Sage Journals: Discover world-class research

Abstract

Past research has linked individual differences in loneliness to Big Five personality traits. However, experience sampling studies also show intrapersonal fluctuations in loneliness. These may reflect situational factors as well as stable individual differences. Here, for the first time, we study the relationship between personality traits and within-person variation in loneliness. In a one-week experience sampling study, n = 285 Nepali participants reported feelings of loneliness three times a day (3597 observations). We use Bayesian mixed-effects location scale models to simultaneously estimate the relationship between Big Five personality traits and (a) mean levels and (b) within-person variability in loneliness. We also test whether these relationships vary depending on whether participants were alone or in the company of others. More neurotic individuals felt lonelier, especially (but not only) when they were alone. These individuals also experienced greater intrapersonal fluctuations in loneliness. These findings extend the differential reactivity hypothesis, according to which individuals vary in loneliness due to differential reactivity to social situations, and accord with the conceptual view of neuroticism as hyperreactivity to social stressors. In addition, we document the role of personality and social context in people’s everyday experience of loneliness in a non-WEIRD population.

Plain language summary

People differ in how much loneliness they experience. When asked how lonely they feel on average, some people report that they feel very lonely; others, that they do not feel lonely at all. However, people also differ in how much their feelings of loneliness vary from situation to situation. Two people who report the same average level of loneliness may have very different experiences if one of them persistently feels somewhat lonely, whereas the other sometimes feel very lonely and sometimes not at all. In this study, we investigated how such ups and downs in loneliness relate to the individual’s personality. 285 Nepali participants first reported their personality in terms of five major dimensions (the Big Five). Then, three times a day for one week, we asked them how lonely they felt, collecting a total of 3,597 reports. We found that participants with higher levels of neuroticism – one of the Big Five dimensions – on average felt lonelier, and especially so when they were alone. Moreover, these individuals also experienced greater ups and downs in loneliness. In addition, we also found that participants higher in agreeableness felt less lonely on average. These results support the view that people high in neuroticism are more sensitive to social stressors and experience more frequent emotional ups and downs.

Keywords

loneliness personality big Five experience sampling location scale model

Introduction

Loneliness is the feeling that one’s social relations are lacking in quantity and/or quality (Hawkley & Cacioppo, 2010; Perlman & Peplau, 1981; Rotenberg, 1994). Experiencing persistent loneliness is associated with lower levels of physical, mental, and subjective well-being (Hawkley & Cacioppo, 2007; Holt-Lunstad et al., 2015; VanderWeele et al., 2012; van Winkel et al., 2017; Wei et al., 2005; but see Rohrer & Lucas, 2020, for a cautionary note on interpreting similar links between subjective well-being and health as causal). On the one hand, trait loneliness is linked to individual differences in personality. A recent meta-analysis found that all Big Five traits were associated with individual differences in (trait) loneliness (Buecker et al., 2020). The strongest and most robust associations were a negative link between extraversion and loneliness and a positive link between neuroticism and loneliness. On the other hand, people’s experiences of loneliness are also influenced by their social context and fluctuate from situation to situation. For example, adolescents experience more loneliness when they are alone than when they are with somebody (van Roekel et al., 2018). This indicates that the experience of loneliness varies both between people as well as within individuals across time.

The focus on associations between personality traits and trait loneliness on the one hand and between situational factors and state loneliness on the other hand may hide context-dependent effects of personality traits on (state) loneliness (Funder, 2008). The expression of personality traits may, for example, depend on the affordances of the situation (Thielmann et al., 2020). One prominent example in the context of loneliness is the differential reactivity hypothesis. In its original formulation, this hypothesis states that high trait loneliness is associated with greater state loneliness specifically when individuals are on their own (Cacioppo et al., 2003; van Roekel et al., 2018). Interpreted more broadly, depending on their personality traits, individuals may vary in their reactivity to situational factors. Neuroticism, in particular, may be associated with greater reactivity to being alone. Consequently, the link between personality traits and state loneliness may be context-dependent. However, past studies on links between personality and loneliness have largely overlooked such situation-specific factors.

Crucially, individuals may also differ in the extent to which they experience fluctuations in loneliness, independent of their trait level. Yet, there has been little attention to what makes some people experience frequent fluctuations in loneliness while others have a more stable trait level. Among broad Big Five personality traits, neuroticism has been defined as hyperreactivity to social stressors (Bolger & Schilling, 1991; Costa & McCrae, 1992; Eysenck & Eysenck, 1985; Hills & Argyle, 2001; Hisler et al., 2020), resulting in heightened situation-to-situation or day-to-day variability in affect (Ching et al., 2014; Geukes et al., 2017; Kuppens et al., 2007; Mader et al., 2023). Similarly, neuroticism – as well as other personality traits – may be associated with the degree to which a person’s experiences of loneliness fluctuate throughout the day. One person may have a consistently low (or high) level of loneliness, whereas another may cycle through frequent ups and downs, experiencing both low and high levels of loneliness within the span of a day or week. Estimating the degree to which such individual differences in within-person variability occur, and identifying personality traits that are predictive of greater or lesser variability, provides a richer picture of individual differences in the experience of loneliness.

Loneliness as a trait and state

The degree to which people experience loneliness differs both between individuals and within individuals across time. Research on loneliness, however, has largely treated trait and state loneliness as distinct constructs (e.g. van Roekel et al., 2018). A trait approach views loneliness as a relatively stable property of the person (Fung et al., 2017; van Roekel et al., 2018). A recent meta-analysis of longitudinal studies found high rank-order stability for trait loneliness during adulthood (Mund et al., 2020). One-year stability exceeded r = .60 for much of adulthood before declining in very old age. Across the lifespan, loneliness exhibits a U-shaped trajectory (Mund et al., 2020). These findings suggest that loneliness is trait-like to a similar degree as broad-factor personality traits such as the Big Five.

In contrast, the state approach views loneliness as a state fluctuating from one moment to another. From this perspective, an individual can oscillate between feeling more or less lonely from situation to situation or day to day. Experience sampling and daily diary studies have reported significant within-person variability in loneliness (Buecker et al., 2023; Harper et al., 2020; Rinderknecht, 2020; Tam & Chan, 2019). For example, Tam and Chan (2019) reported an intraclass correlation (ICC) for state loneliness of ICC = 0.43; suggesting that loneliness varied more within persons (i.e. from situation to situation; 57%) than between persons (43%). Similarly, Harper et al. (2020) and Rinderknecht (2020) found that around 50% of the variance in loneliness occurred within persons. These findings evidence significant within-person variability in loneliness from occasion to occasion beyond a more stable trait level.

Here, we take the view that loneliness has both trait and state components. This is consistent with contemporary personality theories, which posit that traits are manifested in fluctuating states (e.g. Baumert et al., 2017; Horstmann & Ziegler, 2020). Whole trait theory, in particular, holds that traits are reflections of the density distributions of states; that is, that individuals experience varying (personality) states, but differ between each other in the distribution of these states (Fleeson, 2001; Fleeson & Jayawickreme, 2015). Individual differences can thus refer to measures of central tendency (or level, e.g. the arithmetic mean) and to measures of dispersion (or variability, e.g. the variance). We expect that individuals experience varying levels of loneliness across situations, and that they differ from each other in both the level (i.e. location) and variability (i.e. scale) of their experienced states. Moreover, we explore whether these associations between personality and state loneliness depend on the social context.

Individual and contextual differences in loneliness

Personality predictors of loneliness

Past research has shown that personality traits are robust correlates of trait loneliness. A recent meta-analysis found that extraversion, agreeableness, conscientiousness and openness were negatively related to loneliness, whereas neuroticism was positively related to loneliness (Buecker et al., 2020). When controlling for other Big Five personality traits, the small association between openness and loneliness was not significant. For other personality traits the association was significant, albeit reduced in magnitude when compared to the bivariate associations. The strongest associations existed for extraversion (r = –.37) and neuroticism (r = .36). With the exception of one unpublished experience sampling dataset, these correlations were, however, based on one-time measures of trait loneliness. Importantly, these sizable associations with multiple personality traits suggest that loneliness is not merely a nuance of one broad personality dimension (Mõttus et al., 2017), but may be multiply determined.

Contextual predictors of loneliness

Prima facie, the most plausible predictor of loneliness is contact with other people. Indeed, individuals experience more (state) loneliness when they are alone than when they are with somebody else (Larson, 1990; van Roekel et al., 2018). However, social isolation – the objective lack of relationships and social interaction – is only weakly correlated with trait loneliness (Cacioppo et al., 2009; Hawkley et al., 2003; Tiwari, 2013). In a large sample of older Americans, several indicators of social isolation only correlated with loneliness at r = .20 (Coyle & Dugan, 2012). This correlation may even reflect social withdrawal following experiences of loneliness (van Winkel et al., 2017). Loneliness and social isolation also have distinct associations with physical and mental well-being (Beller & Wagner, 2018; Holt-Lunstad et al., 2015). Thus, there exists a clear discrepancy between objective social isolation and subjective feelings of loneliness.

Relationships and social interactions can also vary in quality. It has been suggested that it is the quality and closeness of relationships rather than their quantity which matters for well-being (Perlman & Peplau, 1981; Russell et al., 2012). While some studies have found that contact with close others is more strongly associated with (lower) loneliness (van Roekel et al., 2015; Sandstrom & Dunn, 2014, Study 1), others found that even contacts with ‘weak ties’ – acquaintances or even strangers – may alleviate loneliness (Epley & Schroeder, 2014; Rinderknecht et al., 2021; Sandstrom & Dunn, 2014, Study 2). The highest-quality studies suggest that positively valenced contact may buffer against loneliness, whereas negatively valenced contact may even increase them (Tam & Chan, 2019; van Roekel et al., 2014, 2015). Unfortunately, causal evidence on the effects of the quality and closeness of contact is lacking.

Person X situation interactions

Crucially, personality and contextual causes of loneliness may also interact with each other. The differential reactivity hypothesis posits that individual differences in trait loneliness are driven by differential reactivity to situational stressors (Cacioppo et al., 2003; van Roekel et al., 2018). In several studies of adolescents, individuals with high (measured) trait loneliness reported feeling lonelier when they were alone, compared to individuals with low trait loneliness; this difference was smaller when the participants had company (van Roekel et al., 2018). This suggests that differential reactivity to being alone is one of the social-cognitive mechanisms which produce individual differences in trait loneliness. However, individuals with high trait loneliness still experienced more state loneliness even when they had intimate company, indicating that differential reactivity to being alone cannot fully account for individual differences in trait loneliness.

Differential reactivity to context factors may also explain why broad personality traits (such as the Big Five) are associated with individual differences in trait loneliness. In particular, neuroticism has been associated with heightened reactivity to situational stressors, which is thought to result both in heightened negative affect and in greater emotional volatility (Bolger & Schilling, 1991; Costa & McCrae, 1992; Eysenck & Eysenck, 1985; Hisler et al., 2020; Mader et al., 2023). Thus, the well-documented association between neuroticism and trait loneliness may be driven by differential reactivity to situational stressors, in particular, being alone. Consequently, we would expect this association to be stronger when individuals are alone (rather than in company). Importantly, more neurotic individuals should experience not only heightened levels of loneliness but also greater fluctuations in state loneliness across situations.

Few studies have directly examined person × situation interactions in loneliness (Reissmann et al., 2021). However, several recent studies on loneliness during the COVID-19 pandemic have found that measures which restricted social contact were associated with changes in the associations between personality and loneliness. In Germany, social contact restrictions were associated with greater increases in loneliness among more extraverted, neurotic and conscientious individuals (Alt et al., 2021; Entringer & Gosling, 2022). More extraverted individuals, in particular, may suffer from such restrictions because they experience a greater need for social contact. The otherwise negative association between extraversion and loneliness may thus be driven by selection into social situations, and may be diminished when individuals are alone. Correlations between personality traits and trait loneliness, which make up the bulk of the literature, may obscure such context-dependent links between personality and loneliness. In contrast, experience sampling studies make it possible to test whether the associations between personality traits and loneliness differ depending on whether the participant is alone or in the company of others.

Personality and within-person variability in loneliness

As whole trait theory suggests, individual differences encompass the entire shape of the distribution of states (Fleeson, 2001; Fleeson & Jayawickreme, 2015). Thus, individuals may differ not only in the level of loneliness they experience across situations but also in the extent of fluctuations from situation to situation. Some individuals may feel chronically high (or low) levels of loneliness; others, however, may experience more frequent ups and downs. Such individual differences in within-person variability may arise from differential reactivity to the social context (Cacioppo et al., 2003; Danneel et al., 2018; Lay et al., 2019; van Roekel et al., 2018). This is consistent with the prominent definition of neuroticism as hyperreactivity to social stressors (Bolger & Schilling, 1991; Costa & McCrae, 1992; Eysenck & Eysenck, 1985; Hills & Argyle, 2001; Hisler et al., 2020): if both the more and the less neurotic individual experience similar kinds of situations, but only the more neurotic individual is affected in their loneliness by being alone, the more neurotic individual will experience greater fluctuations in loneliness from situation to situation than the less neurotic individual. Thus, differential reactivity to social stressor can explain why personality traits may be associated with both the mean level (i.e. location) and cross-situational variability (i.e. scale) of loneliness.

Intriguingly, the link between personality traits and within-person variability may itself differ across social contexts. This may occur when classes of situations contain different subsets of stressors. For example, social situations may vary in the quality of social contact (e.g. van Roekel et al., 2014). Some studies in adolescents suggest that highly (trait) lonely individuals experience the greatest benefits from social contact, but that this is specific to positive interactions with close others (van Roekel et al., 2014, 2018). Extending this finding to neuroticism, highly neurotic individuals might experience stable (high) levels of loneliness when alone, but varying levels of loneliness when in the company of others, depending on the quality of the social contact; less neurotic individuals would experience largely stable low levels of loneliness independent of their social context. In this scenario, unless one accounts for the quality of social contact, within-person variability in loneliness would not only differ depending on the individual’s personality, but also depending on the particular combination of personality and situation.

Research on loneliness has, so far, largely focused on personality and social contexts as predictors of trait and state levels of loneliness. However, theory suggests that these factors should also be associated with differences in intrapersonal variability. Moreover, previous studies have shown that broad personality traits predict such within-person variability in affect more generally. For example, previous studies have linked neuroticism to greater and extraversion to lower within-person variability in affect (Ching et al., 2014; Eid & Diener, 1999; Geukes et al., 2017; Kuppens et al., 2007; Mader et al., 2023). These personality traits may also be associated with intrapersonal variability in loneliness. Such associations may be further moderated by the social context, in particular, the presence of others. This is a hitherto overlooked dimension of individual difference in loneliness, which may open up new opportunities for interventions.

The present study

In the present study, we pursued several interlinked goals. First, we assessed the associations between Big Five personality traits and mean levels of state loneliness. Second, we tested whether these associations varied across situations in which individuals were alone versus in the company of others. Third, we tested whether Big Five personality traits were also linked to within-person variability in loneliness. Fourth, we examined whether these links between personality and within-person variability differed across situations in which individuals were alone versus in the company of others. In sum, we provide a comprehensive test of the associations between Big Five personality traits and the level and intrapersonal variability of loneliness, as well as differences in these associations across varying social contexts.

Methods

Ethics and open scientific practices

This paper uses data from a broader study on personality, loneliness and affect. The study was approved by the national ethical review board of Nepal, the Nepal Health Research Council (NHRC), with the ERB Protocol Registration no. 413/2020 P. Data, analysis code and a list of all measures are available on the OSF (https://osf.io/u2gbe/).

Sample

The sample size was determined in the context of the broader study and before the present analyses were planned. The initial sample consisted of N = 315 participants. Of these, n = 11 did not provide any experience sampling responses and a further n = 19 did not complete the intake survey and are not included in the analyses reported here. In addition, we dropped two duplicate responses from the experience sampling survey. The final sample thus consisted of n = 285 individuals (age: M = 22.50, SD = 5.14, range 17–55; gender: 87% female, 13% male) who provided a total of k = 3597 responses. This constitutes a response rate of 60%. Additionally, for some models we dropped k = 74 responses from our analyses because we could not determine whether the participant was alone or in the company of others, which also lowered the sample size by n = 1. This subsample thus consisted of n = 284 individuals who provided a total of k = 3523 responses.

Participants were recruited through snowball sampling by contacting students through social media sites and email lists. As the participants were either enrolled in or had completed undergraduate or postgraduate education in English, the survey was administered in English. Most participants were students (67%); the remainder were employed or self-employed (26%) or unemployed (6%). 95% of the participants lived in multi-person households (M = 4.39; range 2–14, excluding two participants who indicated households of 24 and 80). Most participants indicated being unmarried (89%) and living with their nuclear family (75%; remainder joint or communal living arrangements).

Historical and local setting

The data was collected during the first wave of the COVID-19 pandemic in Nepal, over the course of twelve days from April 20, 2020 to May 1, 2020. Participants could enrol themselves in the study between April 20 and April 25, 2021. The government of Nepal announced the first lockdown effective from March 24, 2020 till July 21, 2020 (Sharma et al., 2021). During this time, all businesses, schools, colleges and government offices were closed down except for a few hours a day. Mobility was restricted as public transportation was shut down and private vehicles were only allowed to move with a governmental pass. The general population was required to stay home except to obtain essential goods or in case of emergencies. Thus, participants were largely confined to their homes during the study period.

Much attention has been paid in recent years to psychology’s reliance on samples from Western, educated, industrialised, rich and democratic (WEIRD) populations (Henrich et al., 2010). We use this framework to contextualise our sample. Nepal is a South Asian country of about 29 million people. The net primary school enrolment rate has only reached 80% in recent years World Bank, (n.d.); however, our sample consisted mostly of students in tertiary education, which makes up a small part of the population. The economy is largely agricultural, and the per-capita gross national income was only US$1190 in 2020 (World Bank, (n.d.). Nepal is a relatively new democracy, having experienced a democratic revolution in 2006. Within the WEIRD framework, our sample is thus non-Western, relatively educated, less industrialised, less rich and (newly) democratic.

Measures

Personality was assessed using the Big Five Inventory (Goldberg, 1993). The Big Five Inventory is a 44-item self-report measure of the Five Factor Model of Personality (FFM), designed to assess the dimensions extraversion, agreeableness, conscientiousness, neuroticism and openness. Means, standard deviations and zero-order correlations for the BFI subscales are shown in Table 1. Reliabilities of the subscales in terms of McDonald’s omega total were acceptable to good, ranging from

ω_{t}

= .70 (Agreeableness) to

ω_{t}

= .81 (neuroticism). Personality profiles of participants who completed the intake but did not provide any ESM responses did not differ substantively from those who did provide ESM responses (Table 2).

Table 1.

Descriptive statistics for the BFI subscales. Means and standard deviations are shown for the final dataset (‘Included’) and for participants who did not provide ESM responses (‘Excluded’), With 95% credible intervals and Cohen’s d for differences between both subgroups. The correlation matrix shows 95% credible intervals with reliabilities $ω_{t}$ on the diagonal.

	Included		Excluded		Zero-order correlations
	M	SD	M	SD	95% CI	d	E	A	C	N	O
Extraversion	3.12	.68	3.07	.49	[–0.37, 0.41]	0.07	.74
Agreeableness	3.82	.56	3.93	.44	[–0.43, 0.21]	−0.19	[.07, .29]	.70
Conscientiousness	3.31	.61	3.24	.48	[–0.30, 0.43]	0.13	[.23, .43]	[.21, .41]	.74
Neuroticism	3.28	.80	2.99	.56	[–0.15, 0.81]	0.38	[–.40, −.20]	[–.39, −.19]	[–.56, −.39]	.81
Openness	3.66	.42	3.65	.41	[–0.26, 0.24]	0.02	[.04, .26]	[.05, .27]	[.01, .22]	[–.09, .13]	.72

Note. E = Extraversion, A = Agreeableness, C = Conscientiousness, N = Neuroticism, O = Openness.

Table 2.

Associations between personality dimensions and social contact (vs. being alone). Estimates from a series of univariate models and from a multivariate model.

	Univariate			Multivariate
	B	Error	95% CI	B	Error	95% CI
Intercept				0.14	0.02	[0.11, 0.17]
Extraversion	0.09	0.02	[0.04, 0.14]	−0.16	0.04	[–0.24, −0.07]
Agreeableness	0.05	0.03	[–0.01, 0.12]	0.01	0.03	[–0.05, 0.07]
Conscientiousness	0.05	0.03	[–0.00, 0.11]	0.09	0.03	[0.04, 0.14]
Neuroticism	−0.05	0.02	[–0.09, −0.01]	0.04	0.03	[–0.02, 0.11]
Openness	−0.12	0.04	[–0.20, −0.04]	−0.01	0.03	[–0.07, 0.04]

State loneliness was assessed using a single item, ‘How lonely are you feeling right now?’ Responses were recorded on 0-100 rating scale (0 = not lonely at all, 100 = very lonely). The mean reported loneliness was 28.12 (estimated marginal mean = 28.31). Participants whose responses were excluded because they did not complete the personality measure reported only marginally more loneliness, M = 29.93, EMM = 34.25, B = 5.94, 95% CI = [-6.20, 17.65].

Participants were also asked to describe who they were with at the time, where they were at the time and what they were doing at the time. We manually coded these open text responses based on whether they indicated the presence of another person (e.g. ‘family’, ‘brother’, ‘sister’; k = 2394) or being alone (e.g. ‘alone’, ‘no one’; k = 1253); k = 79 uncategorisable. Most contacts were close family members (family, parents, siblings, children and partners; k = 2273), other family members (k = 64), work contacts (k = 22) and undefined others (k = 43). Responses whose social context could not be classified based on the description were rated higher in loneliness (M = 48.1) than both situations in which participants had company (M = 24.8) and in which they were alone (M = 33.2). To ensure that our classification did not skew the results, we ran two additional specifications of the model containing this variable, assigning all unclassifiable responses either to ‘alone’ or to ‘in company’. Neither specification changed the substantive interpretation of our results (Tables S1, S2).

Procedures

Data were collected using the Expiwell mobile app (https://app.expiwell.com/). Participants were contacted through email and Facebook. They were asked to download the mobile app. Usage of the app was explained through a tutorial video. Each participant was given a unique code which would lead them to the survey. At this point, participants provided consent to participate in the study. Subsequently, the participants completed the intake survey, which included the BFI as well as a number of other measures not relevant to the present study (a full list of measures is available on the OSF).

The experience sampling phase started from the day the participants enrolled in the study and lasted for seven days. Each day, participants were asked to complete three surveys in the morning, afternoon and evening. The morning survey was made available from 6:30 AM to 11:59 AM. The afternoon survey was made available from 12 PM to 4:59 PM and the evening survey was made available from 5 PM to 9 PM. Participants were alerted through a notification in the app. Each survey included the loneliness measure as well as questions about the context (e.g. where were they, who were they with, what were they doing?), and a measure of affect not analysed here. The evening surveys also included a day reconstruction measure for daily activities. The data were securely stored on an online server.

Analytic approach

Mixed-effects location scale models

To simultaneously estimate the relationship between individual differences in personality and (a) between-person variation in loneliness and (b) within-person variance in loneliness, we employed Bayesian mixed-effects location scale models (MELSM; Bürkner, 2017a; Hedeker et al., 2008; Rast et al., 2012). MELSMs are an extension of the more familiar linear mixed-effects models. Linear mixed-effects models assume that the (between-subjects) variance of random effects and the (within-subjects) error variance are homogeneous across subjects. MELSMs relax these assumptions by including (sub-)models for the between-subjects and the within-subjects variance (Hedeker et al., 2008). This allows predictors to influence these sources of variation.

Consider first a basic mixed-effects model for repeated observations j = 1, 2, …, k (level 1) nested within subjects i (level 2),

y_{i j} = x_{i j}^{'} β + ν_{i} + ϵ_{i j,}

where

y_{i j}

indicates the score on y of subject i on occasion j,

x_{i j}

is a p × 1 vector of predictors, and β is the corresponding p × 1 vector of regression coefficients.

ν_{i}

indicates the random effects, which are assumed to be normally distributed with mean zero and variance

σ_{ν}^{2}

. The errors

ϵ_{i j}

are likewise assumed to be normally distributed with mean zero and variance

σ_{ϵ}^{2}

. In this model, between- and within-subjects variances are thus homogeneous across subjects and occasions.

The MELSM relaxes the homogeneity assumptions of linear mixed-effects models by allowing the within-person variance $σ_{ϵ}^{2}$ to differ across individuals i and across time points j. Because this is a variance, we use an exponential function to ensure that the estimates are positive,

σ_{ϵ_{i j}}^{2} = \exp (w_{i j}^{'} τ + ω_{i}) .

Introducing this scale submodel allows the value of the variance to change with the predictors $w_{i j}$ and their coefficients τ; $ω_{i}$ indicates the random effects representing i’s variability above and beyond the effect of these predictors. Predictors of the within-person variance $σ_{ϵ_{i j}}^{2}$ can be occasion-specific (i.e. at level 1) or person-specific (i.e. at level 2). Because we use an exponential function in the scale submodel, the resulting coefficients are on a log scale. Positive values of τ mean that larger values of some predictor w* are associated with greater within-subjects variance. That is, the larger w*, the more subjects vary from occasion to occasion. The inverse is true of τ < 0. If $y_{i j}$ are scores on self-reported loneliness and $w_{i}$ are participants’ neuroticism scores, a positive τ would mean that more neurotic individuals exhibit greater ups and downs in their self-reported loneliness.

Zero inflation

A particular challenge arises because the outcome variable, loneliness, exhibits significant inflation at both ends. This is commonly observed in slider scale data (Vuorre, 2019). We therefore modelled the outcome as both left- and right-censored. Censoring assumes a continuous underlying (latent) variable. Other treatments for inflated data instead assume distinct data generation processes, for example, zero-or-one-inflated beta regression (Liu & Kong, 2015). In the case of loneliness measured using a slider scale, a continuous latent variable appears (more) plausible. This also facilitates the interpretation of the estimates.¹

Model specification

We implement a Bayesian MELSM using brms (Bürkner, 2017b), an R package for implementing Bayesian mixed-effects models using the probabilistic programming language Stan (Stan Development Team, 2022). We use weakly informative default priors (see supplementary materials) and model loneliness on a Student’s t distribution.² We include random intercepts and random slopes for the variable coding for the presence of others in both submodels and allow the random effects to be correlated across submodels. We centred the continuous predictors and contrast-coded the variable coding for the presence of others (−0.5 = ‘alone’, 0.5 = ‘in company’). All models were run with four chains and 10,000 iterations and converged without issue (no divergent transitions after warm-up; all $\hat{R} \leq 1.01$ ; see supplementary materials for full model performance information).

Inference criteria

We report point estimates and 95% credible intervals (CIs) based on quantiles. Credible intervals are based on the posterior distribution and describe a range which contains 95% of the values in the posterior distribution. Therefore, a 95% credible interval can be interpreted as ‘given the observed data, the effect has 95% probability of falling within this range’ (Makowski et al., 2019). In line with recommendations for Bayesian mixed-effects models, we do not adjust for multiple comparisons (Gelman et al., 2012).

Software

Mixed-effects location scale models were fitted using brms (Bürkner, 2017b). Data wrangling was done using R and tidyverse (R Core Team, 2022; Wickham et al., 2019). Factor analyses were conducted using the R packages psych (Revelle, 2022) and lavaan (Rosseel, 2012).

Results

Descriptive results

Zero-order and partial correlations

Table 3 presents zero-order and partial correlations between the average loneliness score across time points and the Big Five traits. The partial correlations show a positive correlation between neuroticism and loneliness and negative correlations indistinguishable from zero between all other traits and loneliness.

Table 3.

Zero-order and partial correlations between mean loneliness and Big Five traits, with 95% credible intervals and Bayes factors.

	Zero-order Correlations			Partial Correlations
	r	95% CI	BF	r	95% CI	BF
Extraversion	−.18	[–.30, −.08]	18.63	−.05	[–.17, .06]	0.21
Agreeableness	−.22	[–.33, −.11]	>100	−.11	[–.22, .00]	0.75
Conscientiousness	−.26	[–.37, −.16]	>100	−.09	[–.19, .03]	0.41
Neuroticism	.32	[.22, .42]	>100	.20	[.08, .31]	52.87
Openness	−.06	[–.17, .06]	0.22	−.02	[–.14, .09]	0.15

Distribution and within-person variability of loneliness

The average reported loneliness was 28.12 (out of 100). To estimate the within-person variability of loneliness, we fitted an intercept-only mixed-effects model without scale parameters to the raw data. This yielded an ICC = .53 (i.e. about half of the variance in loneliness occurred within persons). Loneliness ratings exhibited significant inflation at both extremes of the scale, with 26.7% of responses reporting ‘0’ (i.e. no loneliness) and 4.6% reporting ‘100’ (i.e. maximum loneliness). We therefore apply models accounting for censoring at both extremes.

To further examine within-person variability, we added a null submodel for the scale (i.e. a model including only a scale-related random effect). This revealed variance due to the random intercept of the location, $σ_{ν}^{2} = 2.27$ , 95% CI = [2.12, 2.42], and due to the scale-related random effect, $σ_{ω}^{2} = 0.93$ , 95% CI = [0.82, 1.04]. The former indicates that individuals differed in the average level of loneliness they reported, whereas the latter indicates that they differed in the within-person variance of loneliness. The random intercept of the location and the scale-related random effect were weakly correlated, r = .18, 95% CI = [.02, .34], indicating that individuals who reported feeling lonelier on average also experienced greater fluctuations in loneliness.

Inferential results

To examine the associations of personality traits with the level (i.e. location) and within-person variance (i.e. scale) of loneliness, we estimated a series of MELSMs. First, we fitted a model including only the Big Five personality traits as predictors in the location and the scale submodel. We assess the main effects of personality traits on loneliness using this model (Model 1). Second, we entered the social context (alone = −0.5, in company = 0.5) to assess whether people felt lonelier when they were alone rather than in company (Model 2). Third, to test whether individuals with different personality traits reacted differently to varying social contexts, we also included the second-order interactions between each personality trait and the social context (Model 3). We included random intercepts for participants and, in the latter two models, random slopes for social context. We first focus on the effects on the location (i.e. level) of loneliness (Table 4), then on the effects on the scale (i.e. within-person variance; Table 5). Random effects and random effect correlations are reported in Tables 6 and 7.

Table 4.

Location effects. Fixed effects parameter estimates for the effects of personality and social context on the level of loneliness. Social context was contrast-coded (alone: −0.5 vs. in company: 0.5). 95% credible intervals not containing zero are shown in bold.

	Model 1				Model 2				Model 3
	B	Error	LL	UL	B	Error	LL	UL	B	Error	LL	UL
Intercept	20.91	1.90	17.16	24.63	21.12	1.88	1.88	17.43	21.13	1.87	17.47	24.84
Openness	−5.43	4.67	−14.57	3.83	−5.34	4.50	−14.19	3.52	−6.06	4.64	−15.02	3.09
Conscientiousness	−6.15	3.92	−13.67	1.45	−6.86	3.69	−14.01	0.48	−6.68	3.68	−13.88	0.56
Extraversion	−1.11	3.06	−7.09	4.96	−0.17	3.02	−6.08	5.78	0.04	3.03	−5.85	5.99
Agreeableness	−8.05	3.68	−15.57	−0.78	−8.53	3.70	−15.93	−1.33	−7.69	3.63	−14.79	−0.56
Neuroticism	8.38	2.96	2.57	14.22	8.46	2.91	2.85	14.13	8.98	2.95	3.29	14.76
Social context					−2.80	1.03	−4.95	−0.97	−4.01	0.98	−6.01	−2.18
SC × openness									0.22	2.00	−3.89	3.97
SC × conscientiousness									1.53	1.47	−1.37	4.48
SC × extraversion									−3.58	1.33	−6.29	−1.05
SC × agreeableness									−0.36	1.68	−3.58	3.02
SC × neuroticism									−2.79	1.17	−5.17	−0.57

Note. Error = standard deviation of the posterior distribution; LL, UL = lower and upper limit, 95% credible interval; SC = Social Context. Ns = 3597 (Model 1), 3523 (Models 2, 3).

Table 5.

Scale effects. Fixed effects parameter estimates for the effects of personality and social context on the within-person variance of loneliness. Social context was contrast-coded (alone: −0.5 vs. in company: 0.5). 95% credible intervals not containing 1 are shown in bold.

	Model 1				Model 2				Model 3
	exp(τ)	Error	LL	UL	exp(τ)	Error	LL	UL	exp(τ)	Error	LL	UL
Intercept	9.65	1.08	8.38	11.16	9.72	1.08	8.41	11.18	9.50	1.08	8.20	11.00
Openness	1.28	1.17	0.95	1.73	1.30	1.16	0.97	1.74	1.28	1.16	0.95	1.73
Conscientiousness	0.85	1.13	0.67	1.09	0.90	1.13	0.71	1.14	0.89	1.13	0.70	1.14
Extraversion	1.10	1.11	0.90	1.35	1.04	1.11	0.85	1.28	1.06	1.11	0.87	1.31
Agreeableness	0.96	1.13	0.76	1.22	0.91	1.13	0.72	1.16	0.90	1.13	0.71	1.14
Neuroticism	1.29	1.10	1.07	1.57	1.28	1.10	1.06	1.55	1.34	1.10	1.10	1.62
Social context					1.03	1.08	0.89	1.19	1.05	1.08	0.91	1.21
SC × openness									0.90	1.18	0.65	1.25
SC × conscientiousness									1.03	1.14	0.79	1.34
SC × extraversion									0.90	1.12	0.72	1.13
SC × agreeableness									1.27	1.14	0.98	1.65
SC × neuroticism									0.94	1.11	0.76	1.16

Note. Error = standard deviation of the posterior distribution; LL, UL = lower and upper limit, 95% credible interval; SC = Social Context. Ns = 3597 (Model 1), 3523 (Models 2, 3). 95% credible intervals not containing 0 are shown in bold.

Table 6.

Standard deviations of random effects parameters for location and scale.

	Model 1				Model 2				Model 3
	Est	Error	LL	UL	Est	Error	LL	UL	Est	Error	LL	UL
Location effects
Intercept	30.11	1.63	27.07	33.43	29.59	1.59	26.63	32.87	29.51	1.62	26.42	32.86
Social context					2.61	1.59	0.29	6.13	4.48	1.31	1.87	7.11
Scale effects
Intercept	0.88	0.06	0.78	1.00	0.85	0.06	0.75	0.97	0.85	0.06	0.74	0.97
Social context					0.30	0.14	0.03	0.56	0.24	0.13	0.02	0.51

Note. Error = standard deviation of the posterior distribution; LL, UL = lower and upper limit, 95% credible interval. Ns = 3597 (Model 1), 3523 (Models 2, 3).

Table 7.

Random effects correlations. 95% credible intervals not containing zero are shown in bold.

	Model 1				Model 2				Model 3
	Est	Error	LL	UL	Est	Error	LL	UL	Est	Error	LL	UL
Intercept location, Slope location					−0.08	0.28	−0.67	0.46	−0.10	0.20	−0.48	0.29
Intercept location, Intercept scale	0.11	0.09	−0.06	0.27	0.13	0.09	−0.04	0.30	0.13	0.09	−0.04	0.30
Slope location, Intercept scale					−0.53	0.24	−0.92	−0.02	−0.48	0.20	−0.84	−0.07
Intercept location, Slope scale					0.43	0.26	−0.17	0.86	0.32	0.31	−0.42	0.84
Slope location, Slope scale					−0.10	0.39	−0.81	0.66	−0.08	0.40	−0.81	0.69
Intercept scale, Slope scale					0.31	0.26	−0.27	0.78	0.33	0.30	−0.37	0.83

Note. Error = standard deviation of the posterior distribution; LL, UL = lower and upper limit, 95% credible interval; Slope = random slope for social context. Ns = 3597 (Model 1), 3523 (Models 2, 3).

Personality and social context predict the level of loneliness

We first consider individual differences in the mean level of loneliness across time points (Model 1, Table 4). Overall, more neurotic individuals felt lonelier, B = 8.38, 95% CI = [2.57, 14.22], whereas more agreeable individuals felt less lonely, B = −8.05, 95% CI = [−15.57, −0.78]. For all other traits, the 95% credible interval contains zero, which is consistent with the partial correlations reported in Table 3. These effects were robust to the inclusion of the social context (Model 2, Table 4). Unsurprisingly, participants felt less lonely when they were in company than when they were alone, B = −2.80, 95% CI = [−4.95, −0.97].

The multiplicative model identifies two person × situation interactions on the level of loneliness (Model 3, Table 4). Whether participants were alone or in company moderated the effect of extraversion, B = −3.58, 95% CI = [−6.29, −1.05], as well as the effect of neuroticism, β = −2.79, 95% CI = [−5.17, −0.57]. Figure 1 illustrates these interactions. To further probe the shape of the significant interactions, we computed estimated marginal effects of extraversion and neuroticism in both social contexts. For extraversion, Figure 1 shows a crossover interaction: more extraverted individuals felt lonelier when alone, but less lonely when in company. However, 95% credible intervals for marginal effects of extraversion included zero when participants were alone, B = 1.83, 95% CI = [−4.35, 7.98], and when they had company, B = −1.75, 95% CI = [−7.84, 4.16]. For neuroticism, Figure 1 shows that less neurotic individuals experienced little loneliness regardless of social context, whereas highly neurotic individuals experienced loneliness, especially when they were alone. This was borne out by the marginal effects, which showed that the effect of neuroticism was stronger when participants were alone, B = 10.37, 95% CI = [4.36, 16.30], than when they were in company, B = 7.59, 95% CI = [1.86, 13.30].

Figure 1.

Conditional effects of Big Five personality traits on average loneliness when participants were alone (blue) and when they were in company (red). 95% credible intervals for the interaction terms involving extraversion and neuroticism did not contain zero; the other intervals did.

Neuroticism predicts within-person variability in loneliness

Big Five traits also predicted how much an individual’s feelings of loneliness varied from time point to time point (Model 1, Table 5). More neurotic individuals experienced greater ups and downs in their loneliness, exp(τ) = 1.29, 95% CI = [1.07, 1.57]. No other personality traits predicted within-person variability. In addition, participants did not fluctuate more (or less) when they were alone than when they were in the company of others, exp(τ) = 1.03, 95% CI = [0.89, 1.19] (Model 2, Table 5). Finally, neither the effect of neuroticism on within-person variability nor that of any other personality trait differed across social contexts (Model 3, Table 5).

Account for personality reduces the association between level and variability of loneliness

The models also yield some insights through the random effects. In particular, once we control for the (fixed) effects of personality traits, the location and scale of loneliness are no longer correlated, r = .11, 95% CI = [−.06, 27] (Model 1, Table 7). This suggests that the correlation between mean levels and within-person variability in loneliness is due to shared underlying personality traits. However, even after accounting for personality traits and social context, there remains unexplained variation in the location and scale of loneliness (Model 3, Table 6).

Discussion

Some people feel lonelier than others, an individual difference which is robustly associated with Big Five personality traits (Buecker et al., 2020). However, most people also vary in their experience from situation to situation; in particular, they feel lonelier when alone and less lonely when they are in the company of close others (van Roekel et al., 2015). Yet, following the differential reactivity hypothesis, individuals may also differ in how they respond to contextual factors. Partly as a consequence, some individuals may experience persistently low or high levels of loneliness, whereas others may experience greater fluctuations. Here, we use experience sampling data to examine whether Big Five personality traits and social context can explain differences in the level and intrapersonal variability of loneliness. We find that more neurotic individuals both experienced higher levels of loneliness and fluctuated more from situation to situation. The link between neuroticism and the level of loneliness was context-dependent: more neurotic individuals consistently felt lonelier, but especially so when they were alone rather than in company. However, this interaction did not extend to the degree of intrapersonal fluctuation: more neurotic individuals experienced greater ups and downs in loneliness across situations in which they were alone and across situations in which they had company.

Neuroticism and differential reactivity

The differential reactivity hypothesis posits that individuals vary in trait loneliness because they differ in their reactivity to being alone: highly lonely people feel lonelier when they are alone, but not necessarily when they have company (Cacioppo et al., 2003; van Roekel et al., 2018). Past research on Dutch adolescents had shown that the presence of close others moderates the link between trait loneliness and state loneliness, although highly (trait) lonely individuals still felt lonelier when they were in company, compared to less lonely individuals (van Roekel et al., 2018). In contrast to this approach, we did not assess trait loneliness directly. Rather, in line with whole trait theory (Fleeson, 2001; Fleeson & Jayawickreme, 2015), we take the trait of loneliness to be represented by the distribution of state loneliness. However, consistent with the differential reactivity hypothesis, we found that the random location intercept and the random slope intercept of loneliness were correlated, meaning that individuals who reported greater loneliness also experienced greater variability in loneliness across situations.

More broadly, differential reactivity means that individuals differ in their responses to situational cues. In particular, neuroticism has been conceptualised as hyperreactivity to social stressors (Bolger & Schilling, 1991; Costa & McCrae, 1992; Eysenck & Eysenck, 1985; Hills & Argyle, 2001; Hisler et al., 2020). Indeed, we found that highly neurotic individuals experienced greater loneliness when they were alone rather than in company. Yet, contrary to suggestions that social contacts could serve as a buffer against loneliness (Wieczorek et al., 2021), more neurotic individuals felt lonelier than less neurotic individuals even when they had company. Thus, though neurotic individuals do benefit from social contact, an alternative reading of our data could be that low neuroticism serves as a buffer against the negative effects of being alone, in particular, feeling lonely. Recent studies have shown that some people experience solitude (i.e. being alone) mostly negatively, whereas others experience it mostly positively (Danneel et al., 2018; Lay et al., 2019). Such differential reactivity to being alone may form part of the social-cognitive mechanism which links individual differences in neuroticism to individual differences in the level and intrapersonal variability of state loneliness.

One consequence of hyperreactivity to social stressors is increased emotional volatility (Bolger & Schilling, 1991; Costa & McCrae, 1992; Eysenck & Eysenck, 1985; Hills & Argyle, 2001; Hisler et al., 2020). Past empirical findings have linked neuroticism to within-person variability in affect more generally (Eid & Diener, 1999; Hisler et al., 2020; Mader et al., 2023).Consistently, our study shows that more neurotic individuals also experienced greater within-person variability in loneliness. A recent study suggested that such associations between personality traits and within-person variability in affect may be due to confounding of variability with the mean level of (negative) affect (Kalokerinos et al., 2020). Our analytic approach avoids this issue by concurrently modelling influences on mean level and within-person variance. Incidentally, Mader et al. (2023) used simulations to show that the data of Kalokerinos et al. (2020) are best explained by a model which is nearly identical to ours. In particular, like us, Mader et al. (2023) use censoring to account for the possibility of emotional states outside the range of the response scale. After re-analysing the original data, they conclude that ‘more neurotic individuals indeed experience more variability in negative emotion in everyday life’. Our results provide initial evidence that this is also true for loneliness, although further replications, including in other populations, would be welcome.

Extraversion and restrictions on social contact

Next to neuroticism, a recent meta-analysis found that, among the Big Five traits, extraversion was most strongly associated with loneliness (Buecker et al., 2020). In contrast, we found no such association when controlling for the other Big Five traits. However, several recent studies found that highly extraverted individuals experienced the greatest increase in loneliness when restrictions on social contact were put in place (Alt et al., 2021; Entringer & Gosling, 2022). Highly extraverted individuals may have responded more negatively to such restrictions because they have a greater need for social contact. Since our data were collected under similar lockdown conditions, which may have increased the overall level of loneliness (Ernst et al., 2022), these may have diminished the link between extraversion and loneliness.

Intriguingly, we found that the link between extraversion and loneliness was moderated by social context. Plotting this interaction shows that when they were alone, participants felt lonelier the more extraverted they were; conversely, when they had company, participants felt less lonely the more extraverted they were (Figure 1(a)). This pattern is consistent with the suggestion that extraversion may be positively associated with loneliness when individuals are restricted in their opportunities for social contact. However, marginal effects in each social context showed that neither of these associations was statistically significant. Future research may use more reliable measures of loneliness and more detailed measures of social context to further investigate whether there indeed exists a context-dependent link between extraversion and loneliness, and in particular, whether more extraverted individuals indeed experience greater loneliness than less extraverted individuals when they are alone.

Despite the restrictions put on social contact, it is still possible that individuals differed in the frequency of social interactions. We therefore tested whether personality predicted the proportion of situations individuals experienced alone versus in company. For this purpose, we regressed the social context on the five personality traits in a generalised linear mixed model. Indeed, individuals who were more extraverted, B = .09, 95% CI = [.04, .14], and less open, B = −.16, 95% CI = [−.24, −.07], reported being in the company of others more often (for full results, see Table 2). Thus, even under restrictions on social contact, more extraverted individuals appear to have been able to seek out, or otherwise experienced, more frequent social interactions. This provides some evidence that person–situation transactions may partially explain associations between personality traits and loneliness.

Generalisability

While most prior literature on personality and loneliness focuses on American and European samples, we collected a large experience sampling dataset in Nepal. Yet, our results generally replicate prior findings about the association between Big Five traits and mean levels of loneliness. The zero-order correlations in our study are within the prediction intervals obtained from a recent meta-analysis (Buecker et al., 2020). Like Buecker et al. (2020), we find partial correlations to be somewhat weaker. However, our results only diverge from their meta-analytic estimates in the case of extraversion, for which we find a non-significant negative partial correlation. This may be partly due to restrictions on social contact (Entringer & Gosling, 2022). Overall, we thus find quite similar associations between personality and mean levels of loneliness among Nepali participants as those previously reported in WEIRD samples. Even beyond these mean-level associations, our findings concord with the differential reactivity hypothesis and the conceptualisation of neuroticism as hyperreactivity to social stressors. Our study thus provides a sanguine perspective on the generalisability of the nomological net of broad personality models such as the Big Five beyond the usual WEIRD samples.

Limitations

We used experience sampling methods to ask participants several times a day about their feelings of loneliness. This allowed us to study intrapersonal fluctuations in loneliness. A downside of experience sampling is the need for brief measures. In this case, we used a face-valid single-item measure of loneliness. When analysed ‘as is’ (i.e. without censoring), this item has an ICC = .53, which is in line with other experience sampling studies of loneliness (Harper et al., 2020; Rinderknecht, 2020; Tam & Chan, 2019). Unfortunately, with a single-item measure, it is not feasible to establish the specificity, that is, how reliably the measure captures occasion-specific effects. If anything, however, an unreliable measure would work against our models and, in particular, should not be expected to yield the theoretically plausible associations we identify. Still, future research may replicate our analyses using more reliable multi-item measures of loneliness.

Collecting intensive longitudinal data also allowed us to explore whether associations between personality and loneliness were context-dependent. Using a simple indicator of whether participants were alone or in the company of others, we find that social context moderated the links between Big Five traits and both the level of loneliness and the degree to which individuals experienced ups and downs of loneliness. We must acknowledge, however, that our measure of social context was rather crude. Past research based on correlations between trait loneliness and measures of the quantity and quality of social connections has suggested that what matters is less the number of one’s contacts than the depths of one’s relationships with them (Perlman & Peplau, 1981; Russell et al., 2012; van Roekel et al., 2015). Most social contacts reported by our participants involved close family members. It may be that these contacts were overall positively valenced, and that this accounts for the relationship between social context and loneliness. However, further research may use more sophisticated measures of social context, including measures not only of the relationship with the other person(s), but also of the situation itself (e.g. DIAMONDS, Rauthmann et al., 2014; SIS, Gerpott et al., 2018).

A related limitation is that we only accounted for whether others were physically present. However, participants may also have been in contact with others via phone or internet. Correlational studies have consistently linked the use of (computer-)mediated communication to more, rather than less, loneliness, including during the COVID-19 pandemic (Bonsaksen et al., 2021; Kashian, 2022; Sarmiento et al., 2020). However, these broad associations may reflect that lonelier individuals turn to mediated communication in the absence of alternative means of contact (Boursier et al., 2020; Cauberghe et al., 2021). Indeed, some studies suggest that social uses of mediated communication during the COVID-19 pandemic were associated with lower loneliness (Choi & Choung, 2021; Meier et al., 2021). Future experience sampling studies or within-person experiments may explore whether communicating by phone or internet may buffer against loneliness by comparing such instances against situations in which individuals did not communicate at all.

Outlook: the importance of intrapersonal variability

A large body of research has documented links between trait loneliness and mental and physical well-being (Holt-Lunstad et al., 2015). A key finding from our study is that the individuals vary not only in average levels of loneliness but also in the extent to which they experience ups and downs of loneliness. This raises the question whether individual differences in intrapersonal variability in loneliness have independent associations with well-being. A recent study found that social contact alleviated the negative effects of loneliness on mental well-being (Stavrova & Ren, 2023). This suggests that occasional meaningful social connections, even if they are interspersed with periods of intense loneliness, could buffer against the harmful effects of loneliness and mitigate the negative association between neuroticism and mental well-being.

Our findings may also inspire new interventions. Drawing on the distinction between objective social isolation and subjective feelings of loneliness, researchers and practitioners have recognised that encouraging social contact alone is insufficient to alleviate loneliness (Hickin et al., 2021; Käll et al., 2020). Interventions based on cognitive behavioural therapy target perceptual and cognitive biases that result in hypervigilance to negative social information (Cacioppo & Hawkley, 2009; Hickin et al., 2021). This hypervigilance and hyperreactivity is seen as central to neuroticism (Bolger & Schilling, 1991; Costa & McCrae, 1992; Eysenck & Eysenck, 1985; Hills & Argyle, 2001; Hisler et al., 2020), and may underlie the greater intrapersonal variability in loneliness associated with heightened neuroticism. However, our results also show that merely being in the company of others was associated with lower loneliness among even the most neurotic part of our sample. Future research may inform interventions by further elucidating the personality-dependent triggers within situations spent alone or with others which give rise to feelings of loneliness.

Conclusion

Most people feel lonely from time to time, though some people feel lonelier on average than others. Beyond such differences in average loneliness, however, people also differ in the degree to which they experience ups and downs of loneliness. Using experience sampling data, we show that more neurotic individuals experience a greater level of loneliness. This link was strongest when participants were alone rather than in company. Moreover, in line with the differential reactivity hypothesis, more neurotic individuals also experienced greater fluctuations in loneliness from situation to situation. We also find some evidence for a context-specific association between extraversion and loneliness. Thus, the link between personality and loneliness is not merely a matter of averages but also of context-dependent intrapersonal dynamics.

Supplemental Material

Supplemental Material - Big five traits predict between- and within-person variation in loneliness

Supplemental Material for Big five traits predict between- and within-person variation in loneliness by Sujan Shrestha, Kripa Sigdel, Madhusudan Pokharel and Simon Columbus in European Journal of Personality

Footnotes

Acknowledgements

S.S. and S.C. would like to thank Author AID () for facilitating this collaboration.

Author’s note

Sujan Shrestha. Part of the work was completed when Mr. Shrestha was at St. Xavier’s College, Maitighar.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

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

Open science statement

This study was not preregistered. Data, analysis code, and a list of all measures are available on the OSF ().

ORCID iDs

Sujan Shrestha

Kripa Sigdel

Madhusudan Pokharel

Simon Columbus

Supplemental Material

Supplemental material for this article is available online.

Notes

References

Alt

Reim

Walper

(2021). Fall from grace: Increased loneliness and depressiveness among extraverted youth during the German COVID‐19 lockdown. Journal of Research on Adolescence, 31(3), 678–691. https://doi.org/10.1111/jora.12648

Baumert

Schmitt

Perugini

Johnson

Blum

Borkenau

Costantini

Denissen

J. J. A.

Fleeson

Grafton

Jayawickreme

Kurzius

MacLeod

Miller

L. C.

Read

S. J.

Roberts

Robinson

M. D.

Wood

Wrzus

(2017). Integrating personality structure, personality process, and personality development. European Journal of Personality, 31(5), 503–528. https://doi.org/10.1002/per.2115

Beller

Wagner

(2018). Disentangling loneliness: Differential effects of subjective loneliness, network quality, network size, and living alone on physical, mental, and cognitive health. Journal of Aging and Health, 30(4), 521–539. https://doi.org/10.1177/0898264316685843

Bolger

Schilling

E. A.

(1991). Personality and the problems of everyday life: The role of neuroticism in exposure and reactivity to daily stressors. Journal of Personality, 59(3), 355–386. https://doi.org/10.1111/j.1467-6494.1991.tb00253.x

Bonsaksen

Ruffolo

Leung

Price

Thygesen

Schoultz

Geirdal

A. Ø.

(2021). Loneliness and its association with social media use during the COVID-19 Outbreak. Social Media + Society , 7(3), Article 205630512110338. https://doi.org/10.1177/20563051211033821

Boursier

Gioia

Musetti

Schimmenti

(2020). Facing loneliness and anxiety during the COVID-19 isolation: The role of excessive social media use in a sample of Italian adults. Frontiers in Psychiatry, 11(3), 586222. https://doi.org/10.3389/fpsyt.2020.586222

Buecker

Horstmann

K. T.

Luhmann

(2023). Lonely today, lonely tomorrow: Temporal dynamics of loneliness in everyday life and its associations with psychopathological symptoms. Social Psychological and Personality Science, 15(2), 170–181. https://doi.org/10.1177/19485506231156061

Buecker

Maes

Denissen

J. J. A.

Luhmann

(2020). Loneliness and the Big five personality traits: A meta-analysis. European Journal of Personality, 34(1), 8–28. https://doi.org/10.1002/per.2229

Bürkner

P. C.

(2017). Advanced Bayesian multilevel modeling with the R package brms. arXiv. https://doi.org/10.48550/arXiv.1705.11123

10.

Bürkner

P. C.

(2017b). brms: An R package for Bayesian multilevel models using Stan. Journal of Statistical Software, 80(2), 1–28. https://doi.org/10.18637/jss.v080.i01

11.

Cacioppo

J. T.

Fowler

J. H.

Christakis

N. A.

(2009). Alone in the crowd: The structure and spread of loneliness in a large social network. Journal of Personality and Social Psychology, 97(6), 977–991. https://doi.org/10.1037/a0016076

12.

Cacioppo

J. T.

Hawkley

L. C.

(2009). Perceived social isolation and cognition. Trends in Cognitive Sciences, 13(10), 447–454. https://doi.org/10.1016/j.tics.2009.06.005

13.

Cacioppo

J. T.

Hawkley

L. C.

Berntson

G. G.

(2003). The anatomy of loneliness. Current Directions in Psychological Science, 12(3), 71–74. https://doi.org/10.1111/1467-8721.01232

14.

Cauberghe

Van Wesenbeeck

De Jans

Hudders

Ponnet

(2021). How adolescents use social media to cope with feelings of loneliness and anxiety during COVID-19 lockdown. Cyberpsychology, Behavior, and Social Networking, 24(4), 250–257. https://doi.org/10.1089/cyber.2020.0478

15.

Ching

C. M.

Church

A. T.

Katigbak

M. S.

Reyes

J. A. S.

Tanaka-Matsumi

Takaoka

Zhang

Shen

Arias

R. M.

Rincon

B. C.

Ortiz

F. A.

(2014). The manifestation of traits in everyday behavior and affect: A five-culture study. Journal of Research in Personality, 48(1), 1–16. https://doi.org/10.1016/j.jrp.2013.10.002

16.

Choi

Choung

(2021). Mediated communication matters during the COVID-19 pandemic: The use of interpersonal and masspersonal media and psychological well-being. Journal of Social and Personal Relationships, 38(8), 2397–2418. https://doi.org/10.1177/02654075211029378

17.

Costa

P. T.

McCrae

R. R.

(1992). Normal personality assessment in clinical practice: The NEO personality inventory. Psychological Assessment, 4(1), 5–13. https://doi.org/10.1037/1040-3590.4.1.5

18.

Coyle

C. E.

Dugan

(2012). Social isolation, loneliness and health among older adults. Journal of Aging and Health, 24(8), 1346–1363. https://doi.org/10.1177/0898264312460275

19.

Danneel

Maes

Vanhalst

Bijttebier

Goossens

(2018). Developmental change in loneliness and attitudes toward aloneness in adolescence. Journal of Youth and Adolescence, 47(1), 148–161. https://doi.org/10.1007/s10964-017-0685-5

20.

Eid

Diener

(1999). Intraindividual variability in affect: Reliability, validity, and personality correlates. Journal of Personality and Social Psychology, 76(4), 662–676. https://doi.org/10.1037/0022-3514.76.4.662

21.

Entringer

T. M.

Gosling

S. D.

(2022). Loneliness during a nationwide lockdown and the moderating effect of extroversion. Social Psychological and Personality Science, 13(3), 769–780. https://doi.org/10.1177/19485506211037871

22.

Epley

Schroeder

(2014). Mistakenly seeking solitude. Journal of Experimental Psychology: General, 143(5), 1980–1999. https://doi.org/10.1037/a0037323

23.

Ernst

Niederer

Werner

A. M.

Czaja

S. J.

Mikton

Ong

A. D.

Rosen

Brähler

Beutel

M. E.

(2022). Loneliness before and during the COVID-19 pandemic: A systematic review with meta-analysis. American Psychologist, 77(5), 660–677. https://doi.org/10.1037/amp0001005

24.

Eysenck

H. J.

Eysenck

M. W.

(1985). Personality and individual differences. Plenum.

25.

Fleeson

(2001). Toward a structure- and process-integrated view of personality: Traits as density distributions of states. Journal of Personality and Social Psychology, 80(6), 1011–1027. https://doi.org/10.1037/0022-3514.80.6.1011

26.

Fleeson

Jayawickreme

(2015). Whole trait theory. Journal of Research in Personality, 56, 82–92. https://doi.org/10.1016/j.jrp.2014.10.009

27.

Funder

D. C.

(2008). Persons, situations, and person-situation interactions. In John

O. P.

Robins

R. W.

Pervin

L. A.

(Eds.), Handbook of personality: Theory and research (pp. 568–580). The Guilford Press.

28.

Fung

Paterson

Alden

L. E.

(2017). Are social anxiety and loneliness best conceptualized as a unitary trait? Journal of Social and Clinical Psychology, 36(4), 335–345. https://doi.org/10.1521/jscp.2017.36.4.335

29.

Gelman

Hill

Yajima

(2012). Why we (usually) don't have to worry about multiple comparisons. Journal of Research on Educational Effectiveness, 5(2), 189–211. https://doi.org/10.1080/19345747.2011.618213

30.

Gerpott

F. H.

Balliet

Columbus

Molho

de Vries

R. E.

(2018). How do people think about interdependence? A multidimensional model of subjective outcome interdependence. Journal of Personality and Social Psychology, 115(4), 716–742. https://doi.org/10.1037/pspp0000166

31.

Geukes

Nestler

Hutteman

Küfner

A. C. P.

Back

M. D.

(2017). Trait personality and state variability: Predicting individual differences in within- and cross-context fluctuations in affect, self-evaluations, and behavior in everyday life. Journal of Research in Personality, 69(3), 124–138. https://doi.org/10.1016/j.jrp.2016.06.003

32.

Goldberg

L. R.

(1993). The structure of phenotypic personality traits. American Psychologist, 48(1), 26–34. https://doi.org/10.1037/0003-066X.48.1.26

33.

Harper

K. L.

Eddington

K. M.

Silvia

P. J.

(2020). Perfectionism and loneliness: The role of expectations and social hopelessness in daily life. Journal of Social and Clinical Psychology, 39(2), 117–139. https://doi.org/10.1521/jscp.2020.39.02.117

34.

Hawkley

L. C.

Burleson

M. H.

Berntson

G. G.

Cacioppo

J. T.

(2003). Loneliness in everyday life: Cardiovascular activity, psychosocial context, and health behaviors. Journal of Personality and Social Psychology, 85(1), 105–120. https://doi.org/10.1037/0022-3514.85.1.105

35.

Hawkley

L. C.

Cacioppo

J. T.

(2007). Aging and loneliness: Downhill quickly? Current Directions in Psychological Science, 16(4), 187–191. https://doi.org/10.1111/j.1467-8721.2007.00501.x

36.

Hawkley

L. C.

Cacioppo

J. T.

(2010). Loneliness matters: A theoretical and empirical review of consequences and mechanisms. Annals of Behavioral Medicine, 40(2), 218–227. https://doi.org/10.1007/s12160-010-9210-8.

37.

Hedeker

Mermelstein

R. J.

Demirtas

(2008). An application of a mixed-effects location scale model for analysis of ecological momentary assessment (EMA) data. Biometrics, 64(2), 627–634. https://doi.org/10.1111/j.1541-0420.2007.00924.x

38.

Henrich

Heine

S. J.

Norenzayan

(2010). The weirdest people in the world? The Behavioral and brain sciences, 33(2–3), 61–135. https://doi.org/10.1017/S0140525X0999152X

39.

Hickin

Käll

Shafran

Sutcliffe

Manzotti

Langan

(2021). The effectiveness of psychological interventions for loneliness: A systematic review and meta-analysis. Clinical Psychology Review, 88(3), 102066. https://doi.org/10.1016/j.cpr.2021.102066

40.

Hills

Argyle

(2001). Emotional stability as a major dimension of happiness. Personality and Individual Differences, 31(8), 1357–1364. https://doi.org/10.1016/S0191-8869(00)00229-4

41.

Hisler

G. C.

Krizan

DeHart

Wright

A. G.

(2020). Neuroticism as the intensity, reactivity, and variability in day-to-day affect. Journal of Research in Personality, 87(3), 103964. https://doi.org/10.1016/j.jrp.2020.103964

42.

Holt-Lunstad

Smith

T. B.

Baker

Harris

Stephenson

(2015). Loneliness and social isolation as risk factors for mortality: A meta-analytic review. Perspectives on Psychological Science, 10(2), 227–237. https://doi.org/10.1177/1745691614568352

43.

Horstmann

K. T.

Ziegler

(2020). Assessing personality states: What to consider when constructing personality state measures. European Journal of Personality, 34(6), 1037–1059. https://doi.org/10.1002/per.2266

44.

Käll

Shafran

Lindegaard

Bennett

Cooper

Coughtrey

Andersson

(2020). A common elements approach to the development of a modular cognitive behavioral theory for chronic loneliness. Journal of Consulting and Clinical Psychology, 88(3), 269–282. https://doi.org/10.1037/ccp0000454

45.

Kalokerinos

E. K.

Murphy

S. C.

Koval

Bailen

N. H.

Crombez

Hollenstein

Gleeson

Thompson

R. J.

Van Ryckeghem

D. M. L.

Kuppens

Bastian

(2020). Neuroticism may not reflect emotional variability. Proceedings of the National Academy of Sciences of the United States of America, 117(17), 9270–9276. https://doi.org/10.1073/pnas.1919934117

46.

Kashian

(2022). Can mediated communication moderate the relationship between being in isolation and quarantine for COVID-19 and loneliness? Cyberpsychology: Journal of Psychosocial Research on Cyberspace , 16(4). Article 6. https://doi.org/10.5817/CP2022-4-6

47.

Kubinec

(2022). Ordered beta regression: A parsimonious, well-fitting model for continuous data with lower and upper bounds. Political Analysis, 31(4), 519–536. https://doi.org/10.1017/pan.2022.20

48.

Kuppens

Van Mechelen

Nezlek

J. B.

Dossche

Timmermans

(2007). Individual differences in core affect variability and their relationship to personality and psychological adjustment. Emotion, 7(2), 262–274. https://doi.org/10.1037/1528-3542.7.2.262

49.

Larson

R. W.

(1990). The solitary side of life: An examination of the time people spend alone from childhood to old age. Developmental Review, 10(2), 155–183. https://doi.org/10.1016/0273-2297(90)90008-R

50.

Lay

J. C.

Pauly

Graf

Biesanz

J. C.

Hoppmann

C. A.

(2019). By myself and liking it? Predictors of distinct types of solitude experiences in daily life. Journal of Personality, 87(3), 633–647. https://doi.org/10.1111/jopy.12421

51.

Liu

Kong

(2015). zoib: An R package for Bayesian Inference for beta regression and zero/one inflated beta regression. R Journal, 7(2), 34–51. https://doi.org/10.32614/RJ-2015-019

52.

Mader

Arslan

R. C.

Schmukle

S. C.

Rohrer

J. M.

(2023). Emotional (in) stability: Neuroticism is associated with increased variability in negative emotion after all. Proceedings of the National Academy of Sciences of the United States of America , 120(23), Article e2212154120. https://doi.org/10.1073/pnas.2212154120

53.

Makowski

Ben-Shachar

M. S.

Lüdecke

(2019). bayestestR: Describing effects and their Uncertainty, existence and significance within the Bayesian framework. Journal of Open Source Software, 4(40), 1541. https://doi.org/10.21105/joss.01541

54.

Meier

J. V.

Noel

J. A.

Kaspar

(2021). Alone together: Computer-mediated communication in leisure time during and after the COVID-19 pandemic. Frontiers in Psychology, 12(3), 666655. https://doi.org/10.3389/fpsyg.2021.666655

55.

Mõttus

Kandler

Bleidorn

Riemann

McCrae

R. R.

(2017). Personality traits below facets: The consensual validity, longitudinal stability, heritability, and utility of personality nuances. Journal of Personality and Social Psychology, 112(3), 474–490. https://doi.org/10.1037/pspp0000100

56.

Mund

Freuding

M. M.

Möbius

Horn

Neyer

F. J.

(2020). The stability and change of loneliness across the life span: A meta-analysis of longitudinal studies. Personality and Social Psychology Review, 24(1), 24–52. https://doi.org/10.1177/1088868319850738

57.

Rast

Hofer

S. M.

Sparks

(2012). Modeling individual differences in within-person variation of negative and positive affect in a mixed effects location scale model using BUGS/JAGS. Multivariate Behavioral Research, 47(2), 177–200. https://doi.org/10.1080/00273171.2012.658328

58.

Rauthmann

J. F.

Gallardo-Pujol

Guillaume

E. M.

Todd

Nave

C. S.

Sherman

R. A.

Ziegler

Jones

A. B.

Funder

D. C.

(2014). The situational Eight DIAMONDS: A taxonomy of major dimensions of situation characteristics. Journal of Personality and Social Psychology, 107(4), 677–718. https://doi.org/10.1037/a0037250

59.

Perlman

Peplau

L. A.

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

Gilmour

(Eds.), Personal Relationships in Disorder (pp. 31–56). London, UK: Academic Press.

60.

R Core Team . (2022). R: A language and Environment for statistical computing. R Foundation for Statistical Computing.

61.

Reissmann

Stollberg

Hauser

Kaunzinger

Lange

K. W.

(2021). The role of state feelings of loneliness in the situational regulation of social affiliative behavior: Exploring the regulatory relations within a multilevel framework. PLoS One, 16(6), 1–21. https://doi.org/10.1371/journal.pone.0252775

62.

Revelle

(2022). psych: Procedures for personality and psychological research (Version 2.2.3). https://cran.r-project.org/package=psych

63.

Rinderknecht

R. G.

(2020). Factors Contributing to the experience of state loneliness: Doctoral dissertation. University of Maryland. https://drum.lib.umd.edu/handle/1903/26453

64.

Rinderknecht

R. G.

Doan

Sayer

L. C.

(2021). Loneliness Loves company, some more than others: Social ties, form of Engagement, and their relation to loneliness. Social Problems. spab049. https://doi.org/10.1093/socpro/spab049

65.

Rohrer

J. M.

Lucas

R. E.

(2020). Causal effects of well-being on health: It’s Complicated . PsyArXiv. https://doi.org/10.31234/osf.io/wgbe4

66.

Rosseel

(2012). Lavaan: An R Package for Structural Equation modeling. Journal of Statistical Software, 48(2), 1–36. https://doi.org/10.18637/jss.v048.i02

67.

Rotenberg

K. J.

(1994). Loneliness and interpersonal trust. Journal of Social and Clinical Psychology, 13(2), 152–173. https://doi.org/10.1521/jscp.1994.13.2.152

68.

Russell

D. W.

Cutrona

C. E.

McRae

Gomez

(2012). Is loneliness the same as being alone? Journal of Psychology, 146(1–2), 7–22. https://doi.org/10.1080/00223980.2011.589414

69.

Sandstrom

G. M.

Dunn

E. W.

(2014). Social interactions and well-being: The surprising power of weak ties. Personality and Social Psychology Bulletin, 40(7), 910–922. https://doi.org/10.1177/0146167214529799

70.

Sarmiento

I. G.

Olson

Yeo

Chen

Y. A.

Toma

C. L.

Brown

B. B.

Bellmore

Mares

M. L.

(2020). How does social media use relate to adolescents’ internalizing symptoms? Conclusions from a systematic narrative review. Adolescent Research Review, 5(4), 381–404. https://doi.org/10.1007/s40894-018-0095-2

71.

Sharma

Banstola

Parajuli

R. R.

(2021). Assessment of COVID-19 pandemic in Nepal: A lockdown scenario analysis. Frontiers in Public Health , 9(8), 599280. Article 599280. https://doi.org/10.3389/fpubh.2021.599280

72.

Stan Development Team . (2022). Stan modeling language Users Guide and reference manual. Version 2.31. https://mc-stan.org

73.

Stavrova

Ren

(2023). Alone in a crowd: Is social contact associated with less psychological pain of loneliness in everyday life? Journal of Happiness Studies, 24(5), 1841–1860. https://doi.org/10.1007/s10902-023-00661-3

74.

Tam

K. Y. Y.

Chan

C. S.

(2019). The effects of lack of meaning on trait and state loneliness: Correlational and experience-sampling evidence. Personality and Individual Differences, 141, 76–80. https://doi.org/10.1016/j.paid.2018.12.023

75.

Thielmann

Spadaro

Balliet

(2020). Personality and prosocial behavior: A theoretical framework and meta-analysis. Psychological Bulletin, 146(1), 30–90. https://doi.org/10.1037/bul0000217

76.

Tiwari

S. C.

(2013). Loneliness: A disease? Indian Journal of Psychiatry, 55(4), 320–322. https://doi.org/10.4103/0019-5545.120536

77.

VanderWeele

T. J.

Hawkley

L. C.

Cacioppo

J. T.

(2012). On the reciprocal association between loneliness and subjective well-being. American Journal of Epidemiology, 176(9), 777–784. https://doi.org/10.1093/aje/kws173

78.

van Roekel

Goossens

Verhagen

Wouters

Engels

R. C.

Scholte

R. H.

(2014). Loneliness, affect, and adolescents' appraisals of company: An experience sampling method study. Journal of Research on Adolescence, 24(2), 350–363. https://doi.org/10.1111/jora.12061

79.

van Roekel

Scholte

R. H. J.

Engels

R. C. M. E.

Goossens

Verhagen

(2015). Loneliness in the daily lives of adolescents: An experience sampling study examining the effects of social contexts. The Journal of Early Adolescence, 35(7), 905–930. https://doi.org/10.1177/0272431614547049

80.

van Roekel

Verhagen

Engels

R. C. M. E.

Scholte

R. H. J.

Cacioppo

J. T.

(2018). Trait and state levels of loneliness in early and late adolescents: Examining the differential reactivity hypothesis. Journal of Clinical Child and Adolescent Psychology, 47(6), 888–899. https://doi.org/10.1080/15374416.2016.1146993

81.

van Winkel

Wichers

Collip

Jacobs

Derom

Thiery

Myin-Germeys

Peeters

(2017). Unraveling the role of loneliness in depression: The relationship between daily life experience and behavior. Psychiatry, 80(2), 104–117. https://doi.org/10.1080/00332747.2016.1256143

82.

Vuorre

(2019). How to Analyze Visual analog (slider) scale data? Sometimes I R. https://sometimesir.com/posts/2019-02-18-analyze-analog-scale-ratings-with-zero-one-inflated-beta-models

83.

Wei

Russell

D. W.

Zakalik

R. A.

(2005). Adult attachment, social self-efficacy, self-disclosure, loneliness, and subsequent depression for freshman college students: A longitudinal study. Journal of Counseling Psychology, 52(4), 602–614. https://doi.org/10.1037/0022-0167.52.4.602

84.

Wickham

Averick

Bryan

Chang

McGowan

L. D. A.

François

Grolemund

Hayes

Henry

Hester

Kuhn

Pedersen

Miller

Bache

Müller

Ooms

Robinson

Seidel

Spinu

Yutani

(2019). Welcome to the tidyverse. Journal of Open Source Software, 4(43), 1686. Article 1686. https://doi.org/10.21105/joss.01686

85.

Wieczorek

L. L.

Humberg

Gerstorf

Wagner

(2021). Understanding loneliness in adolescence: A test of competing hypotheses on the interplay of extraversion and neuroticism. International Journal of Environmental Research and Public Health, 18(23), 12412. Article 12412. https://doi.org/10.3390/ijerph182312412

86.

World Bank . (n.d.). Nepal . The World Bank. Retrieved from https://data.worldbank.org/country/NP

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.83 MB

Big Five traits predict between- and within-person variation in loneliness

Abstract

Plain language summary

Keywords

Introduction

Loneliness as a trait and state

Individual and contextual differences in loneliness

Personality predictors of loneliness

Contextual predictors of loneliness

Person X situation interactions

Personality and within-person variability in loneliness

The present study

Methods

Ethics and open scientific practices

Sample

Historical and local setting

Measures

Procedures

Analytic approach

Mixed-effects location scale models

Zero inflation

Model specification

Inference criteria

Software

Results

Descriptive results

Zero-order and partial correlations

Distribution and within-person variability of loneliness

Inferential results

Personality and social context predict the level of loneliness

Neuroticism predicts within-person variability in loneliness

Account for personality reduces the association between level and variability of loneliness

Discussion

Neuroticism and differential reactivity

Extraversion and restrictions on social contact

Generalisability

Limitations

Outlook: the importance of intrapersonal variability

Conclusion

Supplemental Material

Supplemental Material - Big five traits predict between- and within-person variation in loneliness

Footnotes

Acknowledgements

Author’s note

Declaration of conflicting interests

Funding

Open science statement

ORCID iDs

Supplemental Material

Notes

References

Supplementary Material