Toxic positivity intentions: an image management approach to upward social comparison and false self-presentation

Abstract

Media users are often overwhelmed with excessive amounts of positivity on social media. While a healthy positive outlook acknowledges both the positive and negative aspects of life, a growing term in popular media referring to toxic positivity suggests users leverage positivity and avoid negativity for personal gain. Employing a two-study, multi-method design, Study 1 is framed within the impression management literature to conceptually and operationally define perceived toxic positivity intentions (TPI) through qualitative and quantitative measures. Study 2 is framed within the bandwagon and descriptive norm literature to investigate the role of perceived TPI in upward social comparisons and digital self-presentation behaviors. Results of Study 1 indicate a five-factor solution of perceived TPI on social media. Results of Study 2 suggest the relationship between (most) dimensions of perceived TPI and false social media self-presentations is significantly mediated by upward social comparison.

Presenting an edited version of oneself, often highly subjective and not completely truthful, is a central theme in self-presentations on social media (Chua & Chang, 2016; Mun & Kim, 2021; Prins & Wellman, 2023; Rousseau, 2021; Yau & Reich, 2019). Defined as social media self-presentation, users edit photos and strategically present a positive narrative of themselves to their audience for personal gain. Consider how users manage impressions through social media posts that highlight how beautiful, smart, and beyond expectations their family is, how great their partner looks, or how their diet has worked miracles. Sound familiar? Although it is natural to lean toward a positive bias when portraying oneself on social media (Rousseau, 2021; Yau & Reich, 2019), these types of posts do not represent a complete narrative or use positivity in a healthy way, which results in a positivity bias across social media and negative effects for users who are exposed to the positivity (Putra et al., 2023).

Popular media has defined positive communication avoiding negative emotions or experiences as toxic positivity (Solaris, 2019), and the phenomenon is increasingly being explored across academic literature (Pangestu et al., 2022; Prins & Wellman, 2023; Putra et al., 2023; Quinto et al., 2021). Research reveals that toxic positivity is used by social media users as a self-presentation strategy to suppress negativity by only conveying positivity (Prins & Wellman, 2023). Yet, in current research on toxic positivity on social media, perceptions of this self-enhancement strategy and its impact on other users remain understudied. As such, research has not yet addressed what intentions people perceive as toxic positivity and what role the perceptions play in users perceiving their life to be less than others. That is, users could see toxic positivity and think it is fake or is intentionally being malicious. In contrast, a user could see these posts and think the post is prideful, inspiring, or important. Since the difference in perceptions of posts could manifest from the perceived intention of sharing, it is important to investigate the intentions users perceive others to have when using toxic positivity.

Considering the outcome of perceived intentions of posting toxic positivity on social media, research suggests that when individuals perceive they are being deceived by a false self-presentation shared by others, the individuals are likely to reciprocate the deception (Markowitz & Hancock, 2018). In a social media context, where exposure to positively biased posts highlighting the best parts of life is associated with comparisons (Chae, 2018; Chua & Chang, 2016; Pedalino & Camerini, 2022), the current study leverages the descriptive norm literature (Bergquist & Nilsson, 2019; Cialdini, 2005) and positions upward social comparison as an explanatory process by which the perceived intentions behind sharing toxic positivity influence the sharing of one’s own false self-presentation.

Purpose of the present study

To move the literature on toxic positivity as a self-presentation strategy forward, this study explicates the construct of perceived toxic positivity intentions (TPI) and examines the implications of perceived TPI on future self-presentation behaviors. Specifically, this study examines the broad perceptions of intentions behind posting toxic positivity across all social media. This study employs a two-study design that investigates perceptions of posting toxic positivity on social media (Study 1) and empirically investigates the impact of perceiving TPI on false self-presentation behaviors through upward social comparisons (Study 2). Study 1 leverages both qualitative and quantitative methodology to outline, conceptualize, and operationalize the concept of perceived TPI. Then, Study 2 extends the literature by uniquely demonstrating the impact of perceived TPI on social media behaviors through upward social comparison.

Study 1: Literature review

Impression management

Impression management research examines why and how individuals influence perceptions that others form. The underlying principles of impression management were first defined by Goffman (1949) through the introduction of self-presentation processes in social interactions. Generally, people manage impressions to achieve personal goals, and to do so, it often includes managing a positive and socially desirable image among others. Maintaining this favorable image, and controlling perceptions that others form about oneself, is referred to as self-presentation (Goffman, 1949). While current literature often uses the terms impression management and self-presentation interchangeably, self-presentation refers specifically to managing information and impressions made about oneself.

Self-presentations are often complex as individuals share multiple versions of themselves, such as their real self (sharing authentic feelings), their ideal self (sharing ideal attributes such as aspirations), and their false self (sharing something that is not true), all of which could be presented simultaneously across social media (Michikyan et al., 2015). This is especially true for adolescents and emerging adults who spend time experimenting with their identities on social media (Yang & Bradford Brown, 2016). Self-presentations further vary in the breadth (amount), depth (level of intimacy), valence (positivity), authenticity (accuracy), and intentionality of the personal information being shared (Yang & Bradford Brown, 2016; Yang et al., 2017).

Many self-presentations are motivated by personal gain (Yau & Reich, 2019), where social media users are selective in the images and information they post for others to see (Fox & Vendemia, 2016; Lo & McKercher, 2015; Rousseau, 2021) and craft their image to be more positive (Prins & Wellman, 2023; Saha et al., 2021; Vogel & Rose, 2016), visually enhanced (Chua & Chang, 2016; Fox & Vendemia, 2016; Rousseau, 2021), and sometimes even false (Gil-Or et al., 2015). It is important to note, however, that research suggests individuals often form a negative perception of self-enhancement strategies (Hong et al., 2020; Lafrenière et al., 2016). As such, it is important to investigate the perceptions of toxic positivity as a self-presentation tactic. It is important to explore this specific strategy, and the intentions behind it, as there may be negative implications for the user exposed (Brooks et al., 2019; Quinto et al., 2021).

Toxic positivity on social media

Toxic positivity is defined by Psychology Today as “the act of avoiding, suppressing, or rejecting negative emotions or experiences” (n.d.). Popular media interpretations of toxic positivity refer to the phenomenon as the demand for a positive attitude (Scott, 2020; Solaris, 2019). Previous academic publications have explored toxic positivity as forcing others to have a positive mindset (Pangestu et al., 2022; Putra et al., 2023; Quinto et al., 2021) and as avoiding negativity when posting on social media (Prins & Wellman, 2023). Toxic positivity is seen through influencers and content creators as they share their own success stories as advice that others should follow to be successful, often attributing such success to having a positive outlook (Bosveld, 2021). For instance, during the 2019 coronavirus pandemic (COVID-19), fitness coaches often framed the negativity of the virus positively when posting to Instagram, such as being “grateful for home workouts” (Prins & Wellman, 2023). Here, influencers were avoiding the negativity of the global pandemic, and instead, used positivity to lift themselves up on social media. Toxic positivity is also experienced through the use of deception when people only share positive information (Vogel & Rose, 2016; Yau & Reich, 2019), heavily edited photos (Fox & Vendemia, 2016; Lo & McKercher, 2015; Rousseau, 2021), or information that is not true about themselves (Mun & Kim, 2021)––all of which focus on the best parts of oneself while avoiding any negative aspects, or even crafting a fake version altogether.

There is no doubt that individuals are more likely to post about their positive life events that represent a gain for themselves and avoid sharing the negative aspects of their life (Saha et al., 2021). Although positivity and optimism are considered beneficial in self-presentations to form a positive impression, this is also where it can become toxic to other users—when social media users, and their posts, suppress the difficult and challenging aspects of life, or when positivity is used to push others down or to lift oneself up by deception (Brooks et al., 2019; Pangestu et al., 2022; Prins & Wellman, 2023). Previous research indicates that individual differences in processing positive social media posts lead users to react in different ways, especially through the process of comparison (de Vries et al., 2018). This means that if individuals have different emotional outcomes (Pangestu et al., 2022) and different processing tendencies (de Vries et al., 2018), the excessively expressed positive information and enhanced images can be harmful to individuals depending on how they perceive the intentions of the post. For instance, a social media post about a successful weight loss journey could be viewed by one user as bragging and by another user as being informative about effective dieting strategies, and one user may be more negatively impacted than the other.

Based on the literature outlined, Study 1 leans on the self-presentation literature to understand toxic positivity through the development of a perceived TPI measure. To do so, Study 1 of this research focuses on people’s perceptions of the intentions behind using toxic positivity. In doing so, the development of a perceived TPI measure provides a unique contribution to the toxic positivity literature, which also extends motivation and impression management literature, by offering a more nuanced focus toward perceptions of self-enhancement strategies through an individual difference perspective.

Conceptualization and operationalization of TPI

Based on the popular press interpretations and academic literature presented, the current study offers an overarching definition for the concept of toxic positivity within social media. Namely, toxic positivity on social media refers to overly positive posts lacking in negativity. Consistent with a media effects perspective, this process occurs through repeated exposure to positive and idealized posts, images or positive comments. As scholarly research on toxic positivity is limited, this research aims to gain a fundamental understanding of the perceptions that users have toward this type of self-enhancement strategy by broadly exploring toxic positivity as overly positive social media posts that avoid negativity. That is, it is not assumed that positive posts avoiding negativity would all be perceived as having toxic intentions, but rather, it allows the users to define the toxicity through their perceived intentions of the post. Thus, the current study seeks to explicate an operational definition that outlines the perceived intentions behind posting toxic positivity on social media as a self-presentation strategy––referred to as perceived TPI. To do so, Study 1 employs open-ended responses to develop a measure of perceived TPI, defined as the intention of positive posting void of negative emotions or experiences as a social media self-presentation strategy. The following research question is addressed in Study 1:

RQ1: How do social media users perceive intentions to post toxic positivity on social media?

Study 1: Method

Overview

Five stages were involved in a mixed methods approach to scale development and validation (Zhou, 2019): (1) qualitative investigation of the construct (perceived TPI), (2) generation of the item pool through open-ended responses, (3) reduction of items, (4) administration of the final scale items, and (5) item purification and validation through exploratory factor analysis (EFA) for evidence of construct validity. It is important to note that each independent expert reviewer (in Stage 3) has published research on social media use and effects through peer-reviewed academic journals. The final scale items can be found in Appendix Table A1.

Qualitative investigation

Sample

In exchange for extra credit, undergraduate college students at a large southwestern university in the United States were invited to participate in an online survey, which was designed and administered using Qualtrics. These data consist of 239 respondents. Of these respondents, 73.6% were female, 25.5% male, 0.4% identified as non-binary/third gender, and 0.4% indicated “other.” Regarding ethnicity, 54.8% were Caucasian, 5.4% were African American, 1.3% were American Indian or Alaska Native, 17.2% were Asian, 16.3% were Native Hawaiian or Pacific Islander, and 5.0% indicated an ethnicity not listed. Approximately 30.5% indicated they were of Hispanic/Latino/Spanish origin. All respondents used social media regularly. More specifically, 0.8% of respondents reported using social media monthly, 2% reported using platforms weekly, 31% reported using daily, 41.1% reported using hourly, and 24.7% reported using all the time.

Outline of survey

Respondents were presented with: “Social media users post positive messages. At times, their messages are considered overly positive and lacking in any negative emotions or negative experiences.” Respondents were asked to think about their social media experiences and indicate up to three intentions they think other users have for sharing positive posts that lack negative emotions and experiences on social media. The respondents were not asked to consider a specific social media platform when indicating their perceptions.

Item generation and reduction

The authors of this study reviewed the qualitative responses provided by the respondents and generated a pool of items representative of intentions to share toxic positivity (perceived TPI). This process yielded a total of 326 items. To reduce the number of items, the 326 items were grouped by theme and then scrutinized for redundancy and relevancy by four independent social media researchers. To increase face validity, items that were not relevant in the current context were removed. For instance, items that were not representative of an intention or motivation (i.e., “not everything on social media is true” and “not always realistic”) were removed. A total of 82 items were retained.

Administration of final 82 items

Sample

To provide results that can be generalized to most social media users, the 82 items were administered to two different age groups––a U.S.-based sample of Generation Z (ages 18–27) and Millennial (ages 28–42) social media users. These respondents were recruited from Prolific, an online research platform that ensures high data quality for behavioral research (Peer et al., 2022). There was a total of 149 respondents. Of these respondents, 49% were female, 46.3% were male, and 4.7% identified as non-binary or other. On average, respondents were 29 years old (SD = 6.12). Regarding ethnicity, 80.5% were Caucasian, 7.4% African American, 2.0% American Indian or Alaska Native, 7.4% Asian, and 2.7% indicated an ethnicity that was not listed. Approximately 13% indicated they were of Hispanic/Latino/Spanish origin. Regarding education, 1.3% reported having less than a high school education, 14.8% graduated high school, 26.2% had some college, 5.4% earned a 2-year degree, 39.6% earned a 4-year degree, and 12.7% had a professional degree or master's degree or doctorate degree. Regarding income, 47.7% had an annual household income of less than $50k, 34.2% earned between $50k and $100k, 14.1% earned between $100k and $150k, and 4% earned more than $150k. Regarding social media usage, 3.4% reported using social media monthly, 10% reported using platforms weekly, 61.4% reported using daily, 18.7% reported using hourly, and 6.7% reported using all the time.

Descriptively, when looking at how often Generation Z and Millennial social media users perceive other users to be posting toxic positivity, 5.4% reported rarely, 24.2% reported sometimes, 8.1% reported about half the time, 27.5% reported often, 31.5% reported most of the time, and 3.4% reported always. Regarding how many people Generation Z and Millennial social media users perceive to be posting toxic positivity, 1.3% reported none of the people they follow to be posting toxic positivity, 6.7% reported not many people, 19.5% reported a few people, 15.4% reported half of the people, 24.2% reported many people, 30.2% reported most of the people, and 2.7% reported everyone.

Outline of survey

Respondents were presented with: “Social media users post positive messages. At times, their messages are considered overly positive and lacking in any negative emotions or negative experiences. Thinking about your time on social media, please indicate the extent to which you agree or disagree with each item being a motivation for why people post overly positive posts that lack any negative emotions or negative experiences.” Respondents indicated their agreement or disagreement with each of the 82 items being a motivation behind sharing toxic positivity on a 7-point Likert-style scale ranging from strongly disagree (score = 1) to strongly agree (score = 7). The respondents were not asked to consider a specific social media platform when indicating their perceptions of posting toxic positivity. Data were analyzed with SPSS version 27.

Study 1: Results

RQ1 results

EFA

In such an early stage of research, discriminant validity can be confirmed through strategies such as an EFA (Zait & Bertea, 2011). Principal component analysis with varimax rotation was performed. The EFA was evaluated using the following criteria: eigenvalue greater than 1.0 (Kaiser, 1974), loading score for each factor was greater than or equal to 0.60 (Pituch & Stevens, 2015) with cross-loadings less than 0.40, and the factor having at least three items. A total of 44 items were eliminated after the first round of the EFA when the criteria were not met. After eliminating these items, a subsequent EFA eliminated four more items for failing to load above 0.60, resulting in a perceived TPI five-factor structure consisting of 34 items. A screen plot further confirmed the five-factor structure as it clearly showed a substantial drop after the fifth factor. See Appendix Table A1 and Appendix Figure A1.

Five-factor label development

In response to RQ1, respondents perceive there to be five different intentions behind sharing toxic positivity on social media. As discussed below, each intention was labeled and defined based on the items included in the EFA and its conceptual overlap with those identified from prior literature.

Inauthentic self-promotion intentions

Self-presentations are authentic when people present themselves in an honest way that is consistent with who they are offline (Yang & Bradford Brown, 2016; Yang et al., 2017). In contrast, self-presentations are inauthentic when people present a different version of themselves or their current situation, such as hiding the bad parts of themselves or their life. Thus, with perceived intentions such as, “to hide their real self” and “to hide problems,” this dimension relates to perceptions of sharing an inauthentic self-promotion and is defined as the intent to present an inauthentic persona online.

Malicious self-promotion intentions

A primary intention behind posting malicious comments online is to hurt others (Lee & Kim, 2015), which is in line with the findings in this study as users perceive others to share toxic positivity to “make others jealous” and “to make others feel less fortunate.” This dimension relates to sharing a malicious self-promotion and is defined as the intent to post content about oneself to make others experience negative feelings.

Personal self-promotion intentions

Social media is used to present a version of oneself that appears more interesting, popular, and attractive (Yau & Reich, 2019). As such, previous research indicates one motivation for sharing personal information is to be better known (Ghaisani et al., 2017), and similarly, respondents in the current study perceived intentions behind toxic positivity to include motivations such as “to gain attention” and “to make others think they are cool.” This dimension relates to personal promotion and is defined as the intent to make others think positive thoughts about themselves.

Commercial self-promotion intentions

Many people use social media for commercial or financial gain, and social media users perceive this, such as believing others post to gain more likes (Kondakciu et al., 2021). In line with these motivations, individuals in this study perceived intentions behind toxic positivity to include motivations of “earning money” and “gaining social media followers.” Thus, this dimension relates to commercial use and is defined as the intent to maintain a commercially viable persona online.

Positivity promotion intentions

Altruism is a common motivation found for sharing information with others online (Ghaisani et al., 2017), where users do not expect a reward in return. Instead, individuals share information (i.e. reviews) online to help others. In line with this motivation, individuals in the current study perceived intentions behind toxic positivity to include motivations such as “to share information,” “to share ideas,” “to show possibilities to others,” among others. Therefore, this dimension of perceived TPI relates to an altruistic motivation and is defined as the intent to service others in a positive way.

Five-factor EFA solution

The five-factor solution with 34 items accounted for a total variance of 65.28%. Cronbach’s alpha values ranged from 0.85 to 0.91, showing good reliability (Nunnally & Bernstein, 1994). The first dimension, inauthentic self-promotion intention, explained 26.97% of the variance; the second dimension, positivity promotion intention, explained 17.98% of the variance; the third dimension, personal self-promotion intention, explained 9.99% of the variance; the fourth dimension, malicious self-promotion intention, explained 5.39% of the variance; and the fifth dimension, commercial self-promotion intention, explained 4.95% of the variance.

Study 1: Discussion

These data capture perceptions of the intentions behind sharing toxic positivity. Results offer an understanding of the specific intentions users believe others have when leveraging positivity as a self-enhancement strategy, including inauthentic self-promotion, malicious self-promotion, personal self-promotion, commercial self-promotion, and positivity promotion. In doing so, this work extends the growing literature on self-presentation.

Data indicate that Generation Z and Millennial social media users encounter toxic positivity while browsing platforms and perceive 65.1% of others to be sharing toxic positivity at least “often” when posting. It is important to note that none of the respondents in these data indicated that people they follow “never” use toxic positivity. In addition, supporting the notion that social media users are positively biased (Prins & Wellman, 2023; Saha et al., 2021), these results provide further evidence that social media users perceive others to be posting toxic positivity, contributing to the foundational understanding of the newly positioned self-presentation strategy.

As stated, data suggest people perceive five unique intentions behind sharing toxic positivity. These dimensions represent both positive and negative perceptions of TPI, which align with research indicating that people have mixed perceptions of self-enhancement strategies (Hong et al., 2020; Lafrenière et al., 2016). In addition, the perceived TPI intentions found in this study align with research suggesting the motivations behind self-presentations often fall under altruistic and egoistic intentions (Ghaisani et al., 2017). Since this research indicates that perceptions of TPI are not all perceived the same, or even to be negative, it is possible to have varying effects across each dimension. With such different perceptions of TPIs on social media and potential psychological effects, it is important to investigate the role each dimension plays in the effects of social media engagement, including important areas such as upward social comparisons and digital identity presentation.

Study 2: Introduction

Exposure to manicured self-presentations is associated with greater social comparisons and greater malicious envy (Brooks et al., 2019; Chae, 2018) as well as malicious and inauthentic online behaviors, such as manipulating future posts (Rousseau, 2021) and deleting past posts that are considered unfavorable (Mun & Kim, 2021). Within a society where positivity is spread across social media (Saha et al., 2021; Vogel & Rose, 2016; Yau & Reich, 2019), which often includes inauthentic and false versions of self (Michikyan et al., 2015; Mun & Kim, 2021), it is important to explore how perceived TPI across social media may impact the user’s psychological processes (upward social comparisons) and future self-presentation behaviors (false self-presentation). Thus, grounded in the descriptive norm literature, a follow-up study (Study 2) examines the role perceived TPI plays in false self-presentation behaviors on social media and the mediating role of upward social comparison. In doing so, Study 2 (1) adds further rigor to the scale development of perceived TPI (Study 1) through predictive validity, (2) adds to the previously defined, and growing, impression management and digital identity literature surrounding toxic positivity as a self-presentation strategy, and (3) demonstrates how upward social comparison effects are more nuanced when examined as a mediator. A conceptual model of Study 2 can be found in Appendix Figure B1.

Study 2: Literature review

Perceptions of TPI

Communication research has long recognized the power of individual difference, and from this perspective, individuals perceive and process messages differently which results in different effects from media consumption (Chae, 2018; de Vries et al., 2018; Lim & Yang, 2015; Meier et al., 2020; Park & Baek, 2018; Vogel et al., 2015). Research suggests there are differences in comparison processes as some people engage in comparisons while others do not (Tandoc et al., 2015). Indeed, some individuals have a higher tendency to make comparisons toward others and thus may experience greater effects as social media is used more often (Vogel et al., 2015). In addition, there are individual differences found in the emotions experienced from comparisons, where some individuals experience benign envy (Chae, 2018; Lim & Yang, 2015; Meier et al., 2020), malicious envy (Meier et al., 2020), sympathy (Meier et al., 2020), shame (Lim & Yang, 2015), inspiration (Meier et al., 2020), and depression (Meier et al., 2020).

Given such individual differences and varying emotional responses in social comparison processes on social media, it is important to investigate how different perceptions of positive self-presentations on social media influence social comparisons and self-presentation behaviors. As such, the individual differences outlined in perceived TPI (Study 1) may offer a nuanced understanding of how different perceptions of positive posting influence processes of social comparison and subsequent online behaviors. Based on individual differences in perceptions, it is expected that each dimension of perceived TPI will influence social comparison and ultimately impact the impression management processes differently.

Upward social comparison

Social comparison theory (Festinger, 1954) posits that individuals have an internal drive to make comparisons to others. Social comparison research explores comparisons made across different contexts, especially comparisons made on social media (Chae, 2018; Fox & Vendemia, 2016; Lim & Yang, 2015; Meier et al., 2020; Midgley et al., 2021; Park & Baek, 2018; Pedalino & Camerini, 2022; Rousseau, 2021). Midgley and colleagues (2021) suggest upward social comparisons are made more often on social media compared to downward and lateral comparisons, which is also noted by the respondents in Chua and Chang’s (2016) study who say social media makes comparisons unavoidable by providing opportunities for users to share a positive self-image (Saha et al., 2021; Yau & Reich, 2019). Indeed, positive social media posts provoke upward social comparisons when users see a highlight reel of other people’s best life, especially when the post is from an acquaintance that is barely known (Chae, 2018; Pedalino & Camerini, 2022; Vogel et al., 2015). These types of posts influence perceptions that others are doing better in life.

When browsing selective or well-crafted posts, especially that of toxic positivity (Prins & Wellman, 2023), social media users gain a distorted impression of the person that is posting, subsequently invoking a comparison with themselves to that distorted perception, and potentially altering how real or false they present themselves (Chua & Chang, 2016; Rousseau, 2021). Indeed, comparisons made toward an ideal image of others are associated with self-enhancement strategies, such as manipulating and posting images of oneself on social media (Chua & Chang, 2016; Rousseau, 2021), especially to gain the attention of others (Chua & Chang, 2016). Thus, with the phenomenon of toxic positivity utilized as a social media self-presentation strategy, and positive posts being shared constantly through a never-ending stream of information updates, it is important to understand the relationship between social comparisons, perceptions of TPIs, and processes of self-presentation, such as manipulating photos (Chua & Chang, 2016; Rousseau, 2021) and deleting unfavorable posts (Mun & Kim, 2021).

False self-presentation

False self-presentations on social media refer to sharing information about oneself in a manner that is inconsistent with their true self (Gil-Or et al., 2015). False self-presentations are often motivated by exploration (experimenting with different identities), deception (sharing information that is not completely true), and impressing others (conforming to expectations for a favorable impressions) (Chua & Chang, 2016; Michikyan et al., 2015). This can be seen through users who falsify themselves (Mun & Kim, 2021), are selective when sharing photos (Lo & McKercher, 2015), edit photos to perfect their face or body before posting (Chua & Chang, 2016; Rousseau, 2021), or avoid negativity when posting online (Prins & Wellman, 2023).

Influential factors for portraying a false self-presentation include comparisons toward the ideal images (Rousseau, 2021), a desire to impress (Michikyan et al., 2015; Mun & Kim, 2021), and perceived misinformation (Markowitz & Hancock, 2018). Thus, with the prominence of toxic positivity (Prins & Wellman, 2023; Vogel & Rose, 2016) and comparisons being unavoidable on social media (Chua & Chang, 2016), it is important to explore the mechanisms by which perceived TPI influences future false self-presentation behaviors, such as through upward social comparison processes (Rousseau, 2021).

Mediating effect of upward social comparison

Self-presentation norms indicate individuals present the same image as they are perceiving from others in social interactions (Leary, 2019). For instance, self-presentation positivity norms indicate that individuals present themselves just as positively as others in the interaction (Leary, 2019). Research on deception supports this norm indicating that perceptions of lying are associated with future lying behavior (Markowitz & Hancock, 2018). This means, if social media users perceive others to be lying or deceptive in their online social interactions, users will also present false information. Regarding perceptions of toxic positivity, research suggests users respond to toxicity online by being toxic themselves, believing it is an effective strategy against toxic positivity (Patel et al., 2021). Building understanding around the process by which this reciprocal relationship occurs, the current research positions upward comparison as an explanatory process by which perceived TPI impacts false self-presentation behavior.

This reciprocal relationship through upward social comparison can be conceptualized within the bandwagon effect and descriptive norms literature. Leveraged to understand the diffusion of various behaviors, the bandwagon effect describes a process where people adopt behaviors simply because others are doing it. More specifically, it represents a psychological process where a behavior increases among a group simply because it is perceived that a large proportion of “others” are engaging in the behavior. Thus, the bandwagon effect increases when a person feels as though they are lacking by comparison and desires to fit in or be included with a particular group. Here, when individuals believe to be lacking by comparison, they adopt behaviors that are perceived to be done by most people, such as perceiving others to use toxic positivity as a social media self-presentation strategy and joining in on this behavior by altering one’s own online image to fit in with the group. Research supports this notion that individuals will manage a false self-presentation on social media to conform to the behaviors of others (Chua & Chang, 2016). That is, research suggests social comparison tendencies can drive the behavior to craft an enhanced image of oneself on social media.

This process is consistent with the descriptive norm literature that suggests what others are perceived to be doing is a driving force to behavior. Importantly, research (Bergquist & Nilsson, 2019) also suggests that behaviors people are not perceived to be engaging in will likely be avoided. Simply put, behavioral adoption is motivated by the desire to “conform” or “be like” others (Cialdini, 2005) that appear different from oneself. But, for an individual to appear different than oneself, there exists a process of comparison to realize the discrepancies between oneself and others (i.e. self-evaluation), as noted by social comparison theory (Festinger, 1954), and as outlined previously (Chua & Chang, 2016). This points toward the instrumental role social comparison plays in the relationship between perceived TPI and false social media self-presentations. Applying these perspectives, this study suggests that to appear like others, or be accepted, users are more likely to present a false self when they perceive other people to be posting toxic positivity, or when perceived TPI is greater.

Considering the five dimensions of perceived TPI, this study aims to examine the role each dimension plays in upward social comparisons and false self-presentation. Specifically, this study hypothesizes that perceptions of TPI will have different influences on an individual’s false self-presentation through upward social comparisons. Simply, it is expected that upward social comparison will increase from perceptions of “negative intentions” dimensions of TPI and subsequently increase the sharing of a false self, due to a desire to be more like those they are upwardly comparing themselves to. The mediation can be expressed in the following hypothesis:

H1(a-d): Upward social comparison increased by perceptions of TPI perceived as inauthentic self-promotion intentions (H1a), malicious self-promotion intentions (H1b), personal self-promotion intentions (H1c), and commercial self-promotion intentions (H1d) will be associated with greater false self-presentation on social media.

In contrast, perceptions of “positive intentions” TPI could trigger avoidance, and thus decrease upward social comparison. Here, perceived positive TPI would function as a positive outcome of exposure, decreasing the need for upward social comparison, and subsequent false self-presentation. However, the perception of a post as inspirational or informational could signal that they are less or lacking, even when done with perceived positive intent. That is, while the perception of TPI is not negative or egoistic, exposure to the content may still prompt an upward social comparison. Given the potential ambiguity of this connection, this study proposes the following research question:

RQ1: Will TPI perceived as positivity promotion decrease upward social comparison and be associated with less false self-presentation on social media?

Study 2: Method

Sample

Study 2 used an online survey developed using Qualtrics and administered to an online panel managed through Prolific. The sample is U.S.-based and contained 200 respondents unique from Study 1. Of these respondents, 52% were female, 45% were male, and 3% identified as non-binary or other. On average, respondents were 29 years old (SD = 6.12). Sixty-two percent were Caucasian, 7% Hispanic/Latino/Spanish, 13% were African American, 1% American Indian or Alaska Native, 12% Asian, and 5% indicated an ethnicity not listed. Thirteen percent of respondents graduated high school, 33% had some college education, 10% earned a two-year degree, 32% earned a 4-year degree, and 12% had a professional degree or master's degree or doctorate degree. Forty-four percent had an annual household income of $50k or less, 41% earned between $50k and $100k, 9% earned between $100k and $150k, and 6% earned more than $150k. Regarding social media use, 5.1% reported using social media monthly, 10.7% reported using platforms weekly, 58.2% reported using daily, 21% reported using hourly, and 5.1% reported using all the time.

Measurement

TPI

Perceived TPI was measured using the multidimensional scale items from Study 1 (see Appendix Table A1). Respondents were asked to think about their general social media use in the past week and to indicate the extent to which they agree or disagree that people share overly positive posts on social media for each of the following reasons. The 34 items measuring perceived TPI were presented, and respondents indicated their level of agreement with each item ranging from strongly disagree (score = 1) to strongly agree (score = 7). Each dimension established in Study 1 is operationalized independently in this study. The items were collapsed to each of the five dimensions and demonstrated good reliability (α = .90; M = 4.99; SD = 0.20).

Inauthentic self-promotion intention was measured with 10 items (α = .90, M = 4.99, SD = 0.09). Malicious self-promotion intention was measured with five items (α = .90, M = 4.45, SD = 0.27). Personal self-promotion intention was measured with five items (α = .85, M = 5.54, SD = 0.04). Commercial self-promotion intention was measured with five items (α = .89, M = 5.19, SD = 0.11). Positivity promotion intention was measured with nine items (α = .92, M = 4.87, SD = 0.10). A complete list of items for each dimension can be found in Appendix Table A1.

Upward social comparison

Upward social comparison was measured using an 11-item Likert-style scale adapted from previous research (Wu & Srite, 2021) ranging from strongly disagree (score = 1) to strongly agree (score = 7). Respondents indicated their level of agreement with items such as “The people I follow on social media have a better life than me” (α = .92; M = 4.76; SD = 0.28).

False self-presentation

False self-presentation was measured using a four-item Likert-style scale (Michikyan et al., 2015) ranging from strongly disagree (score = 1) to strongly agree (score = 5). Respondents indicated their level of agreement with statements such as “I sometimes try to be someone other than my true self on social media” (α = .82; M = 2.50; SD = 0.15).

Data analysis

Data were analyzed with SPSS version 27. To test the serial mediation models posited in Figure 5, a mediation analysis was conducted with Hayes’s (2018) PROCESS Macro Model 4, (95% confidence intervals, 5,000 bootstrap samples). See Appendix Figures C1–C5 for the conceptual models and results for H1a-d and RQ1.

Study 2: Results

Perceptions of inauthentic self-promotion (H1a)

The mediation analysis showed that perceptions of inauthentic self-promotion intentions on false self-presentation were mediated by upward social comparisons, F(2, 197) = 4.06, R² = .04, p = .02. Perceptions of inauthentic self-promotion intentions were associated with greater upward social comparisons (β = .16, SE = .08, t = 2.01, p = .04, 95% CI =0.00– 0.31), which was associated with a higher level of sharing false self-presentations (β = .17, SE = .07, t = 2.30, p = .02, 95% CI =0.02–0.31). Since the direct effect of inauthentic self-promotion intentions on false self-presentations was not significant (β = .11, SE = .08, t = 1.34, p = .18, 95% CI = −0.05,0.27), the upward social comparisons mediation H1(a) was supported by the data.

Perceptions of malicious self-promotion (H1b)

Perceptions of malicious self-promotion intentions on false self-presentation were mediated by upward social comparisons, F(2, 197) = 5.47, R² = .05, p = .01. Perceptions of malicious self-promotion intentions were associated with greater upward social comparisons (β = .15, SE = .06, t = 2.66, p = .01, 95% CI = 0.04–0.26) and were associated with a greater level of sharing false self-presentations (β = .15, SE = .07, t = 2.09, p = .04, 95% CI = 0.01–0.30). However, the significant relationship between perceived malicious self-presentation intentions and false self-presentation (β = .13, SE = .06, t = 2.13, p = .03, 95% CI =0.01–0.25) indicates that upward social comparisons only partially mediated the effect of malicious self-promotion intentions on false self-presentations.

Perceptions of personal self-promotion (H1c)

Perceptions of personal self-promotion intentions on false self-presentation were mediated by upward social comparisons, F(2, 197) = 3.25, R² = .03, p = .04. Respondents with perceptions of personal self-promotion intentions were associated with greater upward social comparisons (β = .23, SE = .08, t = 3.09, p = .002, 95% CI = 0.09–0.39) and associated with a higher level of sharing false self-presentations (β = .17, SE = .07, t = 2.34, p = .02, 95% CI =0.03–0.32). Since the direct effect of personal self-promotion intentions on false self-presentations was not significant (β = .04, SE = .08, t = .49, p = .62, 95% CI = −0.12, 0.20), the upward social comparisons mediation H1(c) was supported by the data.

Perceptions of commercial self-promotion (H1d)

Perceptions of commercial self-promotion intentions on false self-presentation were mediated by upward social comparisons, F(2, 197) = 3.26, R² = .03, p = .04. Respondents with perceptions of commercial self-promotion intentions were associated with greater upward social comparisons (β = .20, SE = .07, t = 3.14, p = .002, 95% CI =0.08–0.33) and associated with a higher level of sharing false self-presentations (β = .17, SE = .07, t = 2.33, p = .02, 95% CI =0.03–0.32). Since the direct effect of commercial self-promotion intentions on false self-presentations was not significant (β = .04, SE = .07, t = 0.50, p = .61, 95% CI = −0.10, 0.17), the upward social comparisons mediation H1(d) was supported by the data.

Perceptions of positivity promotion (RQ1)

Perceptions of positivity promotion intentions on false self-presentation were not mediated by upward social comparisons, F(2, 197) = 4.12, p =.02 (RQ1). Although not significant, respondents with perceptions of positivity promotion intentions were associated with less upward social comparisons (β =−.09, SE = .07, t =−1.20, p = .23, 95% CI = −0.23, 0.06), but were still associated with sharing a high level of false self-presentations (β = .19, SE = .07, t = 2.62, p = .01, 95% CI = 0.04–0.33). The direct association of positivity promotion intentions on false self-presentations was not significant (β = .10, SE = .07, t = 1.39, p = .17, 95% CI = −0.04, 0.25).

Discussion

To better understand and establish an academic application to the popular concept of toxic positivity, this study rooted toxic positivity in the impression management literature and argued the importance of individual perceptions. By employing a two-study design, the current research developed a measure estimating the perceived intentions for posting overly positive posts lacking in any negative emotions or negative experiences, referred to as perceived TPI (Study 1). Further, this research empirically investigated the impact of perceived TPI on false social media self-presentations through upward social comparisons (Study 2). As such, this research contributes to the currently small, but quickly advancing, literature on toxic positivity by explicating an academic measure that outlines the perceptions of social media users.

The current research found that individual differences in perceptions of TPI are associated with differences in upward social comparison processes. Here, negative perceptions of TPI, including perceived inauthentic self-promotion intentions, perceived malicious self-promotion intentions, perceived personal self-promotion intentions, and perceived commercial self-promotion intentions, are associated with greater upward social comparisons. In contrast, positive perceptions of TPI, including perceived positivity promotion intentions, are associated with less upward social comparison (although not significant). This is in line with research that indicates comparisons are made toward the ideal self-presentations of others while browsing social media (Chae, 2018; Pedalino & Camerini, 2022), and with studies suggesting individual differences in message processing, such as for individuals with different comparison tendencies (Park & Baek, 2018), different social media habits (Tandoc et al., 2015), and different emotions (Lim & Yang, 2015; Meier et al., 2020; Park & Baek, 2018). The current study offers a deeper understanding of comparison processes toward positive social media posts that may vary in perceptions. When intentions behind positive posting are perceived to be toxic, users are likely to experience greater upward social comparisons; in contrast, if intentions are perceived to be positive, users may experience less comparisons. This finding provides empirical evidence for upward social comparisons made from perceived TPI, especially social media posts that are perceived to have malicious intentions.

In addition, the current research offers empirical evidence for factors that influence the manipulation of social media self-presentations (Michikyan et al., 2015; Mun & Kim, 2021; Rousseau, 2021). Specifically, data indicate that upward social comparisons influenced by TPI are associated with false self-presentation behaviors. This finding aligns with studies that indicate comparisons are associated with self-enhancement strategies (Chua & Chang, 2016; Fox & Vendemia, 2016; Rousseau, 2021). However, the current study extends this research toward a more nuanced understanding of upward social comparisons by demonstrating the strength of upward social comparison as a mediator in the self-presentation process. Conforming to perceived social media norms, users are manipulating their online personas and making themselves seem more positive, or more popular, than they are to manipulate impressions, and to fit in with others. For instance, when users perceive manipulative intent, they in turn, compare the information to self and subsequently use deception and manipulation when presenting themselves online, too. This finding supports studies suggesting individuals reciprocate the self-presentation they receive from others (Chua & Chang, 2016; Leary, 2019; Markowitz & Hancock, 2018; Patel et al., 2021).

Interestingly, this study also indicates a significant direct effect between perceived TPI of malicious self-promotion and false self-presentation behaviors, which emphasizes a positive association between perceptions of lying and actual lying (Markowitz & Hancock, 2018), pointing to the strong deceptive behavioral effects of perceived malicious content. With both an indirect and a direct influence on sharing a false self, perhaps perceptions of malicious self-promotion intentions on social media go beyond deception. For instance, individuals who perceive malicious self-promotion TPI, and as a result share false information about themselves, may not only engage in upward social comparison, but may also be angered or offended (Pangestu et al., 2022), leading to a direct behavioral response in retaliation. Future research should look to tease out perceptions of malicious self-promotion TPI and its influence on behaviors and should consider potential avenues that may alleviate such malicious perceptions.

Limitations and future research directions

Research looking to test individual models, such as this study, should look to increase statistical power by collecting a larger sample. That said, sample size reflects the power to detect difference, and thus, the findings should only increase with sample size. Regardless, a more robust sample would allow for potential subgroup analysis, and thus, discovery of potential age, gender, or ethnicity patterns. Similarly, the current sample produced ceiling effects across some of the measures, suggesting additional work is still needed in this area to better understand the range of perceptions users have toward social media content. This could direct future research to consider cohort effects within perceived TPI and subsequent effects. Additionally, this research only looked at false self-presentation as a modeled outcome, and real and ideal self-presentation should also be considered in future research. Finally, this study positioned the toxic positivity phenomenon across various social media platforms, without keeping a specific platform in mind. As such, the current study does not speak to perceived TPI on specific social media platforms.

Future research should not only consider changes to better address these limitations, but also consider the positive and negative impacts of perceived TPI from social media consumption. Substantial literature has connected social media use to diminished psychological well-being (Chae, 2018; Park & Baek, 2018) such as depression and decreased life satisfaction. Future research should further investigate the connection between perceived TPI and psychological well-being, through upward social comparison. In addition, future research could look to further examine the false self-presentation construct to include intent. That is, while the current research indicates that self-presentation is intentional, and the items indicate a conscious decision to present a false or altered self, future research could expand these insights to better understand why or with what malice people present the information falsely. Finally, future research should explore a more specific understanding of the effects that perceived TPI may have on specific platforms, especially as platforms are used for different purposes.

Data availability

The data are available on the Open Science Framework (https://osf.io/epc2s/).

Conflicts of interest: The authors declare that there is no conflict of interest.

References

Appendix A: Study 1 EFA and Perceived TPI Measurement Items

Appendix B: Conceptual Model for Study 2

Appendix C: Conceptual Models and Results for Study 2 H1a–d and RQ1