The Cost of Looking Up: Social Comparison as a Partial Mediator Between Social Anxiety and Vindictive Behavior in a Non-anonymous Ultimatum Game

Abstract

Self-appraisals in relation to internalized standards are fundamental constructs in clinical and evolutionary models of Social Anxiety (SA). Empirical evidence has consistently shown that socially anxious individuals tend to engage in upward social comparisons and aim to adjust their social standing accordingly. However, the consequences of perceiving oneself as inferior to others within social contexts remain under-explored. To this end, a sample of N = 166 college students participated as respondents in an iterated, non-anonymous online Ultimatum Game (UG) against two familiar classmates: a fair allocator and a highly unfair allocator. Participants were informed that their identities would be disclosed to their assigned co-players. In reality, however, all participants played against a computer-simulated opponent. Our findings both replicate and extend previous research by showing that individuals with elevated social anxiety tend to exhibit vindictive behavior in response to unfair monetary UG splits (i.e., costly punishment) compared to their less anxious peers, even without the safeguarding of anonymity. Overall, our data lend empirical support to Social Rank theories of social anxiety and underscore the role of (upward) social comparison as a partial mediator explaining behavioral outcomes (i.e., vindictive behavior) in SA. Theoretical and clinical implications are discussed.

Keywords

Social anxiety social comparison vindictiveness ultimatum game

Theoretical Background

Social anxiety (SA) is characterized by an exaggerated and pervasive fear of social evaluation (American Psychiatric Association, 2013). This disproportionate fear is endured with intense distress and is rooted in a deep-seated belief of negative evaluation and subsequent rejection. Moscovitch (2009) posited that the fear of social evaluation in SA is driven by a chronic feeling of falling short of societal expectations. This perception often leads to more frequent upward social comparisons, as noted by Goodman and colleagues (2021). Extensive prior research has established a connection between social anxiety and dysfunctional social behavior (Rodebaugh et al., 2017; Tone et al., 2019), with a host of subsequent studies reporting that even subclinical levels of social anxiety can result in dysfunctional social interactions (i.e., Alden & Taylor, 2004). Several studies have noted dysfunctional social behaviors in SA, such as abberant conformity designed to avoid social rejection (i.e., Feng et al., 2018) or acting hostile in competitive settings (Tone et al., 2019).

Despite the ubiquity of social impairments in social anxiety, researchers have yet to uncover the underlying causal mechanisms. Arguably, the primary challenges in studying social behavior in SA are predominantly technical or methodological. Replicating the complexities of social interactions in controlled research settings is particularly difficult, and accurately measuring these interactions presents an even greater challenge.

Social Behavior in Social Anxiety

Consensus among most theorists (Alden & Taylor, 2004; Davila & Beck, 2002) suggests that social anxiety is typically characterized by a predominantly avoidant interpersonal style which ranges from avoidance of social performance to avoidance of social interactions altogether (American Psychiatric Association, 2013). However, research has also identified increased interpersonal dependency among individuals with high SA (Alnaes & Torgersen, 1988). Notably, this dependency cannot be fully understood without acknowledging the fundamental human need for belonging and social interaction (Hirschfeld et al., 1977), which can partly contribute to conflict avoidance in socially anxious individuals.

Conflict avoidance, on the other hand, is a fundamental aspect dictating how the socially anxious navigate the world. The so-called “avoidance motivation” (Cremers, Veer, Spinhoven, Rombouts, & Roelofs, 2015) is thought to originate in the ventral parts of the striatum, a dopaminergic-mediated brain loop (Haber & Knutson, 2010) that regulates both reward sensitivity and punishment avoidance (Salamone, 1994). Some scholars (Cremers et al., 2015) even hypothesized that SA is associated with a dopaminergic imbalance in the striatal motivational system that could result in a disproportionate drive to avoid social scrutiny. Along the same lines, Reinforcement Sensitivity Theory (Kimbrel, 2008) points to behavioral inhibition (otherwise a central characteristic of SA) as a potential culprit for punishment avoidance. In theory, behavioral inhibition, self-criticism, shyness, and feelings of inadequacy (Alden & Phillips, 1990) are thought to reflect an avoidant interpersonal style. However, the same attributes have also been associated with a dependent interpersonal style (Darcy et al., 2005). Socially anxious individuals, thus, exhibit overlapping traits associated with both dependent and avoidant styles.

Along these lines, an additional dichotomous categorization has been proposed to describe social behavior in SA. Gilbert (2000) introduced two interpersonal dimensions: dominance (vs. submission) and warmth (vs. coldness) to explain these complex dynamics. Subsequent research (Rodebaugh et al., 2017) indicates that socially anxious individuals are often characterized by low levels of both warmth and dominance. Intriguingly, while these individuals generally prefer warm and submissive peers, they tend to admire those who exhibit higher levels of dominance. In a related finding, Rodebaugh et al. (2016) observed that individuals high in social anxiety demonstrate increased tolerance for individuals perceived as less warm.

However, research examining social behavior and interactions within more ecologically valid, real-world contexts for individuals with social anxiety (SA) remains limited. Lately, the growing field of behavioral economics has begun to elucidate the mechanisms underlying social behavior in SA, including cooperation, trust, and reputation. The field of behavioral economics has made significant contributions by offering sound theoretical and methodological frameworks (i.e., predefined economic games that allow the study of otherwise abstract constructs in laboratory settings) (Fehr & Camerer, 2007).

Reputation, Upward Social Comparison, and Vindictive Behavior in Social Anxiety

Reputation can be conceptualized as the collective perception held by others about an individual, which is formed based on their past actions and behaviors. It is also viewed by some as a measure of the trust and respect an individual has earned from their peers (Moscovitch, 2009).

For example, for individuals with elevated social anxiety, a workplace presentation can be a particularly stressful event. Even the mere suspicion of potential gossip or negative commentary about their work can be triggering. Negative comments about the presentation might be spread among colleagues, damaging the presenter’s reputation. Consequently, individuals with high levels of social anxiety may exert considerable effort to prevent the occurrence of negative gossip.

As previously mentioned, socially anxious individuals are particularly sensitive to how others perceive them and often engage in various “safety behaviors” to manage social impressions (Clark & Wells, 1995). These behaviors may include rehearsing responses or hiding their hands to conceal signs of nervousness.

The question remains, however, as to how socially anxious individuals react when their social image or reputation is threatened. Common responses may include ignoring the threat or reacting negatively towards those who pose a threat to one’s social status (Crusius & Lange, 2014). In the current paper, we hypothesize that high SA individuals are more likely to exhibit maladaptative behaviors in response to threats to their reputation or social standing.

To our knowledge, a limited number of studies have examined the role of reputation or social rank in SA (Versella, Piccirillo, Potter, Olino, & Heimberg, 2016), and even fewer have investigated behavioral responses to reputational threats in this population (Bica et al., 2021). Furthermore, the existing literature is constrained by methodological limitations that restrict the generalizability of findings outside of the laboratory setting. For example, Bica and colleagues (2021) noted vindictive behaviors (i.e., costly punishment) in a sample of high SA individuals under conditions of anonymity over a series of asymmetric UG trials designed to threaten one’s social status. However, participants’ reactions and behaviors were safeguarded by the use of anonymity, which limits the generalizability of these findings to real-world settings where such behaviors might be more constrained. Similarly, Rodebaugh and colleagues (2017) observed that socially anxious individuals report a higher baseline level of vindictiveness. However, when looking at social anxiety severity and vindictiveness independently, they found that vindictiveness alone, rather than the severity of SA, was a predictor of unfair behavior during iPD (iterated Prisoner’s Dilemma) trials (Rodebaugh et al., 2015).

Building on prior research, we propose that social comparison, a relative measure of perceived reputation or social rank, could serve as a potential mediator between social anxiety and vindictiveness. The concept of social comparison was first introduced by Festinger in 1954 and later integrated into the self-esteem literature (Suls & Wills, 1991). Social comparison can be either downward, thus aimed at enhancing one’s self-esteem, or upward, which often involves initial discomfort but can ultimately protect self-esteem by serving as a catalyst for personal growth and aspirations of achieving similar status (Wood, 1989). Recent research (Goodman et al., 2021) provides empirical support to the idea that socially anxious individuals are excessively concerned about the prospect of making an unfavorable social impression, consequently engaging more frequently in upward social comparison. These findings come to substantiate Gilbert’s (2000) claims that anxious individuals often perceive themselves as inferior relative to others.

In previous research, the Ultimatum Game (UG; Güth, Schmittberger, & Schwarze, 1982) has been commonly used to study vindictive behavior and social rank. Building on previous findings that socially anxious individuals are biased to view themselves as inferior, we employed an iterated and asymmetric Ultimatum Game designed to minimize intrapersonal variability across responses. The UG is a well-known behavioral-economic task that involves a simple financial exchange between two randomly assigned players: one allocator and one responder. The allocator gets to decide how to split the given amount between himself/herself and their counterpart, the responder. While the allocator proposes a division of the money, the responder gets to either accept or reject it. If the responder accepts, the proposed split stands, and both players receive the money as proposed by the allocator. However, if the responder rejects the proposal, neither the allocator nor the responder receive any money. In line with economic predictions (Stigler & Stigler, 1950), the players should maximize utility either by proposing the smallest positive amount (as allocators) or by accepting any non-zero amount. In reality, however, both economic predictions were disproved. Studies showed that allocators are quite generous, sharing a fair quota of their money, while the responders typically decline offers deemed unfair despite receiving non-zero offers (Knoch et al., 2006). Rejecting non-zero but unfair offers is referred to as „costly punishment”. This form of directional vindictive behavior is often motivated by the desire to protect one’s social reputation. The social function of costly punishment is further evidenced by findings demonstrating varying acceptance thresholds based on whether the offers were computer generated or proposed by a peer. Unfair offers from peers are more frequently rejected than those generated by computers (Blount, 1995).

Study Overview

By using an iterated multi-shot Ultimatum Game, we examined whether and how individuals with elevated levels of social anxiety react when a familiar peer threatens their reputation. Along the way, we operationalized and tested several clinical and evolutionary hypotheses related to social comparison and social behavior in social anxiety, focusing on vindictive behavior as a reaction to unfair behavior and reputational threats. Additionally, we proposed social comparison as a mediator between social anxiety and vindictive behavior across UG trials aiming to provide empirical support to current theoretical frameworks of social anxiety (Gilbert, 2000). At lower scores on the social comparison scale, participants tend to view themselves as inferior, thus engaging in upward social comparison. Conversely, at higher scores, participants tend to hold a more favorable self-view, thus engaging in downward social comparison. As social rank and reputational concerns carry a strong social impact, we could not accurately grasp social behavior by using the traditional, anonymous UG approach. Thus, we modified the UG’s parameters by not concealing the identity of the UG players (i.e., allocators and respondents). Participants were thus led to believe that their counterparts (responders) will be made aware of their identity, although in reality, all participants played the UG trials against a computer. To prompt social comparison, prior to starting the UG session, responders were presented with the average amount received by a previous player. While prior research (Harlé & Sanfey, 2007; Harlé et al., 2010) used anonymous and unrelated UG participants, we recruited our sample from college students who were familiar with each other.

In line with Gilbert and Trower’s (e.g., Gilbert, 2000; Gilbert & Trower, 2001) evolutionary model of social anxiety, which posits that socially anxious individuals tend to regard social interactions as competitive struggles driven by an inherent need to dominate or at least to mitigate the risk of embarrassment or status loss, we hypothesized that individuals high in social anxiety would employ costly punishment (i.e., accepting less unfair offers) to a greater extent than their low-anxiety peers, as a means of protecting their self-worth and social imagine (Yamagishi et al., 2012). Additionally, we aimed to empirically validate social comparison as a mediator between social anxiety and vindictive behavior in the UG.

Method

Participants

Sample size estimations were informed by previous meta-analytic studies that report a small to medium effect size for social outcomes measured by means of behavioral-economic tasks such as the Ultimatum Game (i.e., Thielmann et al., 2020). In order to obtain a small effect size η² of 0.25, we calculated a minimum of 140 participants needed to achieve a recommended power of 1 − β = .80. Sample size estimations were calculated by means of GPower 3 (Faul et al., 2007). Due to the nature of our sample (individuals who are familiar with each other), we recruited Psychology and Pedagogy students from the same university. To create the impression of playing against a familiar classmate, recruitment announcements were disseminated sequentially, one academic group at a time. As most academic groups at our university consist of 30–40 students, the likelihood of participants being unfamiliar with one another was minimal.

In total, 351 volunteers responded to our invitation to take part in our experiment. Out of the initial sample, only a sample of 167 undergraduate students (125 women, ranging in age from 19 to 35; M_age ₌ 21.9 years) completed all the phases (pre-screening and UG trial). One participant was familiar with the UG game, subsequently being removed from the final sample (N = 166 participants). Given the nature of the sample size, the recruitment process spanned over two academic years. Two distinct pre-screening sessions were conducted at the beginning of each academic year. All participants were pre-screened with the Liebowitz Social Anxiety Scale-Self–Report for social anxiety (LSAS–SR; Baker, Heinrichs, Kim, & Hofmann, 2002). Based on the LSAS-SR scores, participants were categorized either as High social anxiety (High SA; >30 points on LSAS; n = 72) or Low social anxiety (Low SA; <30 points; n = 94) (Rytwinski et al., 2009). Our data show no significant gender [U = 3366.0, p = .93] or age differences [t (164) = −1.33, p = .18] between the two groups (High SA vs. Low SA).

Written informed consent was obtained from all participants prior to the screening stage. After finalizing the UG trials, all participants were debriefed as to the nature and purpose of the experiment. Participants were treated in accordance with the American Psychological Association’s ethical guidelines and the study was conducted in accordance with the latest version of the Declaration of Helsinki.

Procedure and Materials

After accepting the invitation to take part in a study involving financial decisions, participants were invited to complete an online survey consisting of two sections (one demographic section and a set of questionnaires, i.e., LSAS-SR and social comparison ratings). As a prerequisite, participants were asked to select a convenient time slot (from a schedule provided by our team) to play a simple online game involving economic transactions. Participants received two SMS reminders in the days preceding the online Ultimatum Game. We stressed that participants could only log into the UG platform within 5 minutes of their selected time slot. Outside of this strict time slot, the UG platform could not be accessed.

Self-Reported Measures

Social Anxiety (SA)

For the purposes of this study, we used the Liebowitz Social Anxiety Scale—Self–Report (LSAS–SR) to assess social anxiety levels. LSAS consists of 24 items with possible scores ranging from zero (minimum) and 144 (maximum) distributed across 2 subscales (i.e., Social Fears and Avoidance of Social Situations). The LSAS provides an overall index of SA severity. Participants are required to rate on a zero (none) to 3 (severe) scale how much fear or avoidance they would feel if they would have to „Call someone in public”; „Eat in public” or „Go to a Party”. The recommended cut-off to screen for SA is set to greater or equal to 30, which provided the best balance between sensitivity and specificity in validation studies (Rytwinski et al., 2009). In line with previous studies, (Fresco et al., 2001; α = 0.95) the LSAS obtained sound psychometric qualities in our study, α = 0.93 (0.93 for the Social Fears subscale and 0.92 for the Avoidance subscale).

Social Comparison

Judgments of perceived social standing were assessed via the Social Comparison Scale (SCS; Allan & Gilbert, 1995). First developed to measure self-perceptions of social rank and social stance, the SCS consists of 11 bipolar pairs (i.e., inferior-superior, outsider-insider; left out-accepted) and requires participants to indicate on a eleven-point scale (0–10) their perceived stance relative to “other people”. At lower scores, participants tend to evaluate themselves as inferior (thus engaging in upward social comparison). Conversely, at higher levels, participants tend to hold a favorable image of themselves (i.e., engaging in downward social comparison). The SCS is the most commonly employed scale to assess social comparison (McCarthy & Morina, 2020) and has excellent psychometric properties, with internal consistency scores ranging from 0.84 to 0.87 (Allan & Gilbert, 1995). The Cronbach’s alpha obtained in our study is α = 0.83.

Experimental Task

The Ultimatum Game (UG)

The UG variant developed for this study was designed as an iterated online network game. While the traditional UG observes behavior over a one-shot interaction, we opted for the iterated version which involves multiple assessments of how responders behave, thus reducing the bias associated with intrapersonal variability. To ensure more ecological validity, participants were selected from a close-knit sample (i.e., classmates). Before starting the UG, all participants were required to disclose their full name. They were informed that both players’ identity will be disclosed at the end of the session. Afterwards, participants were presented with an overview of the UG and its rules. All participants completed a set of comprehension questions before proceeding with the actual task (i.e., “if the allocator keeps $10 for him/herself, then the responder would receive…”). In case participants failed to answer correctly to all comprehension questions, the software would reiterate the rules. After being „randomly” assigned the role of the responder, the computer displayed the average amount their allocator partner had proposed in previous sessions with different colleagues. This amount was invariably set at 50% of the entire amount ($10) and it was the same for all responders. As responders, participants interacted with 2 other „classmates” whose proposals were manipulated to appear as either fair (10 offers; from 50% to 55%) out of $20 or unfair (10 offers; from 20% to 40%). Inter-trial waiting time (while the allocator decides on the proposed amount) varied from 20 to 40 seconds and responders had up to 45 seconds to decide on whether they accept or not the proposed split. Each UG offer was followed by an 11 points visual analogue scale (VAS) (0 = „not at all”; 10 = „extremely”) used to assess the fairness of each offer (Bică & Crețu, 2022). The purpose of the scale was to cross-check whether offers defined as fair were indeed perceived as fair and offers defined as unfair were indeed perceived as unfair. The UG task was developed using Python and JavaScript.

Analytic Strategy

Primary analyses were conducted in JAMOVI. A series of mixed model analyses of variances [2 Groups (High SA vs. Low SA) x 2 Conditions (Fair vs. Unfair partner)] ANOVAs) were conducted to investigate how the high SA group would respond to unfair behavior from a familiar peer in a non-anonymous context. We also proposed social comparison as a mediator between SA and UG outcomes (i.e., acceptance rates for unfair splits). Given the high intrapersonal variability of SA (particularly at lower levels) and the fact that VD was an outcome variable, mediation was tested only for the high SA sub-group (N = 72). As per Preacher and Hayes’s (2008) recommendation, the unstandardized indirect effect was computed for 5.000 bootstrapped samples [95% CI].

Results

Preliminary Analyses

Perceived Fairness

Data were analyzed by means of a mixed 2 X 2 ANOVA (social anxiety status X fairness condition) on participants’ fairness ratings. In line with previous studies (Bica et al., 2021), we observed a significant main effect of Fairness [F (1, 164) = 839.2, p < .001, partial η² = .83]. We can conclude that the experimental manipulation was successful (i.e., participants rated as fair proposals from 50% to 55% of the amount, and as unfair proposals from 20% to 40% of the entire amount). The results were comparable across subgroups, with high SA participants rating unfair splits as (M_{perceived fairness} = 3.15) in comparison to their Low SA counterparts (M_{perceived fairness} = 3.77). For the perceived fairness of the fair splits, participants high in SA rated unfair splits as (M_{perceived fairness} = 9.44) in comparison to their Low SA counterparts (M_{perceived fairness} = 9.48). Descriptives and main results are rendered in Table 1 and Figure 1.

Figure 1.

Acceptance rates by Group. Note. P < .05 only in the Unfair Partner condition.

Figure 2.

Conceptual model - mediation.

Mediation Analysis – Assumptions Checks

Consistent with Gilbert’s (2000) conceptual framework, we found high levels of social anxiety to be a significant predictor of unfavorable social comparison. Specifically, participants with higher SA scores exhibited a greater tendency to engage in upward social comparisons [R² = .27, F (1, 71) = 25.97, p < .001]. Further, our findings show that unfavorable social comparisons predict vindictive behavior (i.e., costly punishment) following unfair UG splits [R² = .48, F (1, 71) = 64.54, p < .001]. Similarly, we found that high social anxiety levels predict vindictive behavior following unfair UG splits [R² = .47, F (1, 71) = 64.06, p < .001], thus lower acceptance rates for unfair monetary offerings. Similarly, the correlation matrix renders a negative significant correlation between SA and relative standing (i.e., social comparison) (r = −.53, p < .001), suggesting that individuals high in SA tend to evaluate themselves negatively in comparison to others, thus engaging in upward social comparison. Furthermore, our data showed a positive significant correlation between relative standing and (%) acceptance rates for unfair UG splits , (r = .69, p < .001), suggesting that individuals with a positive self-view have higher acceptance rates and consequently exhibit lower levels of vindictive behavior across unfair UG trials. Finally, a significant negative correlation was observed between SA and (%) acceptance rates for unfair UG splits (r = −.69, p < .001), indicating that individuals high in SA tend to accept less unfair offers, suggestive of vindictive behavior towards unfair partners. A correlation matrix with all the variables included in the model can be found in the Online Supplemental Materials.

Main Analyses

Acceptance Rates (%)

For the purpose of this analysis, we used the average (%) acceptance rates across the two conditions (i.e., fair vs. unfair partner). We used a 2 (condition: fair vs. unfair allocator) x 2 (social anxiety status: High vs. Low social anxiety) mixed model analysis of variance in order to assess possible differences in social decision-making patterns. The dependent variable is the participants’ (%) acceptance rates. As expected, we replicated the main effect of fairness (i.e., indeed, participants accepted significantly more equitable offers vs. inequitable offers) [F (1, 164) = 418.9, p < .001, partial η² = .71]. Social anxiety alone yields a significant effect on participants’ acceptance rates, [F (1, 164) = 4.09, p < .05, partial η² = .02]. Interestingly, our data show a significant interaction between Condition and SA status, [F (1, 164) = 7.23, p < .05, partial η² = .04]. Post-hoc pairwise comparisons revealed significant differences between conditions. For instance, for unfair splits, participants high in SA accepted significantly fewer splits (M_{acceptance rate%} = 29.3) in comparison to their Low SA counterparts (M_{acceptance rate%} = 42.5) (p < .05). No such differences were found for fair splits between high SA (M_{acceptance rate%} = 95.4) and low SA (M_{acceptance rate%} = 93.3) (p = .42). The results are rendered in Figure 1.

The Mediating Effect of Social Comparison

Our results revealed a significant direct effect of social anxiety on social comparison (i.e., relative standing), with b = −.05, SE = .01, [-.08 to −.02, 95% bootstrapped CI], p < .001 (a path) and on costly punishment with b = −.51, SE = .09, [-.68 to −.29, 95% bootstrapped CI], p < .001 (c path). The effect of social comparison or relative standing on costly punishment (b path) was also significant with b = 4.95, SE = 1.12, [2.54–7.36, 95% bootstrapped CI], p < .001. The indirect effect of social anxiety on costly punishment through social comparison (or relative standing) (ab path) was significant, yielding a b = −.27, SE = .11, [-.56 to −.09, 95% bootstrapped CI], p < .05, consistent with a partial mediation result explaining 34.8% of the total effect. The model and main results are rendered in Table 2 and Figure 2.

Table 1.

Descriptives.

	SA status	Mean	Median	SD	Minimum	Maximum
Social anxiety	High SA	72.11	77.00	23.53	32.00	110.00
Social anxiety	Low SA	16.35	18.50	7.07	1.00	29.00
Perceived relative standing	High SA	3.18	3.00	2.46	0.00	8.00
Perceived relative standing	Low SA	6.16	7.00	2.67	0.00	10.00
(%) Acceptance rates for unfair splits	High SA	29.31	20.00	26.92	0.00	100.00
(%) Acceptance rates for unfair splits	Low SA	42.55	30.00	34.64	0.00	100.00
(%) Acceptance rates for fair splits	High SA	95.42	100.00	15.19	30.00	100.00
(%) Acceptance rates for fair splits	Low SA	93.30	100.00	18.16	10.00	100.00
Perceived fairness for unfair splits	High SA	3.15	3.00	1.72	1.00	10.00
Perceived fairness for unfair splits	Low SA	3.77	3.00	2.13	1.00	10.00
Perceived fairness for fair splits	High SA	9.44	10.00	1.37	0.00	10.00
Perceived fairness for fair splits	Low SA	9.48	10.00	1.52	1.00	10.00

Table 2.

Bootstrapping Coefficients.

			95% Confidence interval
Effects			Label	Estimate	SE	Lower	Upper	p
Social anxiety	→	Relative standing	a	−0.055	0.013	−0.084	−0.026	<.001
Relative standing	→	(%) Acceptance rates for unfair UG splits	b	4.950	1.120	2.544	7.363	<.001
Social anxiety	→	(%) Acceptance rates for unfair UG splits	c	−0.515	0.090	−0.680	−0.295	<.001
Indirect			a x b	−0.275	0.111	−0.569	−0.093	<.05

Discussion

Although social behavior is a critical area of research in social anxiety, pinpointing the underlying mechanisms responsible for social difficulties remains challenging. While recent advancements in behavioral economics have offered valuable tools and methodological frameworks, the inherent lack of ecological validity associated with these methods limits the generalizability of findings.

Therefore, our primary objective was to investigate social behavior and interactions in a more ecologically valid setting, namely, a non-anonymous Ultimatum Game. The present study offers a novel approach to this challenge and underscores the partial mediating role of (upward) social comparison in social anxiety-related behaviors. The first objective was to investigate whether individuals high in SA would display vindictive behavior (i.e., costly punishment) in response to unfair behavior. To emphasize the social implications of the interaction, the unfair behavior was manipulated to appear as coming from a familiar classmate. In the participants’ eyes, they were the first to receive such unfair treatment from a peer (in previous trials, the allocator had acted fairly towards other participants). This is the first study to employ a non-anonymous UG in the study of SA. Secondly, and most notably, we examined the mediating role of (upward) social comparison in the relationship between social anxiety and vindictive behavior (i.e., costly punishment) in the Ultimatum Game. While previous research has primarily focused on the prevalence and psychological implications of upward social comparison in social anxiety (Goodman et al., 2021), we extended these findings by investigating its influence on behavioral outcomes. The results are discussed below.

Consistent with previous UG findings (Bica et al., 2021), our results demonstrate that individuals generally punish unfair partners, regardless of potential financial losses or social consequences. Extending these findings, we show that individuals with high SA accepted significantly fewer unfair offers compared to their low social anxiety peers, replicating previous research. Furthermore, our findings expand upon previous research on vindictive behavior by demonstrating consistent patterns of behavior among socially anxious individuals across various UG methodologies and experimental conditions. While previous research (Bica et al., 2021) primarily employed anonymous Ultimatum Game (UG) methodologies to observe lower acceptance rates for unfair offers, our study replicated these findings using an open UG variant. Notably, our results indicate that vindictive behavior in UG settings is not solely contingent on anonymity, demonstrating the generalizability of these findings across different experimental conditions. We further argue that such reactions to unfairness are ubiquitous and can interfere with adaptive social functioning in SA. To accurately examine reactions to threats to social rank, we designed an UG that facilitated peer-to-peer social comparison. Consistent with Trower and Gilbert’s (1989) model of social anxiety, which posits that the actions of individuals with SA are influenced by their competitive social schema, our findings indicate that high SA participants prioritized maintaining reputation by rejecting offers from unfair peers.

Notably, our findings support the partial mediating role of (upward) social comparison in the relationship between social anxiety and Ultimatum Game decision-making, contributing to the growing body of research emphasizing the importance of social comparison processes in social anxiety (McCarthy & Morina, 2020). Despite its initial conceptualization by Festinger (1954), social comparison has only recently been integrated into theoretical models of social anxiety.

To our knowledge, this study is the first to propose social comparison as a mediator of behavioral outcomes in social anxiety. As noted by Goodman and colleagues (2021), there is a notable puzzling lack of research on social comparison in social anxiety and its potential impact on social and emotional functioning. Notably, we observed that social anxiety would increase the frequency or intensity of upward social comparisons, thus resulting in lower acceptance rates for unfair monetary splits (i.e., costly punishment). This finding is particularly relevant given that socially anxious individuals exhibit greater instability in their social comparison processes on a daily basis, rendering clinical outcomes (e.g., emotions, behaviors) unpredictable (Goodman et al., 2021). Our findings are in line with Moscovitch’s (2009) assumption that socially anxious individuals view themselves as “lesser” in comparison to their non-anxious counterparts, and social evaluation is thus a preamble to social rejection. Similarly, the mediating effect of social comparison lends empirical support to Gilbert’s Social Rank Theory (2000) which emphasizes the crucial role of avoiding social embarrassment for socially anxious individuals. Consistent with this perspective, the excessive vindictive behavior displayed by individuals with high SA can be interpreted as a counter-response to perceived threats to social status, mediated by social comparison processes. Future research should explore additional mediators that may influence behavioral outcomes.

Overall, our findings demonstrate the broader implications of social comparison processes in social anxiety-related outcomes. While previous research has primarily linked (upward) social comparison to negative emotional experiences in social anxiety (Goodman et al., 2021), our study highlights the role of social comparison in shaping social behaviors.

Collectively, these findings contribute to the growing body of literature examining the relationship between mental health and social comparison processes (McCarthy & Morina, 2020). Furthermore, our results hold potential translational implications for clinical practice. Future research could explore the development of tailored interventions targeting social comparison processes in social anxiety.

Several limitations should be acknowledged. While our design addressed some limitations of previous research by employing a close-knit sample which required participants to disclose their identities, the use of an analogue sample (i.e., participants with elevated levels of social anxiety) restricts the generalizability of our findings to clinical samples.

Future studies should aim to replicate our findings in clinical samples to address the limitations associated with the high intrapersonal variability often observed in analogue samples. By utilizing clinical samples, researchers can analyze social anxiety scores continuously, thereby overcoming a limitation of the present study. Furthermore, a significant limitation of our study is the use of cross-sectional mediation analysis, which precludes the establishment of temporal precedence between the predictor and mediator. To address this, future research could employ longitudinal designs that allow for the examination of potential changes over time in the relationship between social anxiety and social comparison processes.

Moreover, our study registered a substantial attrition rate between the pre-screening phase and the UG task, leading to the exclusion of a significant portion of participants who completed the screening questionnaires. Consequently, there is a risk that our findings may be influenced by a selection bias.

Future replications are warranted to confirm the mediating role of social comparison in clinical samples. It could be that severe anxiety would completely block any vindictive tendencies, regardless of social comparison processes. Future studies should also examine social comparison in relation to other behavioral outcomes. Collectively, our findings point to a novel causal mechanism that could lead to social dysfunction in social anxiety, and provide insights into how social comparison processes mediate behavioral outcomes in social anxiety.

Supplemental Material

Supplemental Material - The Cost of Looking Up: Social Comparison as a Partial Mediator Between Social Anxiety and Vindictive Behavior in a Non-anonymous Ultimatum Game

Supplemental Material for The Cost of Looking Up: Social Comparison as a Partial Mediator Between Social Anxiety and Vindictive Behavior in a Non-anonymous Ultimatum Game by Mariana Călin, and Andreea C. Bică in Psychological Reports

Footnotes

Declaration of Conflicting Interests

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

Funding

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

Ethical Statement

ORCID iD

Andreea C. Bică

Data Availability Statement

The data presented in this study are publicly available in the below github repository: .

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Floricica Mariana Călin is a researcher and academic affiliated with Ovidius University of Constanța. Her research primarily explores topics related to emotional intelligence, self-esteem, attachment styles, and psychological well-being, with a focus on both adolescent and adult populations. She has contributed to studies examining the psychological effects of caregiving and aging, as well as broader issues in mental health and social psychology, such as the impact of depression and fear of aging among caregivers during the COVID-19 pandemic.

Andreea C. Bică is currently a Teaching Assistant at Ovidius University of Constanța. Her research focuses on the impact of social anxiety on decision-making processes, particularly in relation to social rank theory, conformity, and reciprocity-driven behaviors.

References

Alden

L. E.

Phillips

(1990). An interpersonal analysis of social anxiety and depression. Cognitive Therapy and Research, 14(5), 499–512. https://doi.org/10.1007/BF01172970

Alden

L. E.

Taylor

C. T.

(2004). Interpersonal processes in social phobia. Clinical Psychology Review, 24(7), 857–882. https://doi.org/10.1016/J.CPR.2004.07.006

Allan

Gilbert

(1995). A social comparison scale: Psychometric properties and relationship to psychopathology. Personality and Individual Differences, 19(3), 293–299. https://doi.org/10.1016/0191-8869(95)00086-L

Alnaes

Torgersen

(1988). Major depression, anxiety disorders and mixed conditions Childhood and precipitating events. Acta Psychiatrica Scandinavica, 78(5), 632–638. https://doi.org/10.1111/J.1600-0447.1988.TB06397.X

American Psychiatric Association . (2013). Diagnostic and statistical manual of mental disorders (5th ed.). APA.

Baker

S. L.

Heinrichs

Kim

H. J.

Hofmann

S. G.

(2002). The liebowitz social anxiety scale as a self-report instrument: A preliminary psychometric analysis. Behaviour Research and Therapy, 40(6), 701–715. https://doi.org/10.1016/S0005-7967(01)00060-2

Bică

Crețu

R. Z.

(2022). The carrot and the stick: How guilt and shame facilitate reciprocity-driven cooperation. Cognition and Emotion, 37(1), 117–127. DOI: 10.1080/02699931.2022.2146057.

Bică

Crețu

R. Z.

Podina

I. R.

(2021). Money versus social rank: An empirical investigation of unfairness in social anxiety. Cognitive Therapy and Research, 45(4), 642–651. https://doi.org/10.1007/S10608-020-10188-Z/FIGURES/3

Blount

(1995). When social outcomes Aren′t fair: The effect of causal attributions on preferences. Organizational Behavior and Human Decision Processes, 63(2), 131–144. https://doi.org/10.1006/obhd.1995.1068

10.

Clark

D. M.

Wells

(1995). A cognitive model of social phobia. In Heimberg

R. G.

Liebowitz

M. R.

Hope

D. A.

Schneier

F. R.

(Eds.), Social phobia: Diagnosis, assessment, and treatment (pp. 69–93). The Guilford Press.

11.

Cremers

H. R.

Veer

I. M.

Spinhoven

Rombouts

S. A. R. B.

Roelofs

(2015). Neural sensitivity to social reward and punishment anticipation in social anxiety disorder. Frontiers in Behavioral Neuroscience, 8(JAN), 439. https://doi.org/10.3389/FNBEH.2014.00439/ABSTRACT

12.

Crusius

Lange

(2014). What catches the envious eye? Attentional biases within malicious and benign envy. Journal of Experimental Social Psychology, 55(1), 1–11. https://doi.org/10.1016/J.JESP.2014.05.007

13.

Darcy

Davila

Beck

J. G.

(2005). Is social anxiety associated with both interpersonal avoidance and interpersonal dependence? Cognitive Therapy and Research, 29(1), 171–186.

14.

Davila

Beck

J. G.

(2002). Is social anxiety associated with impairment in close relationships? A preliminary investigation. Behavior Therapy, 33(3), 427–446. https://doi.org/10.1016/S0005-7894(02)80037-5

15.

Faul

Erdfelder

Lang

A. G.

Buchner

(2007). G*Power 3: A flexible statistical power analysis program for the social, behavioural, and biomedical sciences. Behaviour Research Methods, 39(2), 175–191. https://doi.org/10.3758/BF03193146

16.

Fehr

Camerer

C. F.

(2007). Social neuroeconomics: The neural circuitry of social preferences. Trends in Cognitive Sciences, 11(10), 419–427. https://doi.org/10.1016/J.TICS.2007.09.002

17.

Feng

Cao

Mobbs

(2018). The pursuit of social acceptance: aberrant conformity in social anxiety disorder. Soc Cogn Affect Neurosci, 13(8), 809–817. DOI: 10.1093/scan/nsy052.

18.

Festinger

(1954). A theory of social comparison processes. Human Relations, 7(1), 117–140.

19.

Fresco

D. M.

Coles

M. E.

Heimberg

R. G.

Liebowitz

M. R.

Hami

Stein

M. B.

Goetz

(2001). The liebowitz social anxiety scale: A comparison of the psychometric properties of self-report and clinician-administered formats. Psychological Medicine, 31(6), 1025–1035. https://doi.org/10.1017/S0033291701004056

20.

Gilbert

(2000). The relationship of shame, social anxiety and depression: The role of the evaluation of social rank. Clinical Psychology and Psychotherapy Clin. Psychol. Psychother, 7(1), 174–189.

21.

Gilbert

Trower

(2001). Evolution and process in social anxiety. In International handbook of social anxiety: Concepts, research and interventions relating to the self and shyness (pp. 259–279). John Wiley & Sons Ltd.

22.

Goodman

F. R.

Kelso

K. C.

Wiernik

B. M.

Kashdan

T. B.

(2021). Social comparisons and social anxiety in daily life: An experience-sampling approach. Journal of Abnormal Psychology, 130(5), 468–489. https://doi.org/10.1037/ABN0000671

23.

Güth

Schmittberger

Schwarze

(1982). An experimental analysis of ultimatum bargaining. Journal of Economic Behavior & Organization, 3(4), 367–388. https://doi.org/10.1016/0167-2681(82)90011-7

24.

Haber

S. N.

Knutson

(2010). The reward circuit: Linking primate anatomy and human imaging. Neuropsychopharmacology: Official Publication of the American College of Neuropsychopharmacology, 35(1), 4–26. https://doi.org/10.1038/NPP.2009.129

25.

Harlé

K. M.

Allen

J. J. B.

Sanfey

A. G.

(2010). The impact of depression on social economic decision making. Journal of Abnormal Psychology, 119(2), 440–446. https://doi.org/10.1037/A0018612

26.

Harlé

K. M.

Sanfey

A. G.

(2007). Incidental sadness biases social economic decisions in the ultimatum game. Emotion, 7(4), 876–881. https://doi.org/10.1037/1528-3542.7.4.876

27.

Hirschfeld

R. M. A.

Klerman

G. L.

Gough

H. G.

Barrett

Korchin

S. J.

Chodoff

(1977). A measure of interpersonal dependency. Journal of Personality Assessment, 41(6), 610–618. https://doi.org/10.1207/S15327752JPA4106_6

28.

Kimbrel

N. A.

(2008). A model of the development and maintenance of generalized social phobia. Clinical Psychology Review, 28(4), 592–612. https://doi.org/10.1016/J.CPR.2007.08.003

29.

Knoch

Pascual-Leone

Meyer

Treyer

Fehr

(2006). Diminishing reciprocal fairness by disrupting the right prefrontal cortex. Science (New York, N.Y.), 314(5800), 829–832. https://doi.org/10.1126/SCIENCE.1129156

30.

McCarthy

P. A.

Morina

(2020). Exploring the association of social comparison with depression and anxiety: A systematic review and meta-analysis. Clinical Psychology & Psychotherapy, 27(5), 640–671. https://doi.org/10.1002/CPP.2452

31.

Moscovitch

D. A.

(2009). What is the core fear in social phobia? A new model to facilitate individualized case conceptualization and treatment. Cognitive and Behavioral Practice, 16(2), 123–134. https://doi.org/10.1016/J.CBPRA.2008.04.002

32.

Preacher

K. J.

Hayes

A. F.

(2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40(3), 879–891. https://doi.org/10.3758/BRM.40.3.879

33.

Rodebaugh

T. L.

Bielak

Vidovic

Moscovitch

D. A.

(2016). The effects of social anxiety on interpersonal evaluations of warmth and dominance. Journal of Anxiety Disorders, 38(1), 68–78. https://doi.org/10.1016/J.JANXDIS.2016.01.002

34.

Rodebaugh

T. L.

Lim

M. H.

Shumaker

E. A.

Levinson

C. A.

Thompson

(2015). Social anxiety and friendship quality over time. Cognitive Behaviour Therapy, 44(6), 502–511. https://doi.org/10.1080/16506073.2015.1062043

35.

Rodebaugh

T. L.

Tonge

N. A.

Weisman

J. S.

Lim

M. H.

Fernandez

K. C.

Bogdan

(2017). The behavioral economics of social anxiety disorder reveal a robust effect for interpersonal traits. Behaviour Research and Therapy, 95(1), 139–147. https://doi.org/10.1016/J.BRAT.2017.06.003

36.

Rytwinski

N. K.

Fresco

D. M.

Heimberg

R. G.

Coles

M. E.

Liebowitz

M. R.

Cissell

Hofmann

S. G.

(2009). Screening for social anxiety disorder with the self-report version of the Liebowitz Social Anxiety Scale. Depression and Anxiety, 26(1), 34–38. https://doi.org/10.1002/DA.20503

37.

Salamone

J. D.

(1994). The involvement of nucleus accumbens dopamine in appetitive and aversive motivation. Behavioural Brain Research, 61(2), 117–133. https://doi.org/10.1016/0166-4328(94)90153-8

38.

Stigler

(1950). The development of utility theory. I. Journal of Political Economy, 58(4), 307–327. https://doi.org/10.1086/256962

39.

Suls

Wills

T. A.

(Eds.), (1991). Social comparison: Contemporary theory and research. Lawrence Erlbaum Associates, Inc.

40.

Thielmann

Spadaro

Balliet

Amsterdam

V. U.

(2020). Personality and prosocial behavior: A theoretical framework and meta-analysis. https://doi.org/10.1037/bul0000217.supp

41.

Tone

E. B.

Nahmias

Bakeman

Kvaran

Brosnan

S. F.

Fani

Schroth

E. A.

(2019). Social anxiety and social behaviour: A test of predictions from an evolutionary model. Clinical Psychological Science, 7(1), 110–126. https://doi.org/10.1177/2167702618794923

42.

Trower

Gilbert

(1989). New theoretical conceptions of social anxiety and social phobia. Clinical Psychology Review, 9(1), 19–35. https://doi.org/10.1016/0272-7358(89)90044-5

43.

Versella

M. V.

Piccirillo

M. L.

Potter

C. M.

Olino

T. M.

Heimberg

R. G.

(2016). Anger profiles in social anxiety disorder. Journal of Anxiety Disorders, 37(1), 21–29. https://doi.org/10.1016/J.JANXDIS.2015.10.008

44.

Wood

J. V.

(1989). Theory and research concerning social comparisons of personal attributes. Psychological Bulletin, 106(1), 231–248.

45.

Yamagishi

Horita

Mifune

Hashimoto

Shinada

Simunovic

(2012). Rejection of unfair offers in the ultimatum game is no evidence of strong reciprocity. Proceedings of the National Academy of Sciences of the United States of America, 109(50), 20364–20368. https://doi.org/10.1073/pnas.1212126109

Supplementary Material

Please find the following supplemental material available below.

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

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

0.00 MB

0.50 MB