.24 (sd=.49)\u003c/p\u003e\n \u003cp\u003er = .42\u003c/p\u003e\n \u003cp\u003e(n=3113)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 15.1033%;\"\u003e\n \u003cp\u003eNumber of relations\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e2.63 (sd=1.18)\u003c/p\u003e\n \u003cp\u003e(n=4708)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20.9857%;\"\u003e\n \u003cp\u003e2.56 (sd=1.07)\u003c/p\u003e\n \u003cp\u003e(n=4054)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.7806%;\"\u003e\n \u003cp\u003e2.37 (sd=.92)\u003c/p\u003e\n \u003cp\u003e(n=3577)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 21.1447%;\"\u003e\n \u003cp\u003e2.42 (sd=.94)\u003c/p\u003e\n \u003cp\u003e(n=3116)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eNote: The statistics are derived from individuals who were married at each time point and reported having children at Wave 1, the only time this variable was assessed.\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003eAnalytic approach\u003c/h2\u003e\n\u003cp\u003eAttrition can affect longitudinal data, as those who remain across data waves may differ from individuals no longer in the study. Additionally, people undergo changes in their social relationships \u003csup\u003e32, 33, 43\u003c/sup\u003e and their ability to exert control over their lives \u003csup\u003e44, 45\u003c/sup\u003e.\u0026nbsp;Therefore, the relative standing of individuals within a sample can shift. Consequently, it is crucial to re-standardize the criterion variables for each wave of data, which also allows us to capture the dynamics of change over time.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFor each data wave, our analytical approach included the following steps: 1) standardize the criterion variables (communion and agency) and conduct a cluster analysis. 2) subject the loneliness scores to ANCOVA to test the hypothesized effects of Prototype on levels of loneliness. 3) conduct analyses to determine how prototype membership changed over time and how those changes influenced loneliness scores.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eMain cluster analysis\u003c/h2\u003e\n\u003cp\u003eDrawing on previous research that identified four distinct patterns in how individuals experience their social relationships and personal agency \u003csup\u003e17\u003c/sup\u003e, we employed k-means clustering to classify participants into four mutually exclusive clusters. This theoretical constraint, along with the requirement that individuals can belong to only one cluster, directed our analytical approach. K-means clustering was chosen based on comparative analyses that demonstrate its superior performance compared to other clustering methods \u003csup\u003e46\u003c/sup\u003e. To maximize data retention while ensuring analytical rigor, we applied pairwise deletion during the clustering analysis.\u0026nbsp;To examine the consistency of the clustering solutions, analyses using listwise deletion were also conducted, which yielded similar clustering results (see supplemental materials).\u003c/p\u003e\n\u003cp\u003eSubsequently, the elbow method suggested a four-cluster solution and possibly a two-cluster solution. However, the four-cluster solution demonstrated superior differentiation of the data. In contrast, the two-cluster solution merely dichotomized observations into favorable (positive scores on both dimensions) and unfavorable clusters, reducing explanatory power (see supplementary materials for the two-cluster solutions). We also generated ASW metrics (average silhouette width) for the four-cluster solutions. The ASW metric ranges from \u0026minus;1 to 1, where \u0026minus;1 indicates the worst clustering solution, and values close to 0 suggest overlap among clusters. Positive values signify a good match to the assigned cluster. The ASW scores were as follows: W2 = .36; W3 = .36; W4 = .35; and W5 = 0.35. Therefore, four clusters (prototypes) were kept based on theory and analyses of relevant metrics. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe results shown in Panel\u0026nbsp;1 and Table 2 indicate the four clusters align with the four predicted prototypes.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e-\u0026nbsp;Prototype\u0026nbsp;1 (Separated) includes individuals who have high agency but low communion.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e-\u0026nbsp;Prototype\u0026nbsp;2 (Neglected) consists of individuals who score low on both dimensions.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e-\u0026nbsp;Prototype\u0026nbsp;3 (Muted) features individuals with lower agency but higher communion.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e-\u0026nbsp;Prototype\u0026nbsp;4 (Empowered) is characterized by individuals who score high on both dimensions.\u003c/p\u003e\n\u003cp\u003eThe larger number of individuals in Prototype 4 (empowered) likely reflects the positive skew observed in quality measures of social relationships \u003csup\u003e33\u003c/sup\u003e. Nevertheless, each\u0026nbsp;prototype\u0026nbsp;had a sufficient number of individuals.\u003c/p\u003e\n\u003cp\u003eTable 2: The data represent four clusters across Waves 2-5, showing sample sizes (n) and standardized scores (z) for the Communion and Agency dimensions.\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003eCluster\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003eWave\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003en\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003eCommunion (z)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003eAgency (z)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003ePrototype 1 (Separated)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e1085\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e-0.64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e860\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e-0.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e809\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e-0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e0.29\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e548\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e-0.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e0.26\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003ePrototype 2 (Neglected)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e582\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e-1.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e-1.47\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e475\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e-1.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e-1.43\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e470\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e-1.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e-1.41\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e461\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e-1.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e-1.44\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003ePrototype 3 (Muted)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e996\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e0.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e-0.69\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e1019\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e0.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e-0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e749\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e0.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e-0.69\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e795\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e0.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e-0.41\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003ePrototype 4 (Empowered)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e1353\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e0.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e0.93\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e1082\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e0.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e1030\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e0.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e0.92\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 29.76%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 10.24%;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e840\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e0.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 20%;\"\u003e\n \u003cp\u003e1.04\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch2\u003eLoneliness results for the different prototypes\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eAmong the prototypes, ANCOVAs revealed consistent patterns of loneliness across all waves. In Wave 2 (F(3, 3057) = 181.53, p \u0026lt; .001), Prototype 2 (neglected) showed the highest loneliness (M = 1.64, SE = .02), followed by Prototype 1 (separated) (M = 1.32, SE = .01) and Prototype 3 (muted; M = 1.32, SE = .01), while the empowered Prototype showed the lowest loneliness scores (M = 1.16, SE = .01). Prototype 1 and Prototype 3 did not differ (F \u0026lt; 1), but all other comparisons were significant (ps \u0026lt; .001; Bonferroni adjusted significance value per wave for nine comparisons = .005).\u003c/p\u003e\n\u003cp\u003eThis pattern replicated across subsequent waves, with analyses controlling for prior loneliness. Wave 3 showed a significant prototype effect (F(3, 2315)=65.80, p \u0026lt;.001), replicating Wave 2\u0026apos;s pattern. Prototype 2 (neglected) had the highest loneliness (M = 1.47; SE = .02) versus Prototype 1 (separated, M = 1.26; SE = .01), Prototype 3 (muted, M = 1.27; SE = .01), and Prototype 4 (empowered, M = 1.20; SE = .01). Prototype 4 differed from all others (all ps \u0026lt; 0.001). Prototypes 1 and 3 did not differ (F\u0026lt;1).\u003c/p\u003e\n\u003cp\u003eWave 4 showed similar patterns (F(3, 2063)=67.14, p \u0026lt;.001). Prototype 2 (neglected) had the highest loneliness scores (M = 1.47, SE = .02) versus all others (Prototype 1: M = 1.27, SE = .01; Prototype 3: M = 1.29, SE = .01; Prototype 4: M = 1.17, SE = .01; all ps \u0026lt; 0.001). Prototype 4 had the lowest loneliness scores and differed from all others (ps \u0026lt; 0.001), while Prototypes 1 and 3 did not differ.\u003c/p\u003e\n\u003cp\u003eWave 5 maintained these findings (F(3, 1685)=50.58, p \u0026lt;.001). Prototype 2 (M = 1.44, SE = .02) differed from all others (Prototype 1: M = 1.25, SE = .01; Prototype 3: M = 1.22, SE = .01; Prototype 4: M = 1.18, SE = .01; all ps \u0026lt; 0.001). Prototype 4 differed from all others (ps \u0026lt; 0.001), with Prototypes 1 and 3 showing no difference.\u003c/p\u003e\n\u003ch2\u003eChange in Prototypes Across Time\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eThe analyses revealed consistent prototypes across data waves that predict loneliness, even though the clusters are based on re-standardized criterion variables. While the same prototypes appeared consistently, this does not mean the same individuals ended up in the same Prototype at the next time interval. Subsequent analyses assessed change over time and, more importantly, whether these changes resulted in differences in loneliness. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePanel 2 presents change percentages among prototypes over time (see supplemental materials for Sankey diagrams). The prototype an individual belongs to in the previous wave has a strong chance of being the same one they will be in subsequently. Another observation is that individuals in Prototype 4 (empowered) are highly unlikely to move to Prototype 2 (neglected), and the reverse is also true.\u003c/p\u003e\n\u003ch2\u003ePrototype Transitions and Loneliness\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eI examined how cluster transitions influenced levels of loneliness by categorizing these transitions into favorable and unfavorable groups. Based on earlier findings indicating that Prototype\u0026nbsp;2 (neglected) is associated with the highest levels of loneliness, Prototype\u0026nbsp;4 (empowered) with the lowest, and Prototypes 1 and 3 are in between, we classified transitions to or stays in Prototype\u0026nbsp;2 as unfavorable, while transitions to or stays in Prototype\u0026nbsp;4 were deemed favorable. Transitions from Prototype 2 to Prototype 1 or 3 were categorized as favorable, whereas transitions from Prototype 4 to Prototype 1 or 3 were categorized as unfavorable. Finally, we did not analyze transitions that simply swapped memberships between Prototype\u0026nbsp;1 (separated) and Prototype\u0026nbsp;3 (muted), given the findings that they did not differ in loneliness. We conducted analyses using both this binning approach and a simplified version that focused solely on transitions or stays\u0026nbsp;that occurred for Prototype 2 and Prototype 4. The Bonferroni-adjusted significance value per wave pair for four comparisons is .012.\u003c/p\u003e\n\u003cp\u003eFor change in Prototypes from Wave 2 to Wave 3, ANCOVA results (controlling for Wave 3 covariates and previous loneliness) indicated a significant effect of bin (F(1, 1452)=77.60, p \u0026lt;.001). Individuals in unfavorable Prototypes (M = 1.35, SE = .01) had higher loneliness scores than those in favorable Prototypes (M = 1.22, SE = .01). With the simpler assumptions (analyzing only Prototype 2 and 4 transitions/stays), results remained similar (F(1, 1024)=124.97, p \u0026lt;.001). Prototype 2 individuals (neglected, M = 1.47, SE = .02) had higher loneliness scores than Prototype 4 individuals (empowered, M = 1.20, SE = .01).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFor Waves 3-4, the effect persisted under both sets of assumptions. The primary analysis showed a significant bin effect (F(1, 1293)=116.33, p \u0026lt;.001), with unfavorable Prototype individuals (M = 1.37, SE = .01) reporting higher loneliness than those in more favorable Prototypes (M = 1.20, SE = .01). The restricted analysis revealed similar patterns (F(1, 984)=160.93, p \u0026lt;.001), with Prototype 2 individuals (M = 1.46, SE = .02) showing higher loneliness than Prototype 4 (M = 1.17, SE = .01).\u003c/p\u003e\n\u003cp\u003eFor Waves 4-5, the pattern was similar. The primary analysis showed a significant bin effect (F(1, 1127) = 60.02, p \u0026lt; .001), with unfavorable Prototype individuals (M = 1.32, SE = .01) reporting higher loneliness than favorable Prototype individuals (M = 1.19, SE = .01). The restricted analysis confirmed these findings (F(1, 814) = 79.94, p \u0026lt; .001), with Prototype 2 individuals (M = 1.41, SE = .02) showing higher loneliness than Prototype 4 (M = 1.19, SE = .01).\u003c/p\u003e\n\u003cp\u003eAnalyses without covariates (available in supplemental materials) showed similar, though stronger, patterns.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eFor each data wave, the results indicated the same four Prototypes based on individuals' reports of the support from their social relations and their personal agency. Further, loneliness levels varied significantly across these Prototypes, with the neglected Prototype reporting the highest loneliness levels and the empowered Prototype the lowest. The muted and separated Prototypes showed intermediate loneliness levels, similar to each other but distinct from both Prototype 2 (neglected) and Prototype 4 (empowered). This pattern remained stable across Waves 3, 4, and 5, despite re-standardizing the criterion variables. While Prototype membership showed stability between waves, considerable movement occurred among them. Analyses of these transitions across three consecutive wave pairs revealed that individuals who moved to or remained in the unfavorable Prototype experienced greater loneliness compared to those who moved to or remained in the more favorable Prototype.\u003c/p\u003e \u003cp\u003eThe results challenge the dominant view that loneliness primarily arises from social or relational deficits. The findings indicate that personal agency plays an equally important role. The interplay between social connection and agency, as demonstrated in our prototypes, illustrates how loneliness arises from complex patterns in how individuals configure their social relations and their positions within them. The emergence of these patterns across data waves, even as individuals transition to different prototypes, suggests that these configurations reflect meaningful social-psychological states for individuals.\u003c/p\u003e \u003cp\u003eFurthermore, the sequencing between Prototype transitions and the subsequent experience of loneliness offers compelling evidence that relationship-agency configurations affect loneliness levels. Although longitudinal designs cannot definitively determine causation due to potential unmeasured confounding factors, the observed shifts in Prototype membership occurring before changes in loneliness deepen our theoretical understanding beyond cross-sectional correlations. If these findings are confirmed by future research, they imply that interventions could be tailored to assist individuals not only in improving their social relationships but also in recognizing the need for personal development so they can be more independent, socially attractive, and assert themselves within their relationships.\u003c/p\u003e \u003cp\u003eLimitations and Future Directions\u003c/p\u003e \u003cp\u003eThe current research has limitations. The measurement intervals between waves may have overlooked shorter-term fluctuations in social support, strain, and changes in personal agency. Additionally, the findings could be influenced by cultural context due to variations in norms and traditions.\u003c/p\u003e \u003cp\u003eFuture research could benefit from designs that use more frequent measurements, and that also examine factors thought to promote or impede positive relationships, along with the development of personal agency. Additionally, cross-cultural studies might reveal how different social norms and values influence the relationship between agency, social connection, and loneliness. However, considering global trends toward achieving greater agency in life, along with the positive wellbeing effects that accompany such changes \u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e, we may also discover more consistency across different populations \u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. Finally, it would be useful to examine individual differences, such as in attachment styles, to determine whether different ways of relating to others predispose certain individuals to end up in one prototype over another.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis research examined the nature of people\u0026rsquo;s social relationships and their sense of personal agency as key factors predicting loneliness. The findings suggest that meaningful relationships not only serve as sources of social support but also allow individuals to express their agency and to be themselves. These results should contribute to the discussion on loneliness and its causes, helping to shift the perspective away from a narrow view of loneliness as primarily arising from relational deficits.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eCompeting Interest Declaration\u003c/p\u003e\n\u003cp\u003eI declare no competing financial or non-financial interests as defined by Nature Portfolio.\u003c/p\u003e\n\u003cp\u003eAuthor Contribution\u003c/p\u003e\n\u003cp\u003eI am solely responsible for conceptualizing the research, processing and analyzing the data, and writing the manuscript. However, Tanfu Shi assisted with generating the following figures (Panels 1 and 2 in the manuscript and Panel 1 in the supplemental materials).\u003c/p\u003e\n\u003cp\u003eAuthorship\u003c/p\u003e\n\u003cp\u003eI have read the Nature Portfolio Authorship Policy and confirm that the manuscript complies with it. I am the sole author and not an LLM.\u003c/p\u003e\n\u003cp\u003eData Availability\u003c/p\u003e\n\u003cp\u003eAccess to the data can be obtained by applying to the owners and curators of the Irish Longitudinal Study on Ageing (TILDA) at Trinity College Dublin (https://tilda.tcd.ie).\u003c/p\u003e\n\u003cp\u003eCode Availability\u003c/p\u003e\n\u003cp\u003eThe code for processing and analyzing the data and for generating the figures is available from the author.\u003c/p\u003e\n\u003cp\u003eHuman Research Participants\u003c/p\u003e\n\u003cp\u003eThis study is based on secondary analyses of existing data collected by the TILDA researchers. As they note, \u0026ldquo;Ethical approval for each wave was obtained from the Faculty of Health Sciences Research \u003cem\u003eEthics\u003c/em\u003e Committee at Trinity College Dublin.\u0026rdquo; (tilda.tcd.ie/publications/reports/pdf/Report_CohortMaintenance.pdf). I received IRB approval from the University of Illinois to conduct secondary analyses of the data.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eThe New York Times. The loneliness epidemic in America [Opinion]. Retrieved from https://www.nytimes.com/2023/04/30/opinion/loneliness-epidemic-america.html (2023).\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. Social connection for mental health and well-being: Addressing isolation and loneliness (2023).\u003c/li\u003e\n\u003cli\u003eWorld Health Organization. Commission on Social Connection. Retrieved January 30, 2025, from https://www.who.int/groups/commission-on-social-connection (n.d.).\u003c/li\u003e\n\u003cli\u003eHawkley, L. C., \u0026amp; Cacioppo, J. T. Loneliness matters: A theoretical and empirical review of consequences and mechanisms. \u003cem\u003eAnnals of Behavioral Medicine\u003c/em\u003e, \u003cstrong\u003e40,\u003c/strong\u003e 218\u0026ndash;227 (2010).\u003c/li\u003e\n\u003cli\u003eCacioppo, S., Grippo, A. J., London, S., Goossens, L., \u0026amp; Cacioppo, J. T. Loneliness: Clinical import and interventions. \u003cem\u003ePerspectives on Psychological Science\u003c/em\u003e, \u003cstrong\u003e10,\u003c/strong\u003e 238\u0026ndash;249 (2015).\u003c/li\u003e\n\u003cli\u003eNature Portfolio. How to submit to Nature collections. Retrieved February 18, 2025, from https://www.nature.com/collections/eefieehehi/how-to-submit (n.d.).\u003c/li\u003e\n\u003cli\u003ePeplau, L. A., \u0026amp; Perlman, D. Perspectives on loneliness. In L. A. Peplau \u0026amp; D. Perlman (Eds.), \u003cem\u003eLoneliness: A sourcebook of current theory, research, and therapy\u003c/em\u003e (pp. 1\u0026ndash;8). New York, NY: Wiley (1982).\u003c/li\u003e\n\u003cli\u003ePerlman, D., \u0026amp; Peplau, L. A. Toward a social psychology of loneliness. \u003cem\u003ePersonal Relationships\u003c/em\u003e, \u003cstrong\u003e3,\u003c/strong\u003e 31\u0026ndash;56 (1981).\u003c/li\u003e\n\u003cli\u003eMasi, C. M., Chen, H. Y., Hawkley, L. C., \u0026amp; Cacioppo, J. T. A meta-analysis of interventions to reduce loneliness. \u003cem\u003ePersonality and Social Psychology Review\u003c/em\u003e, \u003cstrong\u003e15,\u003c/strong\u003e 219\u0026ndash;266 (2011).\u003c/li\u003e\n\u003cli\u003eAndersson, L. Loneliness research and interventions: A review of the literature. \u003cem\u003eAging \u0026amp; Mental Health\u003c/em\u003e, \u003cstrong\u003e4,\u003c/strong\u003e 264\u0026ndash;274 (1998).\u003c/li\u003e\n\u003cli\u003eBarreto, M., Victor, C., Hammond, C., Eccles, A., Richins, M. T., \u0026amp; Qualter, P. Loneliness around the world: Age, gender, and cultural differences in loneliness. \u003cem\u003ePersonality and Individual Differences\u003c/em\u003e, \u003cstrong\u003e169,\u003c/strong\u003e 110066 (2021).\u003c/li\u003e\n\u003cli\u003eHawkley, L. C., Browne, M. W., \u0026amp; Cacioppo, J. T. How can I connect with thee? Let me count the ways. \u003cem\u003ePsychological Science\u003c/em\u003e, \u003cstrong\u003e16,\u003c/strong\u003e 798\u0026ndash;804 (2005).\u003c/li\u003e\n\u003cli\u003eRodriguez, M., Schertz, K. E., \u0026amp; Kross, E. How people think about being alone shapes their experience of loneliness. \u003cem\u003eNature Communications\u003c/em\u003e, \u003cstrong\u003e16,\u003c/strong\u003e 1594 (2025).\u003c/li\u003e\n\u003cli\u003eTan, A. J., Mancini, V., Gross, J. J., Goldenberg, A., Badcock, J. C., Lim, M. H., ... \u0026amp; Preece, D. A. Loneliness versus distress: A comparison of emotion regulation profiles. \u003cem\u003eBehaviour Change\u003c/em\u003e, \u003cstrong\u003e39,\u003c/strong\u003e 180-190 (2022).\u003c/li\u003e\n\u003cli\u003eBrown, D. E. Human universals and their implications. In N. Roughley (Ed.), \u003cem\u003eBeing humans: Anthropological universality and particularity in transdisciplinary perspectives\u003c/em\u003e (pp. 156\u0026ndash;174). Berlin, Germany: Walter de Gruyter (2000).\u003c/li\u003e\n\u003cli\u003eYbarra, O., Chan, E., Park, H., Burnstein, E., Monin, B., \u0026amp; Stanik, C. Life\u0026apos;s recurring challenges and the fundamental dimensions: An integration and its implications for cultural differences and similarities. \u003cem\u003eEuropean Journal of Social Psychology\u003c/em\u003e, \u003cstrong\u003e38,\u003c/strong\u003e 1083\u0026ndash;1092 (2008).\u003c/li\u003e\n\u003cli\u003eYbarra, O., \u0026amp; Chan, T. The S(quad) model, a pattern approach for understanding the individual and their social network relations: Application to loneliness. \u003cem\u003eFrontiers in Social Psychology\u003c/em\u003e, \u003cstrong\u003e1,\u003c/strong\u003e 1278671 (2023).\u003c/li\u003e\n\u003cli\u003eRyan, R. M., \u0026amp; Deci, E. L. Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. \u003cem\u003eAmerican Psychologist\u003c/em\u003e, \u003cstrong\u003e55,\u003c/strong\u003e 68\u0026ndash;78 (2000).\u003c/li\u003e\n\u003cli\u003eBandura, A. Self-efficacy. In V. S. Ramachaudran (Ed.), \u003cem\u003eEncyclopedia of Human B\u003c/em\u003eehavior (Vol. 4, pp. 71\u0026ndash;81). San Diego, CA: Academic Press (1994).\u003c/li\u003e\n\u003cli\u003eBurt, R. S. Decay functions. \u003cem\u003eSocial Networks\u003c/em\u003e, \u003cstrong\u003e22,\u003c/strong\u003e 1\u0026ndash;28 (2000).\u003c/li\u003e\n\u003cli\u003eOswald, D. L., Clark, E. M., \u0026amp; Kelly, C. M. Friendship maintenance: An analysis of individual and dyad behaviors. \u003cem\u003eJournal of Social and Clinical Psychology\u003c/em\u003e, \u003cstrong\u003e23,\u003c/strong\u003e 413\u0026ndash;441 (2004).\u003c/li\u003e\n\u003cli\u003eOgolsky, B. G., \u0026amp; Bowers, J. R. A meta-analytic review of relationship maintenance and its correlates. \u003cem\u003eJournal of Social and Personal Relationships\u003c/em\u003e, \u003cstrong\u003e30,\u003c/strong\u003e 343\u0026ndash;367 (2013).\u003c/li\u003e\n\u003cli\u003eBerkowitz, L. Resistance to improper dependency relationships. \u003cem\u003eJournal of Experimental Social Psychology\u003c/em\u003e, \u003cstrong\u003e5,\u003c/strong\u003e 283\u0026ndash;294 (1969).\u003c/li\u003e\n\u003cli\u003eLeary, M. R., Jongman-Sereno, K. P., \u0026amp; Diebels, K. J. The pursuit of status: A self-presentational perspective on the quest for social value. In J. T. Cheng, J. L. Tracy, \u0026amp; C. Anderson (Eds.), \u003cem\u003eThe Psychology of Social Status\u003c/em\u003e (pp. 159\u0026ndash;178). New York, NY: Springer (2014).\u003c/li\u003e\n\u003cli\u003ePerry-Jenkins, M., \u0026amp; Gerstel, N. Work and family in the second decade of the 21st century. \u003cem\u003eJournal of Marriage and Family\u003c/em\u003e, \u003cstrong\u003e82,\u003c/strong\u003e 420\u0026ndash;453 (2020).\u003c/li\u003e\n\u003cli\u003eCohen, S., \u0026amp; Lemay, E. P. Why would social networks be linked to affect and health practices? \u003cem\u003eHealth Psychology\u003c/em\u003e, \u003cstrong\u003e26,\u003c/strong\u003e 410\u0026ndash;417 (2007).\u003c/li\u003e\n\u003cli\u003eBowlby, J. \u003cem\u003eAttachment\u003c/em\u003e. New York, NY: Basic Books (1969).\u003c/li\u003e\n\u003cli\u003eAinsworth, M. D. S., Blehar, M. C., Waters, E., \u0026amp; Wall, S. N. \u003cem\u003ePatterns of attachment: A psychological study of the strange situation\u003c/em\u003e. New York, NY: Psychology Press (2015).\u003c/li\u003e\n\u003cli\u003eThe Irish Longitudinal Study on Ageing. TILDA: The Irish longitudinal study on ageing. Retrieved February 18, 2025, from https://tilda.tcd.ie (n.d.).\u003c/li\u003e\n\u003cli\u003eRussell, D. W. UCLA Loneliness Scale (Version 3): Reliability, validity, and factor structure. \u003cem\u003eJournal of Personality Assessment\u003c/em\u003e, \u003cstrong\u003e66,\u003c/strong\u003e 20\u0026ndash;40 (1996).\u003c/li\u003e\n\u003cli\u003eSchuster, T. L., Kessler, R. C., \u0026amp; Aseltine, R. H., Jr. Supportive interactions, negative interactions, and depressed mood. \u003cem\u003eAmerican Journal of Community Psychology\u003c/em\u003e, \u003cstrong\u003e18,\u003c/strong\u003e 423\u0026ndash;438 (1990).\u003c/li\u003e\n\u003cli\u003eAntonucci, T. C. Social relations: An examination of social networks, social support, and sense of control. In J. E. Birren \u0026amp; K. W. Schaie (Eds.), \u003cem\u003eHandbook of the Psychology of Aging\u003c/em\u003e (pp. 427\u0026ndash;453). San Diego, CA: Academic Press (2001).\u003c/li\u003e\n\u003cli\u003eFiori, K. L., Antonucci, T. C., \u0026amp; Akiyama, H. Profiles of social relations among older adults: A cross-cultural approach. \u003cem\u003eAgeing \u0026amp; Society\u003c/em\u003e, \u003cstrong\u003e28,\u003c/strong\u003e 203\u0026ndash;231 (2008).\u003c/li\u003e\n\u003cli\u003eCarstensen, L. L., Isaacowitz, D. M., \u0026amp; Charles, S. T. Taking time seriously: A theory of socioemotional selectivity. \u003cem\u003eAmerican Psychologist\u003c/em\u003e, \u003cstrong\u003e54,\u003c/strong\u003e 165\u0026ndash;181 (1999).\u003c/li\u003e\n\u003cli\u003eHuxhold, O., Fiori, K. L., \u0026amp; Windsor, T. D. The dynamic interplay of social network characteristics, subjective well-being and health: The costs and benefits of socio-emotional selectivity. \u003cem\u003ePsychology and Aging\u003c/em\u003e, \u003cstrong\u003e28,\u003c/strong\u003e 3\u0026ndash;16 (2013).\u003c/li\u003e\n\u003cli\u003eLachman, M. E., \u0026amp; Weaver, S. L. The sense of control as a moderator of social class differences in health and well-being. \u003cem\u003eJournal of Personality and Social Psychology\u003c/em\u003e, \u003cstrong\u003e74,\u003c/strong\u003e 763\u0026ndash;773 (1998).\u003c/li\u003e\n\u003cli\u003eTaylor, S. E., \u0026amp; Seeman, T. E. Psychosocial resources and the SES‐health relationship. \u003cem\u003eAnnals of the New York Academy of Sciences\u003c/em\u003e, \u003cstrong\u003e896,\u003c/strong\u003e 210\u0026ndash;225 (1999).\u003c/li\u003e\n\u003cli\u003eKraus, M. W., Piff, P. K., \u0026amp; Keltner, D. Social class, sense of control, and social explanation. \u003cem\u003eJournal of Personality and Social Psychology\u003c/em\u003e, \u003cstrong\u003e97,\u003c/strong\u003e 992\u0026ndash;1004 (2009).\u003c/li\u003e\n\u003cli\u003eCurl, A. L., Stowe, J. D., Cooney, T. M., \u0026amp; Proulx, C. M. Giving up the keys: How driving cessation affects engagement in later life. \u003cem\u003eThe Gerontologist\u003c/em\u003e, \u003cstrong\u003e54,\u003c/strong\u003e 423\u0026ndash;433 (2014).\u003c/li\u003e\n\u003cli\u003eCacioppo, J. T., Hughes, M. E., Waite, L. J., Hawkley, L. C., \u0026amp; Thisted, R. A. Loneliness as a specific risk factor for depressive symptoms: Cross-sectional and longitudinal analyses. \u003cem\u003ePsychology and Aging\u003c/em\u003e, \u003cstrong\u003e21,\u003c/strong\u003e 140\u0026ndash;151 (2006).\u003c/li\u003e\n\u003cli\u003eRadloff, L. S. The CES-D Scale: A self-report depression scale for research in the general population. \u003cem\u003eApplied Psychological Measurement\u003c/em\u003e, \u003cstrong\u003e1,\u003c/strong\u003e 385\u0026ndash;401 (1977).\u003c/li\u003e\n\u003cli\u003eRussell, D., Peplau, L. A., \u0026amp; Cutrona, C. E. The revised UCLA Loneliness Scale: Concurrent and discriminant validity evidence. \u003cem\u003eJournal of Personality and Social Psychology\u003c/em\u003e, \u003cstrong\u003e39,\u003c/strong\u003e 472\u0026ndash;480 (1980).\u003c/li\u003e\n\u003cli\u003eCarstensen, L. L. Selectivity theory: Social activity in life-span context. \u003cem\u003eAnnual Review of Gerontology and Geriatrics\u003c/em\u003e, \u003cstrong\u003e11,\u003c/strong\u003e 195\u0026ndash;217 (1991).\u003c/li\u003e\n\u003cli\u003eLachman, M. E., Neupert, S. D., \u0026amp; Agrigoroaei, S. The relevance of control beliefs for health and aging. In K. W. Schaie \u0026amp; S. L. Willis (Eds.), \u003cem\u003eHandbook of the Psychology of Aging\u003c/em\u003e (pp. 175\u0026ndash;190). San Diego, CA: Academic Press (2011).\u003c/li\u003e\n\u003cli\u003eHeckhausen, J., Wrosch, C., \u0026amp; Schulz, R. A motivational theory of life-span development. \u003cem\u003ePsychological Review\u003c/em\u003e, \u003cstrong\u003e117,\u003c/strong\u003e 32\u0026ndash;60 (2010).\u003c/li\u003e\n\u003cli\u003ePreud\u0026apos;homme, G., Duarte, K., Dalleau, K., Lacomblez, C., Bresso, E., Sma\u0026iuml;l-Tabbone, M., ... \u0026amp; Girerd, N. Head-to-head comparison of clustering methods for heterogeneous data: A simulation-driven benchmark. \u003cem\u003eScientific Reports\u003c/em\u003e, \u003cstrong\u003e11,\u003c/strong\u003e 4202 (2021).\u003c/li\u003e\n\u003cli\u003eWelzel, C., \u0026amp; Inglehart, R. Agency, values, and well-being: A human development model. \u003cem\u003eSocial Indicators Research\u003c/em\u003e, \u003cstrong\u003e97,\u003c/strong\u003e 43\u0026ndash;63 (2010).\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Panels","content":"\u003cp\u003ePanels 1 and 2 are available in the Supplementary Files section\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Loneliness, Personal Agency, Social Connection, Longitudinal Data, Cluster Analysis, Prototypes","lastPublishedDoi":"10.21203/rs.3.rs-6099787/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6099787/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis longitudinal study examined how social connections and personal agency influenced loneliness over time. Utilizing four waves of data from the Irish Longitudinal Study on Ageing, analyses consistently revealed four distinct prototypes within each data wave that reflect combinations of social connection and personal agency. Subsequent analyses showed that one prototype (low agency, low connection) experienced the highest levels of loneliness, while another (high agency, high connection) reported the lowest. The remaining two prototypes exhibited equivalent, intermediate loneliness despite the stark differences in social connection levels between them. Tracking transitions between prototypes across waves revealed that shifts toward less favorable prototypes predicted increased loneliness, whereas transitions toward more favorable prototypes predicted reduced loneliness. These findings significantly enhance our understanding of loneliness, an experience generally thought to be driven by relational deficits, highlighting the crucial role of personal agency.\u003c/p\u003e","manuscriptTitle":"Beyond Social Deficits:\nPersonal Agency and Social Connection Shape Loneliness Over Time","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-03-26 05:43:53","doi":"10.21203/rs.3.rs-6099787/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"communications-psychology","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"commspsychol","sideBox":"Learn more about [Communications Psychology](http://www.nature.com/commspsychol/)","snPcode":"44271","submissionUrl":"https://mts-commspsychol.nature.com/cgi-bin/main.plex","title":"Communications Psychology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Communications Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"0f0a51fe-8f40-4353-a920-37b59f3a23dd","owner":[],"postedDate":"March 26th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":44971461,"name":"Biological sciences/Psychology"},{"id":44971462,"name":"Biological sciences/Psychology/Human behaviour"}],"tags":[],"updatedAt":"2025-10-30T07:09:56+00:00","versionOfRecord":{"articleIdentity":"rs-6099787","link":"https://doi.org/10.1038/s44271-025-00329-z","journal":{"identity":"communications-psychology","isVorOnly":false,"title":"Communications Psychology"},"publishedOn":"2025-10-29 04:00:00","publishedOnDateReadable":"October 29th, 2025"},"versionCreatedAt":"2025-03-26 05:43:53","video":"","vorDoi":"10.1038/s44271-025-00329-z","vorDoiUrl":"https://doi.org/10.1038/s44271-025-00329-z","workflowStages":[]},"version":"v1","identity":"rs-6099787","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6099787","identity":"rs-6099787","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}