Looking beyond time alone: An examination of solitary activities in emerging adulthood

Abstract

Introduction

Solitude represents an important context for emerging adults' well‐being; but to date, little is known about how emerging adults spend their time alone. The goals of this study were to: (1) describe and characterize solitary activities among emerging adults attending university; (2) examine links between solitary activities and indices of adjustment; and (3) explore the moderating role of affinity for solitude in these associations.

Methods

Participants were N = 1798 university students aged 18–25 years (M _age = 19.73, SD = 1.46; 59.7% female) who completed assessments of how/why they spend time alone and indices of psychosocial adjustment (e.g., well‐being, psychological distress, loneliness, and aloneliness).

Results

Emerging adults who spent time alone predominantly thinking reported poor adjustment outcomes (i.e., higher loneliness and psychological distress, and lower well‐being) and dissatisfaction with solitude, whereas those who engaged in active leisure activities or passive technology use while alone reported lower psychological distress and higher satisfaction with solitude. The negative implications of doing nothing were not attenuated at higher levels of affinity for solitude.

Discussion

These findings suggest that some solitary activities are more beneficial than others.

1. INTRODUCTION

Solitude is a ubiquitous experience, but it has been challenging to assess the complex implications of spending time alone, and findings remain mixed (Coplan, Bowker, & Nelson, 2021). Previous research has focused primarily on negative aspects of solitude, with many studies demonstrating that spending too much time alone is costly throughout development (Davies, 1996; Pietrabissa & Simpson, 2020; Shankar et al., 2011). However, spending time in solitude is not inherently negative (Borg & Willoughby, 2022). Indeed, studies have connected solitude to a variety of benefits, including increased competence, creativity, spirituality, and personal growth (Thomas & Azmitia, 2019; Weinstein et al., 2021). To gain a clearer understanding of how solitude is experienced, it may be important to look beyond time spent alone to consider factors related to the quality of solitude, including what individuals do while alone and their motivations for solitude (Hipson et al., 2021; Nguyen et al., 2018; Ost Mor et al., 2021). Still, current scholarship provides limited guidance regarding how these individual and contextual factors work together to influence the experience and concomitants of solitude.

Emerging adulthood (ages 18–25 years) is an important and unique developmental period for considering experiences of solitude. For example, solitude may create a space in which emerging adults can meet important developmental tasks, such as exercising their autonomy (including how, and with whom, to spend their free time) and developing a strong sense of identity (Arnett, 2014; Nelson & Millet, 2021). Solitude may also afford emerging adults increased opportunities for rest and relaxation (Nguyen et al., 2018). There are some indications that affinity for solitude (i.e., preference for solitude due to enjoyment) increases during the emerging adult years (Thomas & Azmitia, 2019). As such, emerging adults may be in a unique position wherein they have both the freedom and desire to spend increased time alone. Limited available research shows that seeking solitude for positive reasons predicts better outcomes, such as increased relaxation and self‐esteem (Borg & Willoughby, 2022; Nguyen et al., 2018).

Contemporary evidence also suggests that, particularly for individuals who tend to find solitude enjoyable and productive, not spending enough time alone can lead to negative feelings and dissatisfaction with one's solitary time (i.e., increased aloneliness) (Coplan, Hipson, et al., 2019; Coplan, Hipson, & Bowker, 2021). However, there is at least some preliminary evidence to suggest that both quantity and quality of time alone are important in satisfying one's need for solitude. Coplan, Hipson, and Bowker (2021) found that among adolescents who spent comparatively more time alone, lower aloneliness (i.e., greater satisfaction with their time alone) was only found among those who reported solitary activities that seemed to be more meaningful and engaging. Results from other studies also suggest that some solitary activities may be more beneficial than others (Hipson et al., 2021; Thomas et al., 2021). Thus, both time in solitude and the nature of solitary activities may be important with respect to implications for well‐being. Yet, very little is currently known about what emerging adults do while alone. Moreover, we do not know how the implications of different solitary activities for well‐being might vary as a function of individual differences in how much one enjoys spending time alone (i.e., affinity for solitude).

Given limited knowledge about what emerging adults do when they are alone, this study adopted a person‐centered approach to explore whether there are distinct groups of emerging adults characterized by tendencies to engage in different types of solitary activities. We also examined whether solitary activities differentially predict satisfaction with time alone (i.e., aloneliness) and indices of psychosocial adjustment, and the role of affinity for solitude. This study helps clarify mixed findings regarding the implications of solitude and provides a more concrete understanding regarding the specific aspects of solitude that best predict benefits.

1.1. Solitary activities

To date, little research attention has been paid to the different conditions under which solitude may impact adjustment. Yet, experiences of solitude may be influenced (for better or worse) by what people do while alone. For instance, Wilson et al. (2014) found that emerging adults preferred engaging in solitary activities over sitting alone with their thoughts. Moreover, the idea of being in “pure” solitude (i.e., alone with one's thoughts) was so unappealing that many participants chose instead to self‐administer a painful electric shock. This finding has been extended across 11 countries (Buttrick et al., 2019). Although some research suggests that using technology interrupts the benefits of seeking solitude (Diefenbach & Borrmann, 2019), Thomas et al. (2021) also found that pure solitude was linked to more boredom and loneliness in college students as compared to being alone while on a mobile device.

Hipson et al. (2021) asked adolescents to list the three activities they did most when alone over the last week. Subgroups of adolescents were then created based on their characteristic solitary activities. Most adolescents (53%) reported being passively engaged with technology (e.g., watching TV, scrolling on social media) when alone. However, many (32%) also described more constructive solitary activities, including homework, hobbies, listening to music, and active technology use (e.g., texting and video games). In contrast, fewer adolescents (15%) reported spending their alone time in pure solitude (e.g., daydreaming, planning, and negative thinking). Adolescents who tended to spend their time alone in pure solitude (i.e., doing nothing or thinking) reported greater depression, anxiety, and loneliness than their peers who tended to spend their alone time engaged in other solitary activities. Of note, no differences in indices of adjustment were found between adolescents who spent time alone primarily passively engaged with technology and those who reported participating in more constructive activities. These results were consistent with previous findings indicating that doing something makes us feel better than “just thinking” when alone (Lay et al., 2018; Wilson et al., 2014).

Taken together, there are differences in how young people spend time in solitude, and when it comes to spending time alone, doing something is better than doing nothing. Compared to adolescents, however, emerging adults typically have more control over how they spend their time (Wray‐Lake et al., 2010). Accordingly, the content and implications of solitary activities may change from adolescence to emerging adulthood, particularly in the unique context of university. For example, substance use increases exponentially upon entering university (Sussman & Arnett, 2014). Although university students are most likely to engage in substance use in social contexts, if emerging adults routinely drink or do drugs while alone, the implications of “doing something” may look different during these years (Christiansen et al., 2002).

Moreover, during the transition to emerging adulthood, many young people move away from home for the first time when attending university, where they take on adult‐like responsibilities and learn to balance new academic and social demands. This may have implications for how solitude is experienced. For example, Hatano et al. (2022) found that university students underestimated how enjoyable spending time alone “just thinking” would be. Nguyen et al. (2022) also recently found that although emerging adults were much more likely to choose to spend time in solitude sorting pencils rather than sitting with their thoughts, their enjoyment for each task did not differ.

1.2. Affinity for solitude

Whereas solitary activities reflect the what of solitude, solitary motivations reflect the why. There are many reasons why individuals spend frequent time alone. In this study, we focus on affinity for solitude, which is characterized by a desire to seek solitude due to personal enjoyment and productivity rather than avoidance of others (Borg & Willoughby, 2022; Coplan, Hipson, et al., 2019; Hu et al., 2022). There is some preliminary evidence to suggest that affinity for solitude may impact upon links between aspects of solitude and adjustment. For example, in a recent longitudinal study, adolescents who were motivated to seek solitude for positive reasons experienced better socioemotional adjustment over time, including improved self‐esteem, better quality relationships, and lower levels of depression (Borg & Willoughby, 2022). Time alone stemming from one's personal preferences and values has also been linked to increased personal growth, self‐acceptance, emotion regulation, and feelings of relatedness in emerging adults (Nguyen et al., 2018, 2019; Thomas & Azmitia, 2019). Although these findings highlight the benefits of having a positive orientation toward solitude, research in this area is still scarce. As noted previously, during emerging adulthood, desires for solitude and needs for autonomy increase (Arnett, 2014; Thomas & Azmitia, 2019), making this an especially important time to consider affinity for solitude when examining solitary experiences.

Affinity for solitude may also play a role in how episodes of pure solitude are experienced. For example, Coplan et al. (2022) recently used hypothetical vignettes to examine adolescents' affective responses to pure solitude, as well as physical solitude under different conditions of technology use. Results revealed that overall, adolescents reported higher negative and lower positive affect as virtual engagement decreased. However, in the pure solitude condition, affinity for solitude was positively linked to perceived contentment and negatively associated with boredom, loneliness, and sadness. Relatedly, Thomas et al. (2021) found that emerging adults with positive preferences for solitude (labeled “high functioning introverts”) spent more time in pure solitude than their peers who were either extroverted or motivated to spend time alone for negative reasons. Although pure solitude was generally linked to negative mood, high functioning introverts in this study demonstrated positive psychosocial functioning. In still another study, sitting alone for 10 min was associated with decreased positive mood, but results were attenuated for participants who first read about the benefits of solitude (Rodriguez et al., 2020).

Taken together, these findings highlight the importance of considering affinity for solitude when investigating how emerging adults respond to pure solitude. For emerging adults with positive perceptions of solitude, spending time alone with no distractions may be a more positive experience and the conclusion that doing nothing is both undesirable and risky may thus be premature. Still, researchers have yet to examine how solitary activities may work in conjunction with motivations for solitude to impact emerging adults' adjustment.

1.3. This study

Little is currently known about how emerging adults spend their time alone. Moreover, no studies to date have considered how different solitary activities are connected to social motivations and indices of adjustment in emerging adulthood. Accordingly, the primary goals of this study were to: (1) describe and characterize solitary activities among emerging adults attending university; (2) examine links between solitary activities and indices of adjustment; and (3) explore the moderating role of affinity for solitude in these associations. Aspects of adjustment considered included loneliness (i.e., negative feelings arising from spending too much time alone; Asher & Paquette, 2003), aloneliness (i.e., negative feelings arising from not spending enough time alone; Coplan, Hipson, et al., 2019), psychological distress, and well‐being.

We aimed to take a person‐centered approach to examine distinct groups of emerging adults based on their tendencies to engage in different types of activities while alone using latent class analysis. Theoretically derived differences among solitary activity groups in indices of adjustment, including the understudied construct of aloneliness, were also examined, along with the moderating role of affinity for solitude. Looking beyond time alone to examine solitude in a way that integrates the what and why of solitary behavior addresses discrepancies between previous findings on how solitude is experienced, which can help emerging adults engage in solitude in constructive ways.

1.3.1. Research questions

1. What kinds of activities do emerging adults engage in while alone? In particular, are there distinct groups (i.e., classes) of emerging adults who engage in specific types of activities while in solitude?

2. How do solitary activity classes during emerging adulthood differ in time alone and indices of adjustment (i.e., well‐being, psychological distress, loneliness, and aloneliness)?

3. Do findings differ according to affinity for solitude?

1.3.2. Preliminary confirmatory hypotheses

1. Time alone would be negatively associated with well‐being and positively associated with affinity for solitude, loneliness, and psychological distress.

1.3.3. Primary confirmatory hypotheses

• 2Results from latent class analysis would reveal a general distinction between emerging adults who tend to spend time alone doing versus thinking (Hipson et al., 2021);
• 3Emerging adults who spend time alone predominantly doing nothing (e.g., thinking, ruminating) would report poorer overall adjustment (i.e., lower well‐being, higher psychological distress, loneliness, and aloneliness) and less time alone as compared to those who tend to engage in other solitary activities (Hipson et al., 2021; Wilson et al., 2014);
• 4For emerging adults with higher affinity for solitude, the previously hypothesized effects of doing nothing would be attenuated (Coplan et al., 2022; Rodriguez et al., 2020).

1.3.4. Exploratory hypotheses

Additional hypotheses were, to an extent, exploratory:

• 5Should differences in groups of doing‐based solitary activities emerge, there would be a division between emerging adults involved in more constructive (e.g., hobbies, active technology use, exercise) versus passive (e.g., Netflix, scrolling social media) solitary activities (Hipson et al., 2021);
• 6There would not be differences in adjustment between groups of doing‐based activities; however, emerging adults engaged in passive solitary activities would spend more time alone than those engaged in active activities (Hipson et al., 2021);
• 7Should distinct groups of positive (e.g., daydreaming) and negative (e.g., ruminating) thinking emerge, there would not be differences in adjustment.

As well, previous research regarding the role of gender in solitary activities among adolescents and emerging adults is mixed, with some studies showing that men are more like to prefer doing over thinking (Wilson et al., 2014) and others showing no gender differences (Hipson et al., 2021). Although we deemed it important to include gender in our analyses, analyses regarding gender were exploratory and hypotheses were not offered.

Finally, data collection for this study occurred both before and during the COVID‐19 pandemic. Lockdowns and contact restrictions imposed to mitigate the spread of the virus during various phases of the pandemic substantially increased time alone (Del Fava et al., 2021; Tomori et al., 2021), and there is emerging evidence that this social isolation directly contributed to mental health difficulties (Benke et al., 2020; Fried et al., 2022; Leary & Asbury, 2022). Accordingly, COVID‐19 experiences (before vs. during) were included in our analyses on an exploratory basis and hypotheses were not offered.

2. METHOD

2.1. Ethics

Prior to data collection, ethics approval was received by the institutions' respective Research Ethics Boards. Each participant gave informed consent online prior to participation. The preregistration is available to view at: https://osf.io/3d9k8.

2.2. Participants

Participants were N = 1798 emerging adults between the ages of 18 and 25 years (M _age = 19.73, SD = 1.46). Participants identified as male (39.9%), female (59.7%), and other (0.3%). In terms of ethnicity, participants identified as White (51.3%), mixed (17.9%), Asian (12.9%), Black (8.3%), Arab (4.8%), Native American (3.5%), and Hispanic (1.1%). Participants were primarily first‐ (51.8%) and second‐year (28.2%) university students, however, the survey was also available to third‐ (11.1%) and fourth‐year students, or higher (5.3%). At the time of assessment, 62.2% of participants were single and 31.1% were in a relationship. Most participants reported living with others (with others: 86.2%; alone: 7.1%).

2.3. Procedure

Data were collected as part of a larger online survey conducted with university students in the Northeastern United States (n = 1066) and Ontario, Canada (n = 732). Data collection began in October of 2018 and ended in August of 2020, with about two thirds of participants completing the questionnaire before the onset of the COVID‐19 pandemic. Participants in both samples attended large public universities and were enrolled in introductory psychology courses. U.S. participants attended a 1‐hr laboratory visit to complete the survey measures, which were administered through Qualtrics on a laboratory computer. Canadian participants completed questions online (via Qualtrics). All participants received course credit for their time. The first author conducted a different set of analyses on the pre‐COVID‐19 dataset and presented some preliminary findings at a developmental psychology conference (Development 2022) in June of 2022. For the purpose of this study, the full dataset had not yet been examined.

2.4. Measures

To ensure high quality data, 10 inconsistency items were interwoven throughout the survey (e.g., “I've won the Dag Hammarskjöld Prize”). Items were rated on a scale from 1 = Strongly disagree to 7 = Strongly agree, with higher scores indicating a more inconsistent response. Scores were summed and participants scoring at least two standard deviations above the mean were removed from the dataset.

2.4.1. Solitary activities

In response to the prompt “What did you do most often when you were alone – not including sleeping?” participants listed up to three things they did most while alone over the last week. Two independent raters coded the activities into 29 codes modified from Hipson et al. (2021) to reflect the transition to emerging adulthood (see Appendix A for the list of codes). For example, codes were added for work and substance use. Cohen's Kappa was used to establish inter‐rater reliability on a subsample of 200 participants, which was excellent (κ = 0.92).

2.4.2. Affinity for solitude

Affinity for solitude was measured using three items from the Preference for Solitude Scale (PSS; Burger, 1995) focusing on the enjoyment and productive use of solitude (i.e., “Time alone is often productive for me”; “When I have several hours alone, I find the time productive and pleasant”; and “I enjoy being by myself” (Borg & Willoughby, 2022; Coplan, Hipson, et al., 2019; Hu et al., 2022). Participants indicate how much each statement describes them on a scale from 1 = Almost never to 4 = Almost always. Internal consistency was satisfactory (α = 0.77). Previous studies have taken similar subsets of items from broader scales on preference for solitude and demonstrated similar psychometric properties in emerging adult samples (α = 0.75; Borg & Willoughby, 2022).

2.4.3. Time alone

Alone was defined as “by yourself or doing something by yourself‐ not including sleeping”. Two questions were used to assess time spent alone: (1) “How many times were you alone during the last week for a period lasting at least 15 minutes?”; and (2) “During the last week, how many total hours did you spend alone?” (Coplan, Hipson, et al., 2019). Participants responded to the first question on a scale from 1 = Not at all during the last week to 6 = More than four times each day during the last week. Participants responded to the second question on a scale from 1 = Less than 7 h (less than 1 h per day) to 6 = More than 35 h (more than 5 h per day). Reliability was assessed using the Pearson's r correlation coefficient and was determined to be sufficient (r = 0.65). As such, the two items were aggregated to create a single score of time spent alone. This approach is in line with previous research demonstrating a high correlation between the items, as well as convergent and discriminant validity of the aggregate score (Coplan, Hipson, et al., 2019; Coplan, Hipson, & Bowker, 2021; Hipson et al., 2021).

2.4.4. Psychological distress and well‐being

Psychological distress and well‐being were assessed using the Inventory of Depression and Anxiety Symptoms—Second Version (IDAS‐II; Watson et al., 2012). Participants indicate the extent to which they have experienced each item during the last 2 weeks. Items are rated on a scale from 1 = Not at all to 5 = Extremely. Of particular interest for this study were the subscales assessing well‐being (three items, e.g., “I felt optimistic”) and dysphoria (10 items, e.g., “I found myself worrying all the time”). In line with previous research demonstrating high reliability of the IDAS‐II in college students (dysphoria: α = 0.88; well‐being: α = 0.88; Watson et al., 2012), internal consistency for the dysphoria and well‐being subscales was high (dysphoria: α = 0.90; well‐being: α = 0.86). The dysphoria subscale has also been associated with major depressive disorder and generalized anxiety disorder (Watson et al., 2012).

2.4.5. Loneliness

Loneliness was measured using the 20‐item Revised UCLA Loneliness Scale (Russell et al., 1980). Participants indicate how often each statement describes them on a scale of 1 = Never to 4 = Often (e.g., “I am unhappy being so withdrawn”). The UCLA has demonstrated high internal consistency in previous studies with emerging adults (α = 0.87; Thomas et al., 2021). Internal reliability in this study was excellent (α = 0.93).

2.4.6. Aloneliness

Aloneliness was measured using the Solitude and Aloneliness Scale (SolAS; Coplan, Hipson, et al., 2019). The SolAS includes 12 items assessing feelings of dissatisfaction with time spent alone (e.g., “It would be nice if I could spend more time alone each day”) rated on a scale from 1 = Strongly agree to 5 = Strongly disagree. The SolAS has previously demonstrated a single factor solution with excellent internal consistency (α = 0.92; Coplan, Hipson, et al., 2019). Internal reliability in this study was excellent (α = 0.93).

3. RESULTS

3.1. Preliminary analyses

Data cleaning steps and assumption checks were first conducted using IBM SPSS Statistics for Macintosh (v.28), followed by additional preliminary analyses using R Software (v. 1.4.1106). Descriptive statistics and correlations between continuous study variables are reported in Table 1. Of note, time alone was significantly and positively related to psychological distress, affinity for solitude, and loneliness, and negatively related to well‐being and aloneliness.

TABLE 1.

Descriptive statistics and correlations between continuous study variables.

Variables	n	M	SD	1	2	3	4	5	6
1. Age	1798	19.73	1.46
2. Time alone	1798	4.14	1.36	0.07**
3. Affinity for solitude	1770	5.08	1.32	0.04	0.16***
4. Well‐being	1772	2.88	0.84	−0.01	−0.14***	0.21***
5. Distress	1773	2.36	0.92	0.03	0.12***	−0.08***	−0.34***
6. Loneliness	1770	1.97	0.56	0.06*	0.28***	−0.04	−0.40***	0.51***
7. Aloneliness	1774	2.62	0.77	0.02	−0.09***	0.36***	−0.06*	0.25***	0.18***

^*** p < 0.001;

^** p < 0.01;

^* p < 0.05.

Potential differences among main study variables (time alone, affinity for solitude, well‐being, psychological distress, loneliness, and aloneliness) as a function of gender (males vs. females), COVID‐19 (pre vs. during), living situation (alone vs. with others), and relationship status (in a relationship vs. single) were analyzed with a series of t‐tests (with Bonferroni corrections applied; p ≤ 0.008). Results are displayed in Table 2.

TABLE 2.

t‐tests with Bonferroni correction for study variables split by binary demographic variables.

	Gender				COVID‐19
	Male	Female	t	d	Before	During	t	d
	M (SE)	M (SE)			M (SE)	M (SE)
Time alone	4.22 (0.05)	4.09 (0.04)	2.10	0.10	4.05 (0.04)	4.32 (0.06)	−3.88***	−0.20
Affinity	4.99 (0.05)	5.16 (0.04)	−2.66**	−0.13	5.16 (0.04)	4.95 (0.06)	3.04**	0.16
Well‐being	2.92 (0.03)	2.85 (0.03)	1.66	0.08	2.92 (0.02)	2.80 (0.03)	2.76**	0.14
Distress	2.19 (0.03)	2.47 (0.03)	−6.42***	−0.31	2.41 (0.03)	2.27 (0.04)	2.98**	0.15
Loneliness	1.96 (0.02)	1.97 (0.02)	−0.53	−0.03	1.98 (0.02)	1.95 (0.02)	0.77	0.04
Aloneliness	2.53 (0.03)	2.68 (0.02)	−4.13***	−0.20	2.66 (0.02)	2.56 (0.03)	2.56	0.13

	Relationship status				Living situation
	Single	Involved	t	d	Alone	Others	t	d
	M (SE)	M (SE)			M (SE)	M (SE)
Time alone	4.30 (0.04)	3.86 (0.06)	6.14***	0.32	4.94 (0.11)	4.09 (0.03)	7.40***	0.62
Affinity	5.08 (0.04)	5.14 (0.06)	−0.90	−0.05	5.31 (0.11)	5.08 (0.03)	1.99	0.17
Well‐being	2.84 (0.03)	2.95 (0.04)	−2.57	−0.13	2.86 (0.07)	2.87 (0.02)	−0.25	−0.02
Distress	2.40 (0.03)	2.33 (0.04)	1.49	0.08	2.57 (0.08)	2.36 (0.02)	2.30	0.23
Loneliness	2.01 (0.02)	1.88 (0.02)	5.02***	0.25	2.14 (0.05)	1.96 (0.01)	3.25***	0.32
Aloneliness	2.60 (0.02)	2.67 (0.03)	−1.66	−0.09	2.59 (0.07)	2.63 (0.02)	−0.57	−0.05

Note: d = Cohen's d.

^*** p < 0.001;

^** p < 0.008.

Compared to males, females reported significantly higher affinity for solitude, psychological distress, and aloneliness. Gender was not significantly associated with time alone, well‐being, or loneliness. Before the onset of the COVID‐19 pandemic, participants reported significantly higher affinity for solitude, psychological distress, and well‐being, as well as lower time spent alone. COVID‐19 status was not significantly associated with loneliness or aloneliness. Compared to participants who lived with others, participants who lived alone reported significantly higher time alone and loneliness. Living situation was not significantly associated with affinity for solitude, well‐being, psychological distress, or aloneliness. Compared to participants who were in a romantic relationship, those that were single reported significantly higher time alone and loneliness. Relationship status was not significantly associated with affinity for solitude, well‐being, psychological distress, or aloneliness.

3.1.1. Solitary activities

Solitary activity codes and frequencies are reported in Table 3. The most commonly endorsed solitary activity was homework, followed by passive media use, routine, and video‐games. Few emerging adults endorsed thinking‐related (e.g., planning, daydreaming) or relaxation‐based activities (e.g., meditation, yoga), and solitary substance use was also uncommon.

TABLE 3.

Solitary activity codes and frequencies.

Codes	Example behaviors	Overall n (%)
Homework	Homework, assignments, study	1286 (71.5%)
Passive media	TV, Netflix, YouTube, movies	1100 (61.2%)
Routine	Cleaning, eating, showering, driving	527 (29.3%)
Video games	Computer games, games on phone, Xbox	424 (23.6)
Reading	Books, comics	260 (14.5%)
Exercise	Walking, gym, playing sports	217 (12.1%)
Music	Radio, music	206 (11.5%)
Social media	Social media, Facebook, Instagram	198 (11%)
Unspecified screen time	Phone, electronics	188 (10.5%)
Hobbies	Playing music, art, photography	130 (7.2%)
Work	Work	58 (3.2%)
Outdoors	Outside, go for a walk/run	49 (2.7%)
Texting	Texting, Instant messaging	47 (2.6%)
Class	Go to class, watch online lectures	38 (2.1%)
Relaxation	Relax, resting	30 (1.7%)
Nothing	Nothing	26 (1.4%)
Talk on phone	Talking on the phone, FaceTime	24 (1.3%)
Daydream	Thinking	21 (1.2%)
Self‐care	Journaling, self‐care	18 (1%)
Substance use	Drinking alcohol, doing drugs	16 (0.9%)
Animals	Play with my dog/cat	12 (0.7%)
Planning	Organizing schedule, making plans	12 (0.7%)
Meditating	Meditation	10 (0.6%)
Negative	Worrying, crying	8 (0.4%)
Spiritual	Praying, bible, spirituality	5 (0.3%)
Yoga	Yoga	3 (0.2%)
Philanthropy	Volunteering	2 (0.1%)

Note: N = 1798.

3.2. Primary analyses

3.2.1. Latent class analysis

Activities were coded dichotomously, such that participants either reported or did not report engaging in the activity. Although we anticipated running all analyses using R Software (v. 1.4.1106), there is currently no package for R that facilitates parameter constraints in LCA. As such, latent class analyses were conducted using Mplus (v. 1.8.9). Finch and Bronk (2011) recommend a sample size of at least N = 500 for conducting latent class analyses. For cross‐validation purposes, the sample was thus randomly split into approximately two halves (Subsample 1: n = 889; Subsample 2: n = 909) using IBM SPSS Statistics for Macintosh (v.28). Before splitting the sample, continuous variables were examined for outliers and nonnormality. Cook's distance values did not exceed 1, indicating no univariate outliers.

An exploratory approach to LCA was taken in the first half‐sample. Five models were estimated, beginning with a baseline 1‐class model and increasing sequentially. The final number of classes was determined by drawing upon the relevant previous theoretical and empirical literature (e.g., Coplan, Hipson, & Bowker, 2021; Hipson et al., 2021; Nguyen et al., 2019; Thomas & Azmitia, 2019), as well as considerations of model utility and fit indices such as Bayesian information criterion (BIC), sample size adjusted BIC (aBIC), bootstrapped likelihood ratio test (BLRT), and entropy. Better fit was indicated by (1) lower values for BIC and aBIC, (2) significant log‐likelihood for BLRT when comparing models (i.e., p < 0.05), and (3) an entropy value higher than 0.80. BLRT was performed with 999 bootstrapped samples.

Fit indices are reported in Table 4. BIC was lowest for the 1‐class solution. However, aBIC (which is more reliable than BIC; Nylund et al., 2007) was lowest for the 3‐class solution. Although BLRT showed improvement in the log‐likelihood across all models, BLRT estimates were unreliable in the 4‐ and 5‐class models due to bootstrap draws revealing LRT values that were smaller than the observed value, as well as failure to replicate the best log‐likelihood value. As such, we chose to interpret the 3‐class solution (see Figure 1).

TABLE 4.

Summary of fit indices for unconstrained LCA in first half sample.

	Class breakdown	BIC	aBIC	BLRT p	Entropy
1	1	9987.24	9901.49	NA	1
2	0.311, 0.688	9991.50	9816.83	0.001	1
3	0.201, 0.287, 0.512	10,070.64	9807.05	0.001	0.98
4	0.451, 0.194, 0.115, 0.240	10,171.84	9819.33	0.001	0.99
5	0.263, 0.252, 0.076, 0.198, 0.213	10,276.89	9835.45	0.001	0.93

Abbreviations: aBIC, sample‐size adjusted Bayesian information criterion; BIC, Bayesian information criterion; BLRT, bootstrapped likelihood ratio test. Chosen 3‐class model values are bolded.

FIGURE 1.

Latent class probabilities for 3‐class solution.

The first class was labeled Academic/Productive (20.1%) and was comprised almost exclusively of emerging adults who reported doing homework and engaging in routine activities. Members of the Academic/Productive class were also less likely to report using passive media than the other classes (although this was still a relatively common activity). We labeled the second class Leisure (28.7%), which included emerging adults who engaged in the highest levels of leisure activities (e.g., reading, music, and exercise) and passive media use, as well as moderate levels of routine tasks. The third class was labeled Academic (51.2%) and was comprised of just over half of the sample. This class included emerging adults who engaged in high levels of homework and passive media use, moderate levels of leisure activities, and low levels of routine activities.

Overall, there was little variation in solitary activities. Notably, very few participants endorsed thinking‐based activities (probabilities ranged from 0% to 2.6% across all classes). Moreover, several logit thresholds approached extreme values (i.e., probabilities of 0 or 1). For parameters with extreme probabilities, standard errors and p‐values cannot be generated, which presents a major problem for cross‐validation. To substantiate exploratory findings, we had planned to conduct a CLCA on Subsample 2 using indicator threshold and equality constraints obtained from Subsample 1. Specifically, for significant threshold values in the first half‐sample, we proposed to constrain significantly positive threshold parameters to be greater than the lower bound of the 95% CI and less than the upper bound of the 95% CI. We also planned to constrain class‐specific thresholds to be equal when the difference between them was nonsignificant in the first half‐sample. However, for parameters without standard errors and p‐values, such constraints cannot be performed. Thus, the model could not be cross‐validated. Extreme values were also present in the 2‐, 4‐, and 5‐class models.

3.2.2. Multiple linear regression analyses

Given that parameters could not be effectively estimated for cross‐validation, we implemented our backup analysis plan (as preregistered). Specifically, moderated multiple hierarchical regression analyses were conducted using the lm package in R Software (v. 1.4.1106) to test Hypothesis 3 (i.e., Emerging adults who spend time alone predominantly doing nothing [e.g., thinking, ruminating] would report poorer overall adjustment and less time alone as compared to those who tend to engage in other solitary activities), Hypothesis 4 (i.e., For emerging adults with higher affinity for solitude, the previously hypothesized effects of doing nothing would be attenuated), and Hypothesis 6 (i.e., There would not be differences in adjustment between groups of doing‐based activities; however, adolescents engaged in passive solitary activities would spend more time alone than those engaged in active activities). Because we were no longer able to take a person‐oriented approach, Hypothesis 5 could not be tested directly (i.e., Should differences in groups of doing‐based solitary activities emerge, there would be a division between emerging adults involved in more constructive versus passive solitary activities). As well, since so few participants engaged in thinking‐based solitary activities, Hypothesis 7 (i.e., Should distinct groups of positive and negative thinking emerge, there would not be differences in adjustment) was not examined. Missing data (displayed in Table 1) ranged from 0 (i.e., age, gender, and COVID‐19) to 6.67% (i.e., living situation, relationship status). Missing data were imputed using the mice package (van Buuren & Groothuis‐Oudshoorn, 2011) and analyses were conducted using 50 pooled imputed datasets. Genders that did not meet a sample size of n = 30 (i.e., nonbinary) were coded as missing in analyses.

Models were constructed separately for each outcome variable (i.e., time alone, psychological distress, well‐being, loneliness, and aloneliness). Gender was included as a control variable in Model 1 across all analyses. If preliminary analyses revealed that age, COVID‐19 (pre vs. during), living situation (alone vs. with others), and relationship status (in a relationship vs. single) were significantly associated with the outcome variables, they were included as well.

Finally, solitary activities were aggregated to facilitate exploration of preregistered hypotheses. Specifically, aggregated variables representing different groups of solitary activities included passive technology (i.e., passive media, social media), active/leisure (i.e., hobbies, reading, music, exercise, and outdoors) and thinking (i.e., nothing, planning, daydreaming, and negative thinking). If participants endorsed engaging in any of the grouped activities, they received a score of 1. If they did not endorse engaging in any of the grouped activities, they received a score of 0. For each equation, solitary activities (i.e., passive technology, active/leisure, and thinking) and mean‐centered affinity for solitude were added in Model 2, followed by solitary activity x mean‐centered affinity for solitude interaction terms in Model 3.

Time alone

Results predicting time alone are displayed in Table 5. After controlling for relevant demographic variables in Model 1, results for Model 2 indicated that affinity for solitude was significantly positively associated with time alone, but there were no significant effects for solitary activities. As well, there were no significant two‐way interactions between solitary activities and affinity for solitude at Model 3.

TABLE 5.

Hierarchical regression results for time alone.

Variables	b (SE)	t	p	95% CI for b		R 2	ΔR 2
				LL	UL
Model 1						0.07	0.07***
Gender	−0.10 (0.06)	−1.54	0.124	−0.22	0.02
Age	0.06 (0.02)	2.85	0.004	0.02	0.10
COVID‐19	0.30 (0.07)	4.54	<0.001	0.16	0.44
Relationship	−0.47 (0.07)	−6.79	<0.001	−0.61	−0.33
Living	−0.84 (0.12)	−7.00	<0.001	−1.08	−0.60
Model 2						0.09	0.02***
Passive technology	−0.02 (0.07)	−0.27	0.786	−0.16	0.12
Active/Lleisure	−0.03 (0.09)	−0.37	0.713	−0.21	0.15
Thinking	0.02 (0.16)	0.16	0.877	−0.29	0.33
Affinity	0.17 (0.02)	7.11	<0.001	0.13	0.21
Model 3						0.10	0.01
Passive * Affinity	−0.02 (0.05)	−0.33	0.744	−0.12	0.08
Active * Affinity	0.07 (0.07)	1.05	0.296	−0.07	0.21
Thinking * Affinity	−0.12 (0.10)	−1.18	0.238	−0.32	0.08

Abbreviations: Affinity, affinity for solitude; CI, confidence interval; LL, lower limit; UL, upper limit.

^*** p < 0.001.

Psychological distress

Results predicting psychological distress are displayed in Table 6. After controlling for relevant demographic variables in Model 1, results for Model 2 indicated significant negative associations for passive technology activities, active/leisure activities, and affinity for solitude, as well as a significant positive association for thinking. At Model 3, there were no significant two‐way interactions between solitary activities and affinity for solitude.

TABLE 6.

Hierarchical regression results for psychological distress.

Variable	b (SE)	t	p	95% CI for b		R 2	ΔR 2
				LL	UL
Model 1						0.03	0.03***
Gender	0.28 (0.04)	6.41	<0.001	0.20	0.36
COVID‐19	−0.14 (0.05)	−3.10	0.002	−0.24	−0.04
Model 2						0.06	0.03***
Passive technology	−0.11 (0.05)	−2.38	0.017	−0.21	−0.01
Active/Leisure	−0.13 (0.06)	−2.06	0.040	−0.25	−0.01
Thinking	0.57 (0.11)	5.06	<0.001	0.35	0.79
Affinity	−0.07 (0.02)	−4.34	<0.001	−0.11	−0.03
Model 3						0.06	0.00
Passive * Affinity	0.04 (0.03)	1.19	0.234	−0.02	0.10
Active * Affinity	0.03 (0.05)	0.67	0.501	−0.07	0.13
Thinking * Affinity	−0.07 (0.07)	−1.07	0.286	−0.21	0.07

Abbreviations: Affinity, affinity for solitude; CI, confidence interval; LL, lower limit; UL, upper limit.

^*** p < 0.001.

Well‐being

Results for well‐being are displayed in Table 7. After controlling for relevant demographic variables in Model 1, results for Model 2 indicated a significant positive association for affinity for solitude. There was also a significant negative association for thinking, but there were no significant effects for passive technology or active/leisure activities. At Model 3, there were no significant two‐way interactions between solitary activities and affinity for solitude.

TABLE 7.

Hierarchical regression results for well‐being.

Variables	b (SE)	t	p	95% CI for b		R 2	ΔR 2
				LL	UL
Model 1						0.01	0.01
Gender	−0.06 (0.04)	−1.59	0.111	−0.14	0.02
COVID‐19	−0.11 (0.04)	−2.68	0.008	−0.19	−0.03
Model 2						0.06	0.05***
Passive technology	0.02 (0.04)	0.41	0.679	−0.06	0.10
Active/Leisure	0.10 (0.06)	1.88	0.060	−0.02	0.22
Thinking	−0.34 (0.10)	−3.29	0.001	−0.54	−0.14
Affinity	0.14 (0.01)	9.22	<0.001	0.12	0.16
Model 3						0.06	0.00
Passive * Affinity	−0.02 (0.03)	−0.59	0.556	−0.08	0.04
Active * Affinity	0.03 (0.04)	0.59	0.550	−0.05	0.11
Thinking * Affinity	0.02 (0.07)	0.29	0.770	−0.12	0.16

Abbreviations: CI, confidence interval; LL, lower limit; UL, upper limit; Affinity, affinity for solitude.

^*** p < 0.001.

3.2.3. Loneliness

Results for loneliness are displayed in Table 8. After controlling for relevant demographic variables in Model 1, results in Model 2 indicated a significant positive association for thinking, but there were no significant effects for passive technology activities, active/leisure activities, or affinity for solitude. At Model 3, there were no significant two‐way interactions between solitary activities and affinity for solitude.

TABLE 8.

Hierarchical regression results for loneliness.

Variables	b (SE)	t	p	95% CI for b		R 2	ΔR 2
				LL	UL
Model 1						0.03	0.03***
Gender	0.03 (0.03)	1.14	0.254	−0.03	0.09
Age	0.03 (0.01)	2.78	0.005	0.01	0.05
Relationship	−0.15 (0.03)	−5.25	<0.001	−0.21	−0.09
Living	−0.17 (0.05)	−3.32	0.001	−0.27	−0.07
Model 2						0.04	0.01***
Passive technology	−0.04 (0.03)	−1.40	0.162	−0.10	0.02
Active/Leisure	−0.05 (0.04)	−1.41	0.152	−0.13	0.03
Thinking	0.26 (0.07)	3.67	<0.001	0.12	0.40
Affinity	−0.02	−1.91	.057	−0.04	0.00
Model 3						0.04	0.00
Passive * Affinity	0.04 (0.02)	1.87	0.061	0.00	0.08
Active * Affinity	0.06 (0.03)	1.98	0.048	0.00	0.12
Thinking * Affinity	−0.00 (0.04)	−0.09	0.926	−0.08	0.08

Abbreviations: Affinity, affinity for solitude; CI, confidence interval; LL, lower limit; UL, upper limit.

^*** p < 0.001.

Aloneliness

Results for aloneliness are displayed in Table 9. After controlling for relevant demographic variables in Model 1, results in Model 2 indicated significant negative associations for passive technology and active/leisure activities, and significant positive associations for thinking and affinity for solitude. At Model 3, there were no significant two‐way interactions between solitary activities and affinity for solitude.

TABLE 9.

Hierarchical regression results for aloneliness.

Variables	b (SE)	t	p	95% CI for b		R 2	ΔR 2
				LL	UL
Model 1						0.01	0.01
Gender	0.15 (0.04)	4.00	<0.001	0.07	0.23
Model 2						0.15	0.14***
Passive technology	−0.11 (0.04)	−3.01	0.003	−0.19	0.03
Active/Leisure	−0.14 (0.05)	−2.81	0.005	−0.24	−0.04
Thinking	0.25 (0.09)	2.73	0.006	0.07	0.43
Affinity	0.21 (0.01)	16.10	<0.001	0.19	0.23
Model 3						0.15	0.00
Passive * Affinity	−0.00 (0.03)	−0.17	0.867	−0.06	0.06
Active * Affinity	−0.01 (0.04)	−0.24	0.813	−0.09	0.07
Thinking * Affinity	0.13 (0.06)	2.33	0.020	0.01	0.25

Abbreviations: Affinity, affinity for solitude; CI, confidence interval; LL, lower limit; UL, upper limit.

^*** p < 0.001.

3.3. Exploratory analyses

We proposed to explore gender differences in solitary activity groups. Given the aforementioned statistical limitations associated with the LCA, some exploratory analyses were conducted pertaining to potential differences among participants who reported versus did not report engaging in different types of solitary activities (i.e., passive technology, active/leisure, thinking). First, results from Chi‐square analyses indicated significant gender differences among those who engaged versus did not engage in all three types of solitary activities (passive technology: χ ² (1) = 52.94, p < 0.001; active/leisure: χ ² (1) = 13.99, p < 0.001; thinking: χ ² (1) = 4.13, p = 0.042). Females were more likely to spend time alone thinking or engaged in passive technology activities, whereas males were more likely to spend time alone engaged in active/leisure activities.

Finally, we ran a series of t‐tests to explore whether participants who reported versus did not report engaging in different types of solitary activities differed in affinity for solitude. Results indicated no significant differences in affinity for solitude between participants engaging versus not engaging in passive technology activities (t (1176.2) = 0.46, p = 0.649, 95% CI [−0.11, 0.16]; no: M = 5.11, SE = 0.06; yes: M = 5.08, SE = 0.04), active/leisure activities (t (353.29) = 0.96, p = 0.338, 95% CI [−0.08, 0.26]; no: M = 5.10, SE = 0.03; yes: M = 5.01, SE = 0.08), or thinking activities (t (67.13) = −0.35, p = 0.732, 95% CI [−0.48, 0.34]; no: M = 5.08, SE = 0.03; yes: M = 5.15, SE = 0.20).

4. DISCUSSION

The goals of this study were to: (1) characterize solitary activities among emerging adults attending university; (2) examine links between different types of solitary activities and indices of adjustment; and (3) explore whether affinity for solitude moderates these associations. To accomplish these goals, a large sample of emerging adults at two North American universities completed a series of self‐report measures assessing their time spent alone, most frequent solitary activities (over the last week), affinity for solitude, psychological distress, loneliness, aloneliness, and general well‐being. In line with expectations, time alone was negatively associated with well‐being and positively associated with affinity for solitude, loneliness, and psychological distress. These findings suggest that increased quantity of time spent alone may be risky for emerging adults, which is consistent with previous research on the generally negative implications of solitude (see McVarnock et al., 2023 for a recent review). However, the results indicate that the quality of solitude also contributes to aspects of adjustment among emerging adults. We will discuss these findings in detail next.

4.1. Characteristics of solitary activities

In this study, the most commonly endorsed solitary activities were homework, passive media use (e.g., Netflix), and routine (e.g., cooking and cleaning). Very few emerging adults reported engaging in thinking‐based activities, such as planning, daydreaming, or ruminating (with overall percentages ranging from 0.4% to 1.4% endorsement). These findings provide some support for previous studies suggesting that young people do not enjoy engaging in pure solitude (i.e., being alone with their thoughts and no distractions; Buttrick et al., 2019; Wilson et al., 2014).

Compared to previous results among adolescents (Hipson et al., 2021), our findings suggest that there is less variation in how emerging adults spend their solitary time. The transition to postsecondary education is stressful for many students (Center for Collegiate Mental Health, 2020), and increased academic and social demands impact upon how young people spend their time (Arnett, 2014). Research suggests that emerging adults attending university spend much of their leisure time socializing with friends (Doerksen et al., 2014) and engaging with technology (Coyne et al., 2013). However, there also remains a strong focus on academics, with university students engaging in roughly 20 hr of academic‐related activities per week (Greene & Maggs, 2018).

In many cases, emerging adults attending university are living away from home and learning to take care of themselves for the first time. From this perspective, it is intuitive that most emerging adults would use solitude instrumentally to get stuff done (e.g., homework, household tasks). It has been suggested that passive media use, such as television watching, decreases as adolescents get older (Rideout et al., 2010). Although emerging adults may be less likely than adolescents to passively engage with technology overall, our findings suggest that solitary passive technology use may increase after transitioning to university. For example, Hipson et al. (2021) found that 41% of adolescents endorsed passive technology use, whereas passive technology use in this study was endorsed by 61% of emerging adults. Given high levels of passive technology use evident in this study, it can be speculated that emerging adults use passive technology to relax and recover from daily stressors (e.g., by watching Netflix and scrolling through social media). This idea is consistent with previous research indicating that emerging adults are motivated to watch television in part due to its undemanding nature rather than genuine enjoyment (Arnett, 2014).

4.2. Classes

As specified in our registered report, we next employed LCA to identify groups of emerging adults characterized by their engagement in different solitary activities. Three groups of emerging adults were uncovered who engaged in some solitary activities more than others. However, due to statistical limitations associated with lack of variation in solitary activities, we were unable to cross‐validate the 3‐class model. As such, results regarding solitary activity classes should be interpreted with caution.

The first class was labeled Academic/Productive and was characterized by increased time spent doing homework, engaging in routine activities (e.g., cooking, cleaning), and passive technology use (e.g., TV, scrolling social media). The second class, labeled Leisure, was characterized by engagement in constructive leisure activities (e.g., reading, listening to music), passive technology use, as well as some routine tasks. The third (and largest) class, labeled Academic, was characterized by homework and passive technology use. Taken together, these results did not suggest a distinction between emerging adults who tended to spend time alone doing versus thinking, which does not align with previous research with adolescents (Hipson et al., 2021). However, given the aforementioned analytical issues, it is difficult to determine whether differences in solitary activity classes relate to developmental differences between adolescents and emerging adults or whether they are statistical artifacts.

4.3. Solitary activities and adjustment

Our next set of analyses focused on the unique contributions of activity type (i.e., passive technology, active leisure, and thinking) to several adjustment outcomes. As expected, thinking was associated uniquely with indices of maladjustment, including increased psychological distress and loneliness, as well as lower well‐being. It is not surprising, then, that emerging adults in this study spent so little time alone with their thoughts. Findings are consistent with previous research indicating that young people experience pure solitude negatively (Hipson et al., 2021; Wilson et al., 2014).

Although some evidence suggests that thinking positive thoughts may protect against some of the negative outcomes associated with pure solitude (Nguyen et al., 2018), it may be difficult to regulate one's thoughts in practice (Wilson et al., 2014). Compared to deliberate forms of thinking, letting one's mind wander without direction or intent (i.e., unintentional daydreaming) may lead to more ruminative and negative thoughts (Zedelius & Schooler, 2016), which are associated with a range of poor outcomes, including increased psychological symptoms, negative affect, and stress (Bratman et al., 2021; Irie et al., 2019). It is also important to note that other studies have linked “just thinking” while alone to negative outcomes regardless of whether thoughts are positive or negative (Buttrick et al., 2019; Hipson et al., 2021). As such, additional research is necessary to understand how and when engaging in pure solitude is linked to maladjustment.

As hypothesized, subtypes of doing‐based activities did not differ in adjustment (Hipson et al., 2021). That is, passive technology and active leisure activities both predicted benefits related to reduced psychological distress (and were unrelated to well‐being). Emerging adults may thus use passive and active solitary activities as a means of coping with increased stress and mental health challenges during the transition to university (Center for Collegiate Mental Health, 2020). Alternatively, lower levels of distress may inspire motivation to engage in more positive solitary activities as opposed to doing nothing.

Whereas thinking‐based solitary activities were linked to higher loneliness and aloneliness, doing‐based solitary activities were linked to lower aloneliness (and were unrelated to loneliness). Findings here provide initial evidence that spending time alone with one's thoughts may not be an effective approach to satisfying the need for solitude, at least not during the university years. To effectively combat feelings of aloneliness, emerging adults may instead carve out time to engage in passive or active solitary activities. These findings may have important implications for emerging adults, as: (1) desire for solitude increases after adolescence (Thomas & Azmitia, 2019); and (2) feelings of aloneliness in this study were correlated with poorer adjustment outcomes (i.e., higher loneliness and psychological distress, lower well‐being).

4.4. Affinity for solitude

Research on solitude has traditionally focused on motivations for spending time alone, with few studies examining how such motivations translate to solitary behavior (McVarnock et al., 2023). In this study, affinity for solitude was associated with increased time alone, suggesting that emerging adults who are motivated to spend time alone for positive reasons, such as enjoyment, do in fact engage in more frequent solitude. Affinity for solitude was also linked to higher feelings of aloneliness. Research on aloneliness is in its infancy, and to date, only one study has examined aloneliness in emerging adulthood (Coplan, Hipson, et al., 2019). The present findings suggest that emerging adults with higher affinity for solitude are more prone to dissatisfaction with time alone, which is consistent with their heightened desire for solitude.

Time alone and aloneliness were both correlated with poor adjustment, which speaks to the delicate balance emerging adults must achieve between spending too much versus not enough time alone. In addition, despite being positively related to both time spent alone and the desire for more solitude, affinity for solitude was also correlated with positive adjustment (i.e., higher well‐being and lower psychological distress). These results provide additional support regarding the benefits of having a positive orientation toward solitude during adolescence and emerging adulthood (Borg & Willoughby, 2021, 2022; Nguyen et al., 2018) and point to the complex ways that time alone and motivations for solitude might interact to bolster (or hinder) adjustment. Results may have implications for emerging adults with higher affinity for solitude, who require more high‐quality solitary time to function optimally (Coplan, Hipson, et al., 2019). Increased time alone may provide emerging adults who have higher affinity for solitude with space to recharge their social batteries (Leary et al., 2003; White et al., 2022), which may be particularly important given increased social demands in university (Doerksen et al., 2014).

Interestingly, affinity for solitude was not related to engagement in different types of solitary activities. It is possible that subsequent research might still reveal such an association, but our results suggest that although emerging adults higher in affinity for solitude may spend more time alone, they are not necessarily more likely to spend their solitary time in unique ways. Hipson et al. (2021) similarly found that solitary activities were not linked to preference for solitude. The researchers postulated that young people engage in different solitary activities not because they want to be alone, but because they want to engage in the activity itself (e.g., watching Netflix alone to watch Netflix, rather than be alone). Solitary activities may thus function independent of motivations for solitude. Still, individuals seek solitude for a variety of reasons. Findings may differ for those who seek solitude for more negative reasons (e.g., shyness and social avoidance).

A primary aim of this study was to examine how solitary activities and motivations interact in the prediction of time alone and adjustment. There is some evidence to suggest that spending time alone with no distractions is more beneficial for young people with positive orientations toward solitude (Coplan et al., 2022; Rodriguez et al., 2020). As such, we hypothesized that the negative implications of thinking would be attenuated for emerging adults with higher affinity for solitude. Contrary to expectations, however, affinity for solitude did not play a role in how solitary activities were experienced. Findings indicate that pure solitude may be a negative experience, even among emerging adults who are motivated to seek time alone for productivity and enjoyment. Notwithstanding, despite a large sample size, moderation effects can be difficult to detect (Aguinis et al., 2017) and future research is required to further replicate these results.

4.5. Implications, limitations, and future directions

Studies have identified several important factors that play a role in outcomes of solitude; however, very few studies have integrated them. For the first time, we examined time alone in conjunction with solitary activities and motivations in emerging adulthood. Findings suggest that emerging adults' experience of solitude is embedded not only in how much time they spend alone, but also in what they do behind closed doors and why they seek solitude in the first place. If replicated, such findings may be used to help emerging adults develop healthy solitude habits and stay balanced during the transition to university.

This study relied on retrospective reports of emerging adults' time alone over the last week, which allowed us to examine solitude as it naturally occurs in emerging adults' lives. However, data were cross sectional. Thus, although we speculate that engaging in pure solitude leads to poorer adjustment outcomes, it is entirely possible that increased feelings of distress and loneliness motivate emerging adults to spend time alone ruminating or engaging in other forms of maladaptive thinking (Yun et al., 2023). In addition, retrospective reports open the door to recall issues if individuals cannot accurately remember how much time they spent alone or what they did during their time in solitude (Beckett et al., 2001). To examine naturally occurring solitude in real time and determine the direction in which solitary activities and adjustment are connected, additional studies may use longitudinal designs or experience sampling methods (Bernstein et al., 2018). Future studies may also look more closely at ambiguous technology‐based activities. For example, “video games” was endorsed by almost a quarter of participants in this study. However, the nature of such gaming remains unknown, making it difficult for us to categorize video games as an active or passive activity.

As well, data were collected in two waves, with the second wave taking place during COVID‐19 lockdowns and restrictions. This enabled us to gain insights into emerging adults' solitary behavior before versus after the onset of the COVID‐19 pandemic, which is important given that most studies on the implications of COVID‐19 focus on loneliness rather than time alone (Padmanabhanunni & Pretorius, 2021; Witt et al., 2020). Indeed, findings showed that emerging adults spent more time alone (and felt more distressed) during COVID‐19. Given what we know about the risks of too much solitude, it is possible that contact restrictions contributed to increased distress during COVID‐19. Still, feelings of loneliness and aloneliness did not differ before and after the pandemic (as one may have expected given increased solitude), and well‐being was actually higher after the onset of COVID‐19. Previous studies have demonstrated that although they are related, distress and well‐being are distinct constructs, which may be influenced in different ways by external factors (Aeon et al., 2021). Given that data were partially collected during the summer months, it is possible that reduced academic demands contributed to higher well‐being. To better understand experiences of solitude and well‐being during the COVID‐19 pandemic, additional research is necessitated.

Findings from this study indicate that although few emerging adults spend time in solitude primarily thinking, engaging in pure solitude (i.e., thinking) is linked to negative outcomes (i.e., lower well‐being and higher psychological distress, loneliness, and aloneliness). Although previous studies generally indicate that being alone with one's thoughts is a negative experience for young people (Hipson et al., 2021; Wilson et al., 2014), there is some evidence to suggest that spending time in pure solitude confers affective regulation benefits (Nguyen et al., 2022). Indeed, pure solitude may have a deactivation effect, wherein high arousal emotions (e.g., happiness and anger) decrease and low arousal emotions (e.g., relaxation and loneliness) increase (Nguyen et al., 2018). In this study, we did not consider differences in combinations of valence and arousal when examining affective adjustment outcomes. Additional research should include more low arousal positive emotions to determine whether (and under what conditions) there are benefits associated with time alone.

It is important to note that not all emerging adults make the transition to university (Statistics Canada, 2017). As such, present findings are only generalizable to a specific group of emerging adults. Additional studies are needed to determine what emerging adults do alone beyond the university setting. In addition, research on solitude during the childhood years is virtually nonexistent to date, with the vast majority of studies measuring either motivations for social withdrawal or solitary play in the presence of peers (McVarnock et al., 2023). Given that time alone and solitary activities are linked to adjustment in adolescence and emerging adulthood, it is imperative to examine these factors earlier in development. Findings will help researchers gain a more comprehensive understanding regarding what solitude looks like (and how it functions) throughout the life span.

5. CONCLUSION

Findings from this study suggest that although the types of solitary activities people engage in may change from adolescence to emerging adulthood (Hipson et al., 2021), the manner in which solitary activities are linked to adjustment does not. Time alone is associated with poor adjustment outcomes. Yet, solitude may be less harmful (and more beneficial) depending on how emerging adults spend their solitary time. Engaging in passive technology use or active/leisure solitary activities can help satisfy the need for solitude and reduce psychological distress, whereas pure solitude (i.e., just thinking) carries risks related to increased loneliness, aloneliness, and psychological distress, as well as lower well‐being. In addition, seeking solitude for positive reasons is related to better outcomes. Taken together, our findings support the importance of looking beyond time alone to consider factors related to the quality of emerging adults' solitary experiences.

AUTHOR CONTRIBUTIONS

Alicia McVarnock: Conceptualization, methodology, data analysis and interpretation, drafting manuscript, review and editing. Robert J. Coplan: Methodology, project administration, supervision, data interpretation, review and editing. Hope I. White: Data interpretation, review and editing. Julie C. Bowker: Project administration, data interpretation, review and editing. All authors approved the final manuscript.

FUNDING INFORMATION

This research was funded by the Social Sciences and Humanities Research Council of Canada (SSHRC) Insight Grant.

CONFLICT OF INTEREST STATEMENT

The authors do not have any potential competing interest to report.

ETHICS STATEMENT

This study was conducted in accordance with APA ethical guidelines and was approved by the Carleton University and University at Buffalo Research Ethics Boards.

PREREGISTRATION

This article was preregistered with an analysis plan at OSF Registries | Looking beyond time alone (McVarnock et al., 2022).

APPENDIX A. SOLITARY ACTIVITY CODES

A.1.

Code #	Code name	Example behaviors
Screen time
1	Social media	Facebook, Instagram, Pinterest, TikTok
2	Computer‐mediated interaction	Texting, IM, FaceTime
3	Passive screen time	TV, Netflix, Hulu, YouTube, browsing Internet
4	Video games	Computer games, games on phone
5	Phone conversation	Talking on phone
6	Screen time unspecified	Electronics, phone, scrolling
Leisure
7	Reading	Books, comics
8	Listening to…	Listening to music, podcasts
9	Hobbies	Playing music, drawing/art, knitting, puzzles
Relaxation/Spiritual/Meditation
10	Meditating/positive thinking	Clear my head
11	Yoga	Yoga
12	Spiritual	Praying
13	Relaxation	Relax, closing eyes, closing lights
14	Self‐care	Self‐care, journaling
15	Animals	Playing with dog/cat, spending time with pets (not walking outside—see “outdoors”)
Productive
16	Homework	Homework, write, study, read for school
17	Class	Going to class, class
Physical activity
18	Exercise	Exercise, sports, workout, dancing
19	Outdoors	Walking, running, walking dog, yard work
Nothing or Thinking
20	Nothing	Lay in bed, sit in bed, humming
21	Planning	Thinking about things to do, thinking future
22	Daydreaming	Think generic/unspecified/neutral/reflect
23	Negative	Crying, worrying, ruminating, bored, procrastinating
24	Sleep	Sleeping, napping
Routine
25	Routine	Eating, cooking, showering, chores, errands, driving, masturbating
26	Work	Work
27	Philanthropy	Volunteering
Substance use
28	Substance use	Drinking alcohol, doing drugs
Uncodable
29	Uncodable	Not alone, watch Netflix and reading (conflicting themes)

McVarnock, A. , Coplan, R. J. , White, H. I. , & Bowker, J. C. (2025). Looking beyond time alone: An examination of solitary activities in emerging adulthood. Journal of Personality, 93, 81–100. 10.1111/jopy.12905

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available at OSF Registries | Looking beyond time alone.