Social networking sites (SNSs) continue to grow in popularity, playing an increasingly central role in users’ lives (Pew Research Center, 2024; Villanti et al., 2017). This trend is concerning, as a growing body of research points to a connection between SNSs use and declines in various aspects of mental health and quality of life (e.g., Cunningham et al., 2021; Purba et al., 2023; Vannucci et al., 2020). One mental health outcome frequently linked to SNSs use is self-esteem, a psychological construct that is fundamental to wellbeing (Mann et al., 2004). The suspected effect of SNSs use on self-esteem may stem from the myriad opportunities for upward social comparisons-comparing oneself to someone perceived as superior-these platforms offer to users (Appel et al., 2016). Given the importance of this societal issue, there is a pressing need for more empirical data clarifying the role of social comparisons in the relationship between SNSs use and self-esteem.

SNSs are web-based platforms that enable users to (1) create a public or semi-public profile within a defined network, (2) establish a list of connections with other users, and (3) interact with their own and others’ connections on the platform (Boyd and Ellison, 2007). Users engage with SNSs for various purposes, including maintaining interpersonal relationships, relaxing, and expressing opinions (Whiting and Williams, 2013). Importantly, users also report using SNSs both to share glimpses of their lives and to keep track of what others are doing (Whiting and Williams, 2013). Beyond these uses, some studies suggest that SNSs can foster social support, enhance a sense of belonging, and provide opportunities for self-expression and identity exploration, which may positively influence users’ wellbeing (Manago et al., 2012; Nabi et al., 2013).

Over the last decade or so, SNSs have become increasingly integrated into users’ daily lives (Pew Research Center, 2024; Villanti et al., 2017). A recent survey reported that 72% of Americans use social media, with usage rising to 84% among 18–29-year-olds (Pew Research Center, 2021). Similarly, 77.6% of Canadians are active on at least one SNSs, with this figure increasing to 90% among those aged 35 and younger (Gouvernement du Canada, 2021). Instagram now boasts over 2.3 billion active users (Ruby, 2023), Facebook has surpassed 2.9 billion (Baron, 2023), and TikTok, launched in 2016, has already reached over 1 billion monthly active users (Seignol de Swarte, 2022). This steady rise in SNSs use is a matter of concern. While the effects of SNSs use are unlikely to be entirely negative (e.g., Smith et al., 2021; Spottswood and Wohn, 2020), a substantial body of research suggests a strong association with poorer mental health outcomes. Meta-analyses of correlational studies consistently indicate a positive relationship between SNSs use and depressive symptoms (Cunningham et al., 2021; Ivie et al., 2020; Liu et al., 2022; Shannon et al., 2022). Furthermore, increased time spent on SNSs has been linked to heightened feelings of loneliness (Liu and Baumeister, 2016; Zhang et al., 2022), a greater incidence of eating disorders (Holland and Tiggemann, 2016; Sharma and Vidal, 2023), and lower life satisfaction (Huang, 2017). Notably, SNSs use is also associated with reduced self-esteem (Liu and Baumeister, 2016; Saiphoo et al., 2020). This is especially concerning, as self-esteem is often considered a key determinant of mental health, potentially affecting a broad spectrum of mental health outcomes (Mann et al., 2004).

Self-esteem refers to an individual’s overall positive or negative attitude toward themselves (Rosenberg et al., 1995). Researchers commonly distinguish between global and domain-specific self-esteem. Global self-esteem reflects a person’s general sense of self-worth, while domain-specific self-esteem pertains to self-assessments in particular areas, such as physical appearance or academic achievement (Donnellan et al., 2011). In the context of SNSs, where appearance is often highlighted (Popat and Tarrant, 2023), physical self-esteem appears particularly vulnerable. Indeed, on these platforms, individuals frequently present an idealized version of themselves, sharing polished and highly curated images (Manago et al., 2008; Young et al., 2022).

Both global and physical self-esteem appear to play important roles in mental health and wellbeing. Meta-analytic evidence of correlational studies indicates medium-sized correlations between global self-esteem and both life satisfaction (Diener and Diener, 1995) and wellbeing (Muris and Otgaar, 2023). Longitudinal studies have further demonstrated that self-esteem serves as a predictor of depressive symptoms (Sowislo and Orth, 2013), eating disorders (Colmsee et al., 2021), suicidal behavior (Soto-Sanz et al., 2019), self-harm (Junker et al., 2019), and peer victimization (van Geel et al., 2018). Additionally, higher levels of self-esteem have been reported to predict academic achievement (Valentine et al., 2004) and the quality of social relationships (Harris and Orth, 2020). Correlational studies also suggest that global self-esteem is negatively associated with loneliness (Mahon et al., 2006), substance abuse (Martínez-Casanova et al., 2024), and risky sexual behavior (Ahn and Yang, 2023). As for physical self-esteem, related constructs such as body image and body appreciation have been the focus of extensive research. A meta-analysis by Linardon et al. (2022) found that body appreciation is positively correlated with wellbeing and negatively correlated with depression and eating disorders. Additionally, body image has been reported to be negatively associated with self-harm (Oktan, 2017), as well as suicidal ideation and behavior (Angelakis et al., 2016).

Given the importance of self-esteem for mental health, the rise in SNSs use over the last decade is concerning. Indeed, a substantial body of evidence demonstrates that SNSs use is negatively associated with both global (Andreassen et al., 2017; Liu and Baumeister, 2016; Saiphoo et al., 2020) and physical self-esteem (Fardouly and Vartanian, 2016; Saiphoo et al., 2020).

According to social comparisons theory (Festinger, 1954), humans possess a fundamental drive to assess their opinions and abilities. Individuals use social comparisons as a mechanism to enhance their self-understanding by evaluating themselves in relation to others (Festinger, 1954). Similarly, ranking theory posits that social comparisons is an evolutionarily ancient process allowing individuals to assess their position in the social hierarchy (Gilbert, 1992). Social comparisons can take three primary forms: downward, lateral, and upward comparisons (Guyer and Vaughan-Johnston, 2020). Downward comparisons occur when individuals evaluate themselves against others they perceive as inferior on a given dimension, while lateral comparisons involve evaluations against those perceived as similar or equal. Upward comparisons, by contrast, involve assessing oneself against others perceived as superior.

The highly curated and idealized content prevalent on SNSs encourages users to engage in upward social comparisons, where they compare themselves to seemingly superior others (Henriques and Patnaik, 2020). Studies have consistently shown that frequent use of SNSs is linked to an increase in these upward comparisons (Gomez et al., 2022; Schmuck et al., 2019; Vogel et al., 2014). Drawing on social comparison theory, these upward comparisons can be expected to negatively affect self-perceptions, particularly self-esteem. From the perspective of ranking theory, such comparisons may also threaten one’s perceived social standing, thereby increasing psychological vulnerability. Together, these theories inform the central hypothesis of the present research: that upward comparisons mediate the relationship between SNSs use and reduced self-esteem.

Evidence suggests that upward comparisons can negatively affect both global and physical self-esteem. Research on this relationship includes studies that examine direct effects, showing that upward comparisons independently harm self-esteem (e.g., Ho et al., 2016; Rodgers et al., 2015; McComb and Mills, 2022; McComb et al., 2023), as well as studies exploring mediation pathways, where upward comparisons act as a mediator between other variables, such as social media use and self-esteem (e.g., Wang et al., 2017; Schmuck et al., 2019; Vogel et al., 2014). This effect is well-documented across a range of research designs, including correlational (Ho et al., 2016; Wang et al., 2017), longitudinal (Rodgers et al., 2015; Schmuck et al., 2019), and experimental studies (McComb and Mills, 2022; McComb et al., 2023; Vogel et al., 2014). The consistent findings across these methodologies highlight the robust nature of this effect on both forms of self-esteem. Given that upward social comparisons are positively associated with SNSs use and linked to lower self-esteem, it may be a key mechanism mediating the association between higher SNSs use and reduced self-esteem—a hypothesis supported by current research.

Indeed, several studies have examined the mediating role of social comparisons in the relationship between SNSs use and global self-esteem (e.g., Midgley et al., 2021; Vogel et al., 2014; Wang et al., 2017). For instance, Vogel et al. (2014) found that the negative impact of Facebook use on self-esteem is mediated by the tendency to engage in upward social comparisons on SNSs, while downward comparisons had no significant effect. Similarly, Midgley et al. (2021) exposed one group to a SNSs context and another to a non-social media context. Participants in the SNSs condition reported lower self-esteem by the end of the study compared to those in the non-SNSs condition. Their findings indicated that this relationship between exposure to SNSs and self-esteem was mediated by an increased likelihood of making upward comparisons, consistent with the results of Vogel et al. (2014).

Other studies have investigated the mediating role of social comparisons on SNSs in the relationship between SNSs use and physical self-esteem (e.g., Scully et al., 2020; Tiggemann and Zaccardo, 2015; Wang et al., 2017). Overall, these studies indicate that upward social comparisons mediate the relationship between SNSs use and physical self-esteem. For example, Scully et al. (2020) found, in their correlational study, that the tendency to engage in upward comparisons on SNSs mediates the negative relationship between social media use and body satisfaction. Similarly, Tiggemann and Zaccardo (2015), in an experimental study, assigned participants to one of two groups: one exposed to ‘fitspiration’ images on social media, designed to evoke body-related upward comparisons, and the other exposed to travel images. Those exposed to fitspiration images reported lower appearance-based self-esteem after the task, with upward appearance-based comparisons mediating the negative relationship between exposure to fitspiration content and self-esteem.

Upward social comparisons and its suspected negative effects on self-esteem may vary based on both context and individual characteristics. Key questions remain: under what conditions do these comparisons lead to detrimental outcomes, and how do these processes unfold? While research in this area is still emerging, several factors have been proposed that might influence the relationship between SNSs use, upward comparisons, and both global and physical self-esteem. Variables such as the extremity—defined as the perceived gap between the individual and the comparison target—of social comparisons, the specific SNSs platform, social feedback (i.e., responses or evaluations from others), and gender all seem to play a role. For instance, Midgley et al. (2021) suggest that extreme upward comparisons, where the comparison targets are perceived as significantly superior, can negatively mediate the relationship between SNSs use and self-esteem. Nonetheless, research on the extremity of social comparisons in the context of SNSs use and its impact on mental health remains limited. Similarly, the potential differences in mental health effects and underlying mechanisms across platforms remain largely underexplored, despite their likely significance. Limniou et al. (2022) compared Facebook and Instagram and found that Instagram users exhibited more problematic usage patterns, suggesting that Instagram may foster more addictive behaviors. Because Instagram emphasizes visual content like photos and videos, some researchers suggest that it may influence self-esteem and wellbeing differently than platforms such as Facebook (Faelens et al., 2021). Social feedback is another factor gaining attention in the context of SNSs and mental health. Positive social feedback (e.g., receiving more likes) has been associated with enhanced self-esteem (Burrow and Rainone, 2017; Marengo et al., 2021). However, conflicting evidence suggests that this same feedback may also be linked to reduced self-esteem (Limniou et al., 2022). Lastly, research focusing on adolescence reveals that social media’s impact on mental health tends to be more pronounced in females (Barthorpe et al., 2020; Kelly et al., 2019; Twenge et al., 2022).

Initial studies examining the impact of SNSs largely emphasized a straightforward, linear connection between SNSs use and wellbeing, overlooking the more intricate dynamics at play (Kross et al., 2021). It has become increasingly clear, however, that the influence of SNSs varies based on the context, with outcomes ranging from positive to harmful (Cingel et al., 2022; Krause et al., 2019). While there has been a growing interest in uncovering the mechanisms that link SNSs usage to mental health outcomes (Meier and Johnson, 2022; Verduyn et al., 2020), this field of research is still in its infancy. One major gap in the current literature is the limited understanding of the specific mechanisms through which SNSs exert their influence on mental wellbeing (Kross et al., 2021). Moreover, although an increasing amount of data suggests the mediating role of upward social comparisons in the relationship between media usage and both global (e.g., Midgley et al., 2021; Vogel et al., 2014; Wang et al., 2017) and physical self-esteem (e.g., Scully et al., 2020; Tiggemann and Zaccardo, 2015; Wang et al., 2017; Vogel et al., 2014), no research has investigated this role for both types of self-esteem in the same participants. Examining global and physical self-esteem at the same time allows for a more nuanced understanding of how social comparisons impact distinct domains of self-concept. Theoretically, this approach acknowledges that self-esteem is a multidimensional construct, with different domains potentially responding differently to upward comparisons. Methodologically, it ensures within-subject comparability, thereby reducing confounds and enabling more precise modeling of individual differences. Investigating these two concepts simultaneously would then allow for a better understanding of which aspect(s) of self-esteem is/are more affected by the use of these platforms. Lastly, very few studies in this field focus on multiple SNSs simultaneously. For example, the majority of studies either focus solely on Facebook (Gonzales and Hancock, 2011; Tazghini and Siedlecki, 2013; Vogel et al., 2014) or on Instagram (Dion, 2016; Hill and Denman, 2016; Paramboukis et al., 2016). However, the nature of the content found on each of these applications differs greatly. Instagram is primarily image-based and often promotes idealized self-presentation, while Facebook includes more textual content, mixed media, and interactions grounded in existing offline relationships (Sheldon and Bryant, 2016). Additionally, Instagram tends to be more popular among younger users and is associated with appearance-focused comparison, whereas Facebook has a broader age demographic and is linked to different social comparison processes (Huang, 2017). Therefore, the consequences of these different SNSs may not be the same from one application to another, as this meta-analysis has shown (Akkaş and Turan, 2024). Investigating two different SNSs would thus better distinguish the individual impacts of each of these platforms and can serve as a basis for issuing more tailored recommendations.

Therefore, to fill these gaps in the literature, we conducted two studies designed to enhance our understanding of the mechanisms through which social media use affects mental health. By examining how upward social comparisons impact different domains of self-esteem and symptoms of depression across multiple platforms, these studies offer insights that may inform the development of safer, more mindful social media practices, as well as future digital literacy and mental health interventions. In study 1, which took place during the COVID-19 pandemic, we examined the mediating role of exposure to social comparisons (upward and downward) in the relationship between SNSs usage and self-esteem (both global and physical). In study 2, which was pre-registered, we examined the mediating role of perceived exposure to social comparisons (upward and downward) in the relationship between SNSs usage and (i) self-esteem (global, physical, and academic) and (ii) symptoms of depression. We also investigated the role of other variables: the extremity of social comparisons, the specific SNSs platform (Instagram versus Facebook), social feedback (its valence and quantity), and gender. This study has been approved by the Ethics Committee CEREP (Comité d’éthique de la recherche en éducation et en psychologie) of the Université de Montréal (CEREP-21-081-D).

Early research on the effects of SNSs predominantly focused on the linear relationship between SNSs use and wellbeing, often neglecting the complex nature of their interaction (Kross et al., 2021). While there has been a growing effort to elucidate mechanisms linking SNSs use to mental health (Meier and Johnson, 2022; Verduyn et al., 2020), this endeavor remains in its early stages. A key gap in current knowledge is our limited mechanistic insight into how SNSs impact us (Kross et al., 2021).

The aim of these two studies was to gain a deeper understanding of the mechanisms mediating the relationship between SNSs use and self-esteem, with a particular focus on social comparisons, which we believe is central to these effects. More specifically, our results highlighted two key aspects of upward social comparisons—exposure and extremity—as critical factors linking SNSs use to affective outcomes such as self-esteem and depression. Indeed, the perceived exposure to upward comparisons negatively mediated the relationship between SNSs use and self-esteem, as well as the relationship between SNSs use and depressive symptoms. However, the mediation path was positive for extremity, suggesting that SNSs usage is associated to less extreme upward social comparisons which reduces the direct negative link found between the extremity of upward comparisons and affective outcomes. The relations between upward social comparisons—both in terms of exposure and extremity—and detrimental effects on self-esteem and emotional wellbeing were the most consistent of our results. These findings underscore the critical impact of various dimensions of upward social comparisons on SNSs, highlighting their role in the decline of self-esteem and the increase in depressive symptoms. Both the exposure to upward comparisons and their extreme nature are central contributors to the emotional consequences of social media use (Gerber et al., 2018; Schmuck et al., 2019; Irmer and Schmiedek, 2023). These results support previous research suggesting that frequent engagement in upward comparisons may exacerbate negative emotional outcomes, including lower self-esteem and increased depressive symptoms, particularly when the comparisons are extreme (Midgley et al., 2021).

These results are in line with predictions from ranking theory (Gilbert, 1992) and the existing literature (Gerber et al., 2018; Schmuck et al., 2019; Irmer and Schmiedek, 2023), which both suggest that the behavioral and affective consequence of upward social comparisons, which were adaptative during most of our time as a species, can in some contexts be responsible for various negative emotional consequences, including reduced self-esteem and the development of depressive symptoms. Indeed, our findings are in line with the mismatch hypothesis (Lim and Tan, 2024), which suggests that SNSs hijack our evolved social comparisons mechanisms by offering (too) abundant opportunities for upward comparisons. This hypothesis argues that SNSs inundate users with idealized content, which can lead to negative emotional outcomes.

The results from study 1, which revealed a positive relationship between SNSs use and self-esteem, suggest that additional mechanisms may be influencing this dynamic. While this paper focuses primarily on the potential negative effects of SNSs use, there is also evidence of its positive impact, such as fostering inspiration (Meier and Schäfer, 2018) and enhancing perceptions of social support through increased self-disclosure (Huang, 2016). Other positive psychological mechanisms may buffer the negative effects of social comparisons on SNSs. For instance, engaging with content that promotes self-improvement or motivation, such as fitness goals, may encourage upward comparisons that are aspirational rather than harmful, especially when users perceive the content as attainable. Social media can also serve as a space for positive identity reinforcement—for example, sharing personal milestones, achievements, or cultural identity can enhance self-concept clarity and pride being (Manago et al., 2012; Nabi et al., 2013). Furthermore, receiving positive feedback or peer validation through likes and supportive comments may enhance social capital, which is known to strengthen self-esteem and wellbeing (Boyd and Ellison, 2007). Personality traits also likely play a role in how users interpret and respond to social comparisons. For instance, individuals with high self-compassion or emotional regulation skills may be better equipped to buffer the negative impact of upward comparisons by reframing them or reducing self-criticism (Neff and Tóth-Király, 2022). Future research should examine how these protective factors interact with social comparison processes to better understand why some individuals benefit from SNS use while others are negatively affected. Ultimately, recognizing this complexity does not diminish the potential harm of excessive upward comparisons, but rather highlights the importance of understanding the nuanced interplay between social media use, individual differences, and psychological outcomes.

In both studies, when we added gender as a mediator in our models, no significant interaction effects were observed, indicating that the associations between social media use, social comparisons, and self-esteem did not differ significantly between men and women. This finding contradicts some previous research that has reported stronger negative effects of social comparison for women (Barthorpe et al., 2020; Kelly et al., 2019; Twenge et al., 2022). However, several factors may help explain this discrepancy. First, most earlier studies on social media and body image focused primarily on women, often excluding men altogether or underrepresenting them. In contrast, our sample included a more balanced gender representation, reflecting current trends showing increased male engagement with image-based platforms like Instagram. Second, shifts in digital culture may be narrowing the gender gap in social comparison experiences. With the rise of trends such as “fitspiration” and content showcasing muscular male bodies (Tiggemann and Anderberg, 2020), men are increasingly subject to idealized body standards, which may lead to similar psychological outcomes as observed in women. Indeed, recent research has found that men exposed to upward comparison content also report lower self-esteem (Fatt et al., 2019; Tiggemann and Anderberg, 2020; Barron et al., 2021), supporting the idea that men are not immune to appearance-based pressures online, which aligns with our findings. Finally, it is possible that the self-esteem measures used in our study did not fully capture gender-specific concerns. For instance, while the global and physical self-esteem scales offer valuable insights, they may not be sensitive to domains where men experience the most pressure (e.g., muscularity, dominance, athleticism). Future research could benefit from using gender-sensitive measures or exploring body image constructs specific to men, such as drive for muscularity or self-objectification.

One strength of our research was its timing, with study 1 conducted during the COVID-19 pandemic and study 2 taking place afterward. This allowed us to capture potential changes in the relation between user behavior and outcomes during the pandemic. For instance, in study 1, we observed a positive association between Instagram use and global self-esteem, a result not replicated in study 2 or supported by the existing literature. This discrepancy could be attributed to the increased socialization benefits provided by SNSs during the pandemic, which may have temporarily mitigated its usual negative effects. The findings from study 2, which aligned more closely with the established literature, suggest that the behaviors seen in study 1 were more situational and pandemic-related. While our findings in study 1 indicated a small positive association, it’s important to note that other studies found the relationship between SNSs use and mental health remained negative or even increased during the pandemic (Draženović et al., 2023; Haddad et al., 2021; Marciano et al., 2022). Additionally, study 1 found no significant link between Instagram use and physical self-esteem. As noted earlier, the questionnaire used in study 1 assessed participants’ satisfaction with individual body parts, which may have been less suited to our research goals. To address this, we used a different physical self-esteem questionnaire in study 2 which focusses on a more general appreciation of body appearance. This change in the measurement tool may also have contributed to the differences in results observed between the two studies regarding physical self-esteem. Another strength of this study is its comparative analysis of the effects of two different SNSs platforms, an area that has been relatively understudied. Our findings revealed that similar effects were present on both Facebook and Instagram, despite subtle differences in the nature of their content (Limniou et al., 2022). As noted earlier, these similarities may be partly attributed to the growing convergence of features between platforms like Facebook and Instagram, particularly since both are owned by Meta. However, the study also highlighted some important distinctions, such as differences in usage patterns. Young adults appear to be shifting toward Instagram as their primary SNSs, emphasizing the need for future research to focus more closely on this platform. Finally, including academic self-esteem as a control dependent variable enabled us to provide some of the first evidence that the effects of upward social comparisons on SNSs seems to be specific to the type of content generally present on these platforms.

Although this study provides valuable insights, several limitations should be noted. First, the correlational design restricts our ability to draw definitive conclusions about the causal role of upward social comparisons in shaping affective outcomes. Experimental studies would be better suited for establishing causality in these relationships. Second, the online data collection limited control over the accuracy and reliability of participants’ responses. While we implemented safeguards such as attention checks and filtering responses to open-ended questions, ensuring full data reliability in an online setting remains challenging. In-person studies or platforms that allow closer monitoring of participant engagement could help improve data quality in future research. Third, our analyses did not control for potentially important individual difference variables such as body image dissatisfaction, narcissism, or baseline mental health status, which could influence both social comparison processes and self-esteem outcomes. Including such variables in future research would allow for a more precise understanding of the unique effects of upward social comparisons. Finally, our study did not explore the potential influence of demographic or cultural diversity on social media behavior. Research suggests that individuals from different cultural backgrounds may engage with SNS platforms in distinct ways (Vasalou et al., 2010; Gordesli et al., 2024)—for example, using Instagram for self-expression versus Facebook for maintaining family ties. These patterns may influence both the frequency and content of social comparisons, as well as their emotional impact. The generalizability of our findings may therefore be limited, particularly in culturally diverse populations. Future studies should examine cross-cultural differences in social media use and comparison processes to gain a more nuanced understanding of these relationships. Exploring the effects of various SNSs platforms (TikTok, SnapChat, etc.) with diverse user bases and content types may also provide deeper insights into the mechanisms underlying these observed outcomes.

This study highlights perceived exposure to and extremity of upward social comparisons as key mechanisms linking SNSs use to impacts on self-esteem and depressive symptoms. Across two studies, we found that while perceived exposure to upward comparisons consistently mediated negative effects on mental health, extremity of comparisons was uniquely associated with lower self-esteem, suggesting that not all comparisons are equally harmful. Notably, our findings also showed that the frequency of SNSs use was not uniformly detrimental, and in some contexts—particularly during the COVID-19 pandemic—it was associated with higher self-esteem, possibly due to increased social connection or positive feedback. Given these findings, raising public awareness about the risks of excessive upward comparisons-particularly extreme ones—and promoting healthier and more intentional social media use is essential.