Trapped in Online Comparisons: Within Reciprocal Dynamic of Self-Assessment Motivation and Depressive State on Social Media

Abstract

Most research regarding social comparisons on social media has been limited demonstrating their effects on mental health, without explaining the underlying motivational mechanics. It appears that individuals are often motivated to reduce uncertainty about the self. Social media may serve as a tool to access diagnostic information through social comparison. However, because these platforms predominantly exposed to upward comparisons, individuals motivated to self-assess with social comparisons on social media might be negatively impacted, leading to lower mental health. Furthermore, depressive symptoms might also be more motivated to self-assess, thereby exacerbating their lower mental health. This suggests that depressive symptoms can act as both an antecedent and a consequence of self-assessment motivation (SAM). To examine these reciprocal effects, we conducted a 14-day diary study testing our model using dynamic structural equation modeling. The results revealed a reciprocal association between depressive state and SAM, where the previous depressive state was associated with more SAM, and SAM was associated with more depressive state in return, supporting the existence of a vicious cycle.

Why do people compare themselves to others who seem more successful or happier on social media, even though this behavior often leads to a depressive state? Social comparison theory¹ provides a foundational framework as an answer, positing that humans have an inherent need for self-evaluation. Over the decades, this need has been specified into four motivations: self-assessment, self-confirmation, self-improvement, and self-enhancement,² which have been hierarchically articulated, with self-assessment at the bottom and self-enhancement as the ultimate goal.^3,4 According to the former model, self-assessment motivation (SAM) plays a prominent driver, which compels individuals to seek diagnostic information about the self compared to others. This motivation functions by increasing the accuracy of self-contrual by favoring true over false ones,⁵ by choosing tasks that maximize the expected reduction of uncertainty,⁶ and by prioritizing SAM over the avoidance of negative feedback.⁷ We propose that SAM may be a key mechanism at the source of the social comparison on social media driving the negative impacts of such comparison and may operate bidirectionally.

On the one hand, SAM through social comparisons on social media can lead to increased depressive states. A positive publication bias⁸ contributes to an overwhelming volume of posts that collectively suggest the superiority of others compared to oneself, fostering the perception that others have better lives,⁹ that is, upward social comparisons.^10–13 During such comparison, the threat is induced by self-assessment through social comparison, a process that occurs during self-evaluation.¹⁴ Self-assessment can occur independently of social comparison when an absolute standard is available. Early research on SAM, for instance, operationalized it using ability and performance scores, allowing individuals to evaluate themselves without reference to others.¹⁵ However, in the context of social media use, information about the self is fine related to the situation of self in comparison to others,¹⁶ that is, a relative standard. In this sense, self-regulation integrating self-assessment processes¹⁷ involves social comparison when the standard is relative.¹⁸ Therefore, SAM is best understood as an evaluation of the self through social comparison during social media use. Ultimately, engaging in upward comparison often poses a threat to own self-image during social media use^9,19 which contributes to declining mental health.²⁰

On the other hand, depressive symptoms might increase SAM. Individuals with depressive symptoms tend to engage in more social comparison due to lower self-concept clarity.²¹ In this sense, individuals with higher levels of depressive symptoms tend to actively seek out social comparisons,^22,23 also on social media platforms²⁴ and depressive individuals are more affected by the effects of upward comparisons than others.²² Since self-assessment serves to reduce uncertainty about the self, we propose that individuals with more depressive symptoms might also be more motivated to self-assess through social comparison. This dynamic may contribute to the maintenance or even exacerbation of their negative self-views and mood.^25,26

Our research makes three key contributions that address notable gaps in the existing literature. First, research on motivational processes behind social comparison on social media is scarce.²⁸ This study addressed the gap by highlighting SAM as a key driver of its negative effects. Second, few studies have explored the reciprocal relationship between social comparison processes on social media and negative self-state.³² To expand on this aspect, our research tested the bidirectional association between depressive state (focusing on one core major depression criterion, i.e., depressive mood) and SAM during social media use. Lastly, previous research has often been constrained by methodological limitations, such as using very long intervals between wave collections.³³ This has hindered the ability to examine short-term fluctuations in motivational and affective responses, which are crucial for understanding dynamic processes like social comparison, even when overall trends remain stable. Furthermore, most studies²⁴ have focused on between-person analyses, limiting insights into within-person changes or causal processes.³¹ To overcome these limitations, we employed a diary design combined with a new statistical approach, that is, dynamic structural equation modeling.³² This approach allows to capture the within-person, reciprocal dynamics between depressive state and SAM over time, considering them simultaneously as both predictors and outcomes. Such a method is critical to explaining why some individuals are more vulnerable to the negative effects of social media use than others and resolving the inconsistent aggregate findings observed in most studies and reviews.³³ Thus, we conducted a 14-day diary study to investigate within-person effects using the following model (see Fig. 1): Previous depressive state will be positively associated with SAM (Hypothesis 1), and previous SAM will be positively associated with depressive state in return (Hypothesis 2).

FIG. 1.

Bidirectional effects of depressive state and self-assessment motivation.

Method

The first author’s university ethics committee approved the protocols of the study. Eligible participants (inclusion criteria: having an Instagram account and having reached 18 years old or above) completed an online questionnaire. The material, data, and code itself are available online on the Open Science Framework at https://osf.io/mx2gj/.

Participants

A total of 287 participants (78.6% female; age [years] = M [SD] 21.64 [4.37], range: 18–47) completed the questionnaire. The number of participants corresponded to the total number of eligible individuals from a psychology course cohort in Geneva, Switzerland. Participants were all Instagram users with a mean daily usage time (in minutes) of iM = 53.26 (iSD = 33.46). Participants’ number of social media platforms was M = 4.18 (SD = 1.44), where 80.9% used YouTube, 77.8% for Snapchat, 58.4% for TikTok, 57.7% for Facebook, 35.5% for Twitter, and 23.2% used additional platform.

Material and procedure

The study collected data using a 14-day diary. Participants were asked to respond to the same questionnaire for 14 days at the end of their day via a daily reminder on a forum platform. The response rate at the daily level was 80%. All scales ranged from 1 “completely disagree” to 7 “completely agree.” See descriptive statistics (mean, SD, and reliability) in Table 1.

Table 1.

Descriptive Statistics

	Between level			Within level
	M	SD	Omega	M	SD	Omega
Self-assessment motivation	2.459	1.274	0.978	2.442	0.894	0.809
Depressive state	2.159	1.285	0.947	2.139	0.788	0.820

SD, standard deviation.

SAM and social comparison were measured as a single construct, given that self-assessment through social comparison is regarded as a key process. Inspired by recent works³⁴ showing that self-assessment motive is unrelated to a specific social comparison direction (unlike other self-evaluation motives such as self-improvement and self-enhancement), which enables its very function (i.e., decreasing self-uncertainty), SAM was operationalized via the assessment of the self-position compared to others regardless of its direction. Based on the posts related to others participants viewed on Instagram in the last 24 hours, they reported their SAM through social comparison using three items (“I try to determine whether I’m succeeding or failing in my life compared to others,” “I ask myself whether my life is better or worse than other people’s,” “I try to find out whether my life is a success or a failure compared with those of others”). The average individual mean was iM = 2.442 (iSD = 0.894). The intraclass correlation (ICC) indicated that 67% of the variance was between-person variance.

Depressive state was measured with the depressive subscale of the profile of daily mood.³⁵ It allows assessing one specific symptom of depression, that is, depressive mood. Based on the people they had compared themselves to on Instagram in the past 24 hours, participants were asked to respond to which extent they experienced the following moods with three items (“sad,” “hopeless,” and “discouraged”). The average individual mean was iM = 2.139 (iSD = 0.788). The ICC indicated that 72.7% of the variance was between-person variance.

Data analysis

Our data have two levels with daily measures (Level 1) nested within individuals (Level 2). It was analyzed in Mplus Version 8.3.³⁶ We modeled a bivariate dynamic structural equation modeling (DSEM)^37,38 with two related within-person time series, SAM and depressive state (for a commented Mplus script, see Appendix). When estimating parameters in DSEM, we used the bitarations option in Mplus, running 1000 iterations with the Bayesian Markov Chain Monte Carlo algorithm to ensure that the estimation was more stable. We used the default diffuse priors, meaning that the results are driven by the data, not by previous assumptions. To deal with the unequal spacing of the measurement occasions, we made use of the Mplus option tinterval. We set 1-day time windows. Within-person residuals were allowed to vary across individuals. Mplus uses the log of the residual variance, which ensures that all residual variances are positive. Model convergence was satisfactory. The PSR values were within acceptable limits (1.063).^b

Results

In line with our assumption, there exist reciprocal (cross-lagged) effects from SAM to depressive state and from depressive state to SAM (see Table 2 for all results). SAM at a given moment was positively predicted by previous depressive state (b = 0.052; 95% CI [0.022, 0.086]), and depressive state was positively predicted by previous SAM (b = 0.207; 95% CI [0.140, 0.283]).

Table 2.

Unstandardized Fixed and Random Effects from the Bivariate DSEM

	Fixed effects (means)			Random effects (variances)
		95% CI			95% CI
Parameters	Estimate	Lower	Upper	Estimate	Lower	Upper
SAM intercept	2.242	2.080	2.395	1.352	1.091	1.622
Dep intercept	1.925	1.775	2.072	1.175	0.881	1.514
Autoregressive effects
SAM_{t − 1} → SAM_t	0.181	0.122	0.238	0.087	0.064	0.116
Dep_{t − 1} → Dep_t	0.270	0.210	0.342	0.086	0.063	0.119
Cross-lagged effects
SAM_{t − 1} → Dep_t	0.207	0.140	0.283	0.118	0.084	0.160
Dep_{t − 1} → SAM_t	0.052	0.022	0.086	0.023	0.015	0.036
Log residuals
Variance SAM	−1.675	−1.905	1.410	5.160	4.402	6.089
Variance Dep	−2.282	−2.596	−1.983	6.211	5.266	7.276

CI, credible intervals; Dep, depressive state; DSEM, dynamic structural equation modeling; SAM, self-assessment motivation.

Given that the sampled days represent a random selection of typical days in participants’ lives, the model assumes stationarity, meaning that cross-lagged effects remain consistent across days (e.g., SAM_{t − 1} → depressive mood_t0 is equivalent to SAM_t0 → depressive mood_{t + 1}). Therefore, the two significant cross-lagged effects together indicate the presence of a cyclical relationship, akin to what is observed in indirect effects in other models (X_{t − 1} → M_t0 → X_{t + 1}), such as prior depressive mood is associated with increased SAM, which, in turn, is associated to heightened subsequent depressive mood, and vice versa. Moreover, as the credible intervals of the cross-lagged effects did not overlap, SAM had a stronger effect on subsequent depressive state than depressive state had on predicted subsequent SAM.

Discussion

This study found that when an individual’s depressive state increased, one reported engaging in more subsequent SAM, which, in turn, further increased one’s depressive state, leading to a vicious cycle. To the best of our knowledge, this research is the first to explore within-person association between SAM and depressive state, offering key theoretical and methodological insights.

First, it extends the literature on self-assessment, supporting the dual-edged sword postulate³⁹ within the context of social media. While social comparisons might help users reduce uncertainty about themselves, they also carry the cost of being worse off during social media use. Whereas the motivation to seek knowledge about the self for the sake of accuracy alone, without seeking a positive self-view, has been dismissed,⁴⁰ these results showed otherwise. Indeed, if individuals merely assessed their situation to gain a positive self-view, they would not compare unfavorably with others being better off on social media. Thus, these results tend to confirm that individuals might be motivated to obtain an accurate self-perception⁶ regardless of the impact on the self.

Second, this research emphasizes the importance of appropriate temporal data collection and analysis to capture the within-person variability of depressive symptoms, which enables to investigate underlying processes within these fluctuations. Depressive states show daily and weekly variability, with significant changes occurring over short periods,^41–43 and recovery occurring between one week to years.⁴⁴ Studies relying on long intervals between measurements fail to capture the dynamic relationships between depressive symptoms and underlying cognitive, emotional, and motivational factors. Indeed, as the critical within variability would not be captured, these dynamic relationships could not be tested.³⁸ In other words, while the average outcome between the two waves may remain unchanged, individuals could have experienced and recovered from a major depressive episode during that time, which severely compromises the result’s validity. To accurately capture the dynamic nature of depressive symptoms and underlying factors, researchers should consider prioritizing shorter assessment intervals (e.g., daily or weekly) over extended periods (e.g., months or years).

Last, it underscores the critical role of SAM in explaining the negative impact of social media use on mental health.²⁰ While scarce research has examined motivational processes during social media comparisons,²⁸ our findings align with an established theoretical framework.^4,45 Additionally, this study brings further evidence of the risk factors operating within the vicious cycle of depression, that is, negative cognitions,⁴⁶ insomnia,⁴⁷ social rejection expectations,⁴⁸ and upward comparisons.⁴⁹ Furthermore, since SAM is inherently related to uncertainty about the self,⁴ it might drive individuals to engage in self-assessment, potentially leading to a backfire effect, especially on social media platforms like Instagram.

Moreover, although social comparisons on social media predominantly lead to contrastive rather than assimilative effects, thereby inducing threats to the self,²⁰ they can also induce assimilation processes promoting well-being.⁵⁰ As SAM is thought to precede self-improvement motive⁴ and necessary to reduce discrepancy between the self and a standard,¹⁷ it suggests that SAM could yield beneficial impacts to the self while comparing with slightly superior others.⁵¹ However, the hyper-positive portrayals of others contribute to negative self-perceptions,⁵² which are commonly observed on social media and thereby inducing a contrasting effect. Additionally, while this research focuses on the role of SAM in leading to negative impacts on the self during social media use, other self-motives, such as self-improvement and self-enhancement, may produce different effects on the self. Therefore, future research could incorporate these motives to examine the extent to which they influence self-perception during social comparisons on social media platforms, and also directly examine the extent to which the perceived distance between the actual self and superior others—whether small or large—determines whether SAM induces an assimilation or a contrasting effect during social comparison resulting in beneficial versus harmful impacts on the self.

This research presents several limitations. First, while a daily within-person design provides new insights beyond previous cross-sectional studies, it remains insufficient to establish causal inferences. More rigorous methods, such as within-person experimental manipulations,⁵³ could test causal mechanisms. Second, SAM was operationalized through social comparison as a single variable. Although in the context of social media use, SAM is primarily fulfilled through social comparison, a rigorous test of mediation analysis may still be necessary. Future research should aim to replicate these findings using two distinct measures rather than a single one, thereby allowing for a more robust mediation analysis. Third, sample characteristics may affect the generalizability of our findings. Indeed, the sample primarily consisted of young individuals, with a majority being female (80%), two demographic factors associated with higher rates of depressive symptoms.⁵⁴ Future research might consider for a more balanced sex distribution and a broader age range by employing different recruitment methods and sampling strategies, which would enhance the generalizability and applicability of the findings across different populations. Fourth, while DSEM offers powerful insights into temporal processes, unresolved methodological concerns remain,³⁷ such as missing values that might not be random or unmodeled time trends. Finally, depressive symptoms fluctuate over weeks; hence, expanding data collection beyond two weeks could capture more nuanced temporal variability in depressive states.

In closing remarks, since SAM had a stronger effect on subsequent depressive states, interventions may be more effective if they focus on reducing SAM on social media rather than directly targeting depressive mood. Clinicians and mental health professionals could consider designing interventions that help individuals to engage in self-assessment within more balanced and realistic environments. By facilitating exposure to multiple directions of social comparison with various distances between the self and others (i.e., downward, lateral, and upward), such interventions could help individuals receive more nuanced information about the self, making it less threatening. Thus, if SAM is satisfied, individuals may engage less in social comparisons during social media use, thereby mitigating its potentially harmful effects on the self.

Note

Mplus only reports the highest PSR value and not the values for all parameters.

Footnotes

Authors’ Contributions

All authors have read and agreed to the published version of the article. The authors confirm their contributions to the work as follows: R.A.: Conceptualization, methodology, formal analysis (lead), writing (lead), software, visualization, and project administration. A.Q.: Conceptualization, methodology, writing—review and editing, and supervision (lead). L.M.: Formal analysis, writing—review and editing, and supervision.

Author Disclosure Statement

All authors declare that they have no conflicts of interest.

Funding Information

This research received no specific funding.

References

Festinger

. A theory of social comparison processes. Hum Relat, 1954; 7(2):117–140; doi: 10.1177/001872675400700202

Taylor

, Neter

, Wayment

. Self-Evaluation processes. Pers Soc Psychol Bull, 1995; 21(12):1278–1287; doi: 10.1177/01461672952112005

Quiamzade

, Lalot

. Le Soi: Les Fondamentaux. PUG; 2022.

Sedikides

, Strube

. Self-Evaluation: To thine own self be good, to thine own self be sure, to thine own self be true, and to thine own self be better. In: Advances in Experimental Social Psychology. Elsevier; 1997; pp. 209–269.

Trope

. Self-assessment, self-enhancement, and task preference. J Exp Soc Psychol, 1980; 16(2):116–129; doi: 10.1016/0022-1031(80)90003-7

Trope

. Uncertainty-reducing properties of achievement tasks. J Pers Soc Psychol, 1979; 37(9):1505–1518.

Trope

. Self-Assessment and self-enhancement in achievement motivation. In: Handbook of Motivation and Cognition: Foundations of Social Behavior. ( Sorrentino

, Higgins

. eds) Guilford Press: New York; 1986; pp. 350–378.

Reinecke

, Trepte

. Authenticity and well-being on social network sites: A two-wave longitudinal study on the effects of online authenticity and the positivity bias in SNS communication. Comput Hum Behav, 2014; 30:95–102; doi: 10.1016/j.chb.2013.07.030

Chou

H-TG

, Edge

. “They are happier and having better lives than i am”: The impact of using facebook on perceptions of others’ lives. Cyberpsychol Behav Soc Netw, 2012; 15(2):117–121; doi: 10.1089/cyber.2011.0324

10.

Alicke

, LoSchiavo

, Zerbst

, et al. The person who outperforms Me is a genius: Maintaining perceived competence in upward social comparison. J Pers Soc Psychol, 1997; 73(4):781–789; doi: 10.1037/0022-3514.73.4.781

11.

Collins

. For better or worse: The impact of upward social comparison on self-evaluations. Psychol Bull, 1996; 119(1):51–69; doi: 10.1037/0033-2909.119.1.51

12.

Wheeler

. Motivation as a determinant of upward comparison. J Exp Soc Psychol, 1966; 1:27–31; doi: 10.1016/0022-1031(66)90062-X

13.

Ybema

, Buunk

. Affective responses to social comparison: A study among disabled individuals. Br J Soc Psychol, 1995; 34 (Pt 3)(3):279–292; doi: 10.1111/j.2044-8309.1995.tb01064.x

14.

Tesser

. Toward a self-evaluation maintenance model of social behavior. In: Advances in Experimental Social Psychology. ( Berkowitz

.ed) Academic Press; 1988; pp. 181–227; doi: 10.1016/S0065-2601(08)60227-0

15.

Trope

. Seeking information about one’s ability as a determinant of choice among tasks. J Pers Soc Psychol, 1975; 32(6):1004–1013; doi: 10.1037/0022-3514.32.6.1004

16.

Helgeson

, Mickelson

. Motives for social comparison. Pers Soc Psychol Bull, 1995; 21(11):1200–1209; doi: 10.1177/01461672952111008

17.

Carver

, Scheier

. Control theory: A useful conceptual framework for personality–social, clinical, and health psychology. Psychol Bull, 1982; 92(1):111–135; doi: 10.1037/0033-2909.92.1.111

18.

Scheier

, Carver

. Self-directed attention and the comparison of self with standards. J Exp Soc Psychol, 1983; 19(3):205–222; doi: 10.1016/0022-1031(83)90038-0

19.

Alfasi

. The grass is always greener on my Friends’ profiles: The effect of Facebook social comparison on state self-esteem and depression. Personal Individ Differ, 2019; 147:111–117; doi: 10.1016/j.paid.2019.04.032

20.

McComb

, Vanman

, Tobin

. A meta-analysis of the effects of social media exposure to upward comparison targets on self-evaluations and emotions. Media Psychol, 2023; 26(5):612–635; doi: 10.1080/15213269.2023.2180647

21.

Butzer

, Kuiper

. Relationships between the frequency of social comparisons and self-concept clarity, intolerance of uncertainty, anxiety, and depression. Personal Individ Differ, 2006; 41(1):167–176.

22.

Bäzner

, Brömer

, Hammelstein

, et al. Current and former depression and their relationship to the effects of social comparison processes. Results of an internet based study. J Affect Disord, 2006; 93(1–3):97–103; doi: 10.1016/j.jad.2006.02.017

23.

Wayment

, Taylor

. Self-Evaluation processes: Motives, information use, and self-esteem. J Pers, 1995; 63(4):729–757; doi: 10.1111/j.1467-6494.1995.tb00315.x

24.

Nesi

, Miller

, Prinstein

. Adolescents’ depressive symptoms and subsequent technology-based interpersonal behaviors: A multi-wave study. J Appl Dev Psychol, 2017; 51:12–19; doi: 10.1016/j.appdev.2017.02.002

25.

Ahrens

, Alloy

. Social comparison processes in depression. Health Coping Well-Perspect Soc Comp Theory, 1997:389–410.

26.

Giordano

, Wood

, Michela

. Depressive personality styles, dysphoria, and social comparisons in everyday life. J Pers Soc Psychol, 2000; 79(3):438–451; doi: 10.1037/0022-3514.79.3.438

27.

Wood

, Lockwood

. Social Comparisons in Dysphoric and Low Self-Esteem People. American Psychological Association; 1999; doi: 10.1037/10320-004

28.

Cramer

, Song

, Drent

. Social comparison on Facebook: Motivation, affective consequences, self-esteem, and Facebook fatigue. Comput Hum Behav, 2016; 64:739–746; doi: 10.1016/j.chb.2016.07.049

29.

Frison

, Eggermont

. “Harder, better, faster, stronger”: Negative comparison on facebook and adolescents’ life satisfaction are reciprocally related. Cyberpsychol Behav Soc Netw, 2016; 19(3):158–164; doi: 10.1089/cyber.2015.0296

30.

Riehm

, Feder

, Tormohlen

, et al. Associations between time spent using social media and internalizing and externalizing problems among US youth. JAMA Psychiatry, 2019; 76(12):1266–1273.

31.

Hamaker

, Kuiper

, Grasman

. A critique of the cross-lagged panel model. Psychol Methods, 2015; 20(1):102–116.

32.

Asparouhov

, Hamaker

, Muthén

. Dynamic structural equation models. Struct Equ Model Multidiscip J, 2018; 25(3):359–388; doi: 10.1080/10705511.2017.1406803

33.

Valkenburg

, Beyens

, Meier

, et al. Advancing our understanding of the associations between social media use and well-being. Curr Opin Psychol, 2022; 47:101357; doi: 10.1016/j.copsyc.2022.101357

34.

Diel

, Grelle

, Hofmann

. A motivational framework of social comparison. J Pers Soc Psychol, 2021; 120(6):1415–1430; doi: 10.1037/pspa0000204

35.

Cranford

, Shrout

, Iida

, et al. A procedure for evaluating sensitivity to within-person change: Can mood measures in diary studies detect change reliably? Pers Soc Psychol Bull, 2006; 32(7):917–929; doi: 10.1177/0146167206287721

36.

Muthén

, Muthén

. Mplus. In: Handbook of Item Response Theory Chapman and Hall. CRC; 2017; pp. 507–518.

37.

Hamaker

, Asparouhov

, Brose

, et al. At the frontiers of modeling intensive longitudinal data: Dynamic structural equation models for the affective measurements from the COGITO study. Multivariate Behav Res, 2018; 53(6):820–841; doi: 10.1080/00273171.2018.1446819

38.

McNeish

, Hamaker

. A primer on two-level dynamic structural equation models for intensive longitudinal data in Mplus. Psychol Methods, 2020; 25(5):610–635.

39.

Mettee

, Riskind

. Size of defeat and liking for superior and similar ability competitors. J Exp Soc Psychol, 1974; 10(4):333–351; doi: 10.1016/0022-1031(74)90030-4

40.

Brown

, Dutton

. Truth and consequences: The costs and benefits of accurate self-knowledge. Pers Soc Psychol Bull, 1995; 21(12):1288–1296; doi: 10.1177/01461672952112006

41.

Eaton

, Anthony

, Gallo

, et al. Natural history of diagnostic interview schedule/DSM-IV major depression: The Baltimore epidemiologic catchment area follow-up. Arch Gen Psychiatry, 1997; 54(11):993–999.

42.

Lewinsohn

, Clarke

, Seeley

, et al. Major depression in community adolescents: Age at onset, episode duration, and time to recurrence. J Am Acad Child Adolesc Psychiatry, 1994; 33(6):809–818; doi: 10.1097/00004583-199407000-00006

43.

Spijker

, Graaf

, Bijl

, et al. Duration of major depressive episodes in the general population: Results from the Netherlands Mental Health Survey and Incidence Study (NEMESIS). Br J Psychiatry, 2002; 181(3):208–213; doi: 10.1192/bjp.181.3.208

44.

Keller

, Lavori

, Mueller

, et al. Time to recovery, chronicity, and levels of psychopathology in major depression: A 5-year prospective follow-up of 431 subjects. Arch Gen Psychiatry, 1992; 49(10):809–816; doi: 10.1001/archpsyc.1992.01820100053010

45.

Campbell

, Trapnell

, Heine

, et al. Self-concept clarity: Measurement, personality correlates, and cultural boundaries. J Pers Soc Psychol, 1996; 70(6):1114–1114; doi: 10.1037/0022-3514.70.6.1114

46.

Teasdale

. Psychological treatments for depression: How do they work? Behav Res Ther, 1985; 23(2):157–165; doi: 10.1016/0005-7967(85)90024-5

47.

Baglioni

, Spiegelhalder

, Feige

, et al. Sleep, Depression and Insomnia – A Vicious Circle? Cpsr, 2014; 10(3):202–213.

48.

Kirchner

, Schummer

, Krug

, et al. How social rejection expectations and depressive symptoms bi-directionally predict each other – A cross-lagged panel analysis. Psychol Psychother, 2022; 95(2):477–492; doi: 10.1111/papt.12383

49.

Aubry

, Quiamzade

, Meier

. Depressive symptoms and upward social comparisons during Instagram use: A vicious circle. Personal Individ Differ, 2024; 217:112458.

50.

Meier

, Gilbert

, Börner

, et al. Instagram inspiration: How upward comparison on social network sites can contribute to well-being. J Commun, 2020; 70(5):721–743; doi: 10.1093/joc/jqaa025

51.

Blanton

, Buunk

, Gibbons

, et al. When better-than-others compare upward: Choice of comparison and comparative evaluation as independent predictors of academic performance. J Pers Soc Psychol, 1999; 76(3):420–430; doi: 10.1037/0022-3514.76.3.420

52.

Davidai

, Deri

, Gilovich

. There must be more to life than this: The impact of highly-accessible exemplars on self-evaluation and discontent. Self Identity, 2021; 20(1):72–93; doi: 10.1080/15298868.2020.1779121

53.

Schmiedek

, Neubauer

. Experiments in the Wild: Introducing the Within-Person Encouragement Design. Multivar Behav Res; 2020.

54.

Substance Abuse and Mental Health Services Administration. Key Substance Use and Mental Health Indicators in the United States: Results from the 2023 National Survey on Drug Use and Health. Center for Behavioral Health Statistics and Quality, Substance Abuse and Mental Health Services Administration; 2024.