Event-related potentials of social comparisons in depression and social anxiety

doi:10.21203/rs.3.rs-2762762/v1

Download PDF

Article

Event-related potentials of social comparisons in depression and social anxiety

https://doi.org/10.21203/rs.3.rs-2762762/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Social comparison is central in human life and can be especially challenging in depression and social anxiety. We assessed event-related potentials and emotions when receiving feedback on both their own and a co-player’s performance, in participants with depression and/or social anxiety (n=63) and healthy controls (n=72). Participants reported more negative emotions for downward (being better than someone else) and upward (being worse) comparisons vs. even outcomes, with these effects being stronger in depression and social anxiety. At the Medial Frontal Negativity, both controls and depressed participants showed enhanced negativity for upward (participant wrong, co-player correct) comparison vs. both the participant and co-player performing wrong. Socially anxious subjects showed the opposite effect, possibly due to greater expectations about being worse than others. The P300 decreased for downward and upward comparisons compared to even outcomes, which may relate to the higher levels of conflict of social inequality. Depressed and socially anxious subjects showed a blunted P300 increase over time in response to the task outcomes, suggesting deficits in allocating resources for the attention of incoming social information. This may relate to difficulties in these disorders in learning how to deal with the cognitive/emotional demands of social comparison.

Biological sciences/Neuroscience/Cognitive neuroscience

Biological sciences/Neuroscience/Social neuroscience

Biological sciences/Neuroscience

social comparison

depression

social anxiety

event-related potentials

Social comparison is central to human life. People naturally tend to compare their attitudes and abilities to those of others ¹. Whether we consider ourselves tall/short, clever/silly, successful/loser, handsome/ugly, earning good money or not is a judgment that we tend to make in comparative terms ². Comparisons can be made with someone considered to be better (upward), equal (lateral), or worse (downward) than oneself ³. Upward comparisons are useful for learning and improving abilities but are often associated with negative affect and can be threatening to self-esteem. Downward comparisons can be positive for self-enhancement, but depending on the context they can also imply negative feelings such as guilt and concerns about eliciting negative emotions (e.g. envy) in others ^3,4.

Social comparison can be especially difficult for people with disorders such as depression and social anxiety ^5,6. According to cognitive theories, social comparisons can play an important role in these disorders triggering negative evaluations about the self, low mood and social avoidance ^4,7. Consistent with these theories,, significant evidence shows that individuals with symptoms of depression and social anxiety interpret social comparisons in a less self-serving way, tend to believe that others are better than they are and demonstrate strong negative emotions and social avoidance when facing social comparison situations ^5,6,8,9. However, despite the crucial role of social comparisons in the maintenance of symptoms ⁶ in these disorders, no studies have examined yet the neural substrates of social comparison in depression and social anxiety.

In the last decade, there has been a growing interest in studying the neural substrates of social comparisons in healthy subjects ². Studies using electroencephalography (EEG) have shown at least three event-related potentials (ERP) components associated with social comparisons: the Medial Frontal Negativity (MFN), the P300, and the Late Positive Potential (LPP) ^10–14.

The MFN typically presents in frontocentral regions 200-350 ms after feedback, being more negative for negative than positive events ¹⁵. It is thought to reflect reward prediction errors and motivation/affective processes associated with outcome evaluation. During social comparisons, the MFN has been observed to be more negative both for downward and upward comparisons vs. even outcomes ^10,11,16,17. This modulation has been interpreted as a prediction error and as a motivation/affective response in the face of inequality ^10,11.

The P300 is a positive centro-parietal peak at 300-600 ms following feedback. It has been related to attention, motivation, and other cognitive processes, with larger amplitudes for motivational/salient stimuli ¹⁸. Interestingly, the P300 is also sensitive to the cognitive load, decreasing in amplitude with increasing task difficulty and when other tasks compete for cognitive resources ^19,20. In particular, the P300 decreases in response to conflict situations that require cognitive effort ²¹. For example, the P300 has been observed to decrease when participants make deceptive vs. truthful responses ^21–23 and in response to proposals from others with a medium degree of unfairness vs. more extreme fair/unfair propositions ²⁴. In line with this, a social comparison study observed a decreased P300 for downward and upward comparisons compared to even outcomes ¹⁶, which may relate to the higher levels of social conflict that unequal comparisons entail ¹¹, demanding cognitive resources that compete with processes that underlie the P300.

The LPP is a positive centro-parietal deflection starting approximately 300 ms post-stimulus ²⁵. It is considered to reflect sustained attention and affective processing of outcomes, and it is usually larger for negative than positive feedback ¹⁵. Social comparison studies have reported an increased LPP for large differences in earnings between the participant and other players ^11,13, suggesting that upward and downward comparisons have a stronger motivational relevance and elicit higher levels of autonomic arousal compared to even outcomes ¹¹.

ERPs have also become a fundamental tool in the search for the neural substrates of psychopathology ²⁶. Regarding the MFN, some studies have reported a greater differentiation between negative and positive outcomes in depression and social anxiety, linked to an increased impact of aversive events ^27–29. However, other studies have reported a reduced MFN differentiation between positive and negative outcomes in depression and anxiety, associated with negative biases in outcome expectations ^30,31. A decreased MFN to positive stimuli in depression has also been reported, related to blunted reward processing ^32,33.

The P300 has been reported to be decreased in depression and social anxiety during oddball tasks ^34–37, and in depression during flanker ^38,39 and reward tasks ^30,40. These findings have been related to reduced motivation and allocation of resources for the processing of incoming stimuli in these disorders. It is thought that negative cognitive biases, rumination, and worries may underlie impairments in attention in these populations competing for cognitive resources ^41–43.

The LPP has been reported to be reduced in depression in response to pleasant pictures ^44,45 and also in response to unpleasant faces ^46–48; however, an increased LPP has also been observed in depression, in response to negative and self-referential stimuli ⁴⁹. In social anxiety, an increased LPP has been observed possibly related to enhanced emotional processing ^50–52, while a blunted LPP has also been reported linked to attentional avoidance ⁵³.

Here we aimed to investigate emotional and neural responses to social comparisons in participants with symptoms of depression and/or social anxiety and healthy controls. It was hypothesized that downward and especially upward comparisons would be associated with higher levels of negative emotions compared to even outcomes. We expected social comparisons to modulate the MFN, P300, and LPP. Based on the above findings, it was expected that the MFN would show increased negativity for downward and upward comparisons compared to even outcomes; that the P300 would attenuate for downward and upward comparisons compared to even outcomes due to the higher levels of conflict that unequal comparisons entail; and that the LPP would show an increased positivity for upward and downward comparisons vs. even outcomes.

Regarding the effect of psychopathology, it was expected that depression and social anxiety would be associated with stronger negative emotions for unequal comparisons. With respect to ERPs, an enhanced MFN negativity for downward and upward comparisons could be hypothesized in these disorders, due to negative cognitive biases leading to a deeper processing of conflictive/negative events. However, for upward comparison a diminished MFN negativity could also be expected due to negative biases in outcome expectation (i.e., greater expectation of not being as good as others). For the P300, we hypothesized decreased amplitudes in depression and social anxiety especially for downward and upward comparisons, due to negative cognitive biases, rumination, and worries competing for resources with the processes that underlie the P300. For the LPP, both an enhanced and a diminished positivity could be expected in response to downward and upward comparisons since both enhanced emotional processing or emotional disconnection/avoidance could be possible in these disorders. To our knowledge, this is the first study investigating the neural substrates of social comparison in depression and social anxiety.

Participants

This study was conducted following the Declaration of Helsinki and was approved by the Research Ethics Committee of the Faculty of Psychology, Universidad de la República. Written informed consent was obtained from all participants. Subjects (18-35 years old) completed the Beck Depression Inventory-II (BDI-II)^54,55 and the Liebowitz Social Anxiety Scale (LSAS) ^56,57 on a website advertised through the university’s social networks. Applicants were invited to a recruitment session, where they were screened for psychiatric symptoms using the MINI International Neuropsychiatric Interview (MINI-Plus, version 5.0.0) ^58,59. Two groups of participants were formed: a group with symptoms of major depression and/or social anxiety (MD-SA, n=63) and a group of healthy controls (n=72). Inclusion criteria for the MD-SA group were: satisfying MINI-Plus criteria for an episode of major depression and scoring ≥14 on the BDI-II and/or satisfying MINI-Plus criteria for social anxiety and scoring ≥55 on the LSAS, and at least three weeks of not taking psychiatric medication. Controls had no current or past history of psychiatric disorders. The two groups did not significantly differ in sex, age, years of education, etc. (Table 1). The MD-SA group was composed of 23 volunteers meeting criteria only for MD, 30 only for SA, and 10 for both disorders (see Supplementary Material for psychological rating scales).

Since depression and social anxiety are highly comorbid and share a similar affective and cognitive profile ^60,61, statistical analyses were performed comparing controls vs. the whole MD-SA group to increase statistical power for identification of common neural substrates. However, additional analyses were also performed splitting the MD-SA group into participants presenting only MD and only SA (participants with both MD and SA were not included as they were only 10) to test for specific effects of these disorders.

Behavioral task

The social comparison task was similar to tasks previously used in social comparison studies ¹⁰ (Figure 1). In each trial, a blue circle was presented, which changed color to green after 2–2.5 seconds. Participants were told to press the spacebar one second after the color change and that they would earn one/zero points depending on whether they had done a correct/wrong time estimation. Participants were told that they would play simultaneously with another player in a nearby room and that on every trial, both would receive feedback on their own and the co-player’s performance. During the task, four feedback combinations were possible: “You correct-Other correct”, “You correct-Other wrong”, “You wrong-Other correct”, “You wrong-Other wrong”. It was stressed that the earnings of both subjects were independent. In reality, there was no real co-player, and the feedback was preprogrammed to be positive in 50% of trials for both players.

Before the task, participants were presented with the supposed co-player and told that later after the experiment they would be reunited to discuss the task and that each would receive a monetary reward based on their respective points. It was controlled that the participant did not know the co-player, that the interaction was brief, and that both were of the same sex.

The task was programmed in PsychoPy2 and was composed of 168 trials, 42 in each condition. Trials were divided into three blocks (conditions were balanced between the blocks). Participants also performed the Ultimatum Game task, which will be reported elsewhere. The order of tasks was counterbalanced across participants.

After the experiment, participants rated on nine-point Likert scales: their emotions in response to the task’s conditions, their emotions regarding meeting again with the co-player, and their perception of their own and the co-players performance. Participants were debriefed regarding the cover story and rewarded for their collaboration (participants received a cinema ticket or pen drive).

EEG recording and preprocessing

EEG was recorded using 64 Ag/AgCl active channels (Biosemi, Amsterdam, Netherlands) mounted in an elastic cap following the 10/20 international system ⁶². Data were sampled at 256 Hz and online filtered using a fifth-order low-pass sinc filter with a 52 Hz cut-off (-3 dB). Following the BioSemi design, the voltage at each active electrode was recorded with respect to a common mode sense active electrode and a Driven Right Leg passive electrode, replacing the ground electrode. Data were preprocessed offline using the FieldTrip toolbox in MATLAB ⁶³. Data were re-referenced to an average mastoid reference, and offline filtered using a band-pass Butterworth filter at 0.1-30 Hz with a two-pass zero-phase forward and reverse direction (order: 4). Epochs were cut at 200 ms before and 800 ms after feedback. ERPs from each subject and condition were averaged separately, and a baseline voltage averaged over the 200 ms interval preceding feedback was subtracted from these averages (see Supplementary Material).

Relevant ERP components were identified by visual inspection of the time series average (over all subjects and all conditions) and topographical distributions. The MFN was computed as the mean amplitude in the 200-300 ms time window averaged over frontocentral electrodes. The P300 was computed as the mean amplitude in the 300-450 ms time window averaged over middle-line and centro-parietal electrodes. The LPP was computed as the mean amplitude in the 450-800 ms time window averaged over centro-parietal electrodes. Selected time windows and topographies align with those reported in similar studies ^10,13,16,64 (selected electrodes are listed in Figure 3).

Statistical analyses

Emotional reports were analyzed using three-way ANOVAs with participant and co-player outcomes as within-subjects factors and the group as the between-subjects factor. The Greenhouse-Geisser correction was applied in cases where the sphericity hypothesis was not met. Effect sizes were calculated using the partial Eta squared method (η2p). Pairwise comparisons were tested using paired and independent-sample t-tests. For independent-sample t-tests, Levene's correction was applied in cases where the hypothesis of equality of variances was not met, and effect sizes were calculated as Cohen's Delta (δ)=T*√(1/n1+1/n2).

The classical approach to studying ERPs involves averaging epochs of the same condition in each subject. This approach obscures modulations of the ERPs along the task. Recent approaches have used Mixed Linear Models (MLM) to explore ERPs’ modulations along the task ^32,65. We implemented MLMs, selecting models with the best fitting for each ERP based on the Akaike Information Criterion (AIC). For all ERPs, the model included mean voltage as the response variable, subject as a random effect, and self-outcome, other-outcome, group (Control, MD-SA; or Control, MD, SA), trial, and all their interactions as fixed effects. Post-hoc comparisons were performed using the Holm-Bonferroni correction. Statistical analysis was done on Rstudio (version 1.0.153) and SPSS (version 22).

Of the sample of 135 participants, 117 attended the experimental session. For emotional analyses, four participants were excluded for not believing the cover story, leading to a sample of n=113 (59 Controls, 54 MD-SA). In addition, for the EEG analysis 15 participants were excluded due to: technical problems during data acquisition (n=3), artifacts in the EEG signal (n=11), and unbalanced trials between conditions after excluding artifacts (n=1), leading to a sample of n=98 (49 Controls, 49 MD-SA).

Emotional results

The mixed Anova with factors self-outcome, other-outcome and group (Control, MD-SA) (see Supplementary Materials for the main effects of self-outcome, other-outcome. group and self-outcome*group interaction) identified a significant self-outcome*other-outcome interaction for the emotions of happiness (F_1,111=4.08; p<0.046; η²_p=0.04), relief (F_1,111=22.99; p<0.001; η²_p=0.17), sadness (F_1,111=8.97; p<0.003; η²_p=0.7), guilt (F_1,111=19.03; p<0.001; η²_p=0.14), shame (F_1,111=40.11; p<0.001; η²_p=0.26), envy (F_1,111=46.92; p<0.001; η²_p=0.29), anger (F_1,111=8.77; p=0.004; η²_p=0.07), nervousness (F_1,111=16.10; p<0.001; η²_p=0.12) and disappointment (F_1,111=13.72; p<0.001; η²_p=0.11) (Figure 2A). Post hoc pairwise comparisons identified that the even correct outcome (“You correct-Other correct”), elicited less sadness (t₁₁₂=3.94, p<0.001), guilt (t₁₁₂=4.15, p<0.001), and shame (t₁₁₂=2.46, p=0.015) and more envy (t₁₁₂=-2.23, p=0.027) than downward comparisons (“You correct-Other wrong”). Upward comparison (“You wrong-Other correct”) elicited more anger (t₁₁₂=2.78, p=0.006), nervousness (t₁₁₂=4.66, p<0.001), shame (t₁₁₂=5.64, p<0.001), disappointment (t₁₁₂=4.74, p<0.001) and envy (t₁₁₂=6.63, p<0.001), and less relief (t₁₁₂=-7.28, p<0.001) than the even wrong outcome ( “You wrong-Other wrong”). These findings show that emotions were modulated by social comparison.

A significant self-outcome*other-outcome*group (Control, MD-SA) interaction was observed for the emotions of nervousness (F_1,111=5.97; p=0.016; η²_p=0.05), guilt (F_1,111=8.52; p=0.004; η²_p=0.07), shame (F_1,111=9.10; p=0.003; η²_p=0.08) and disappointment (F_1,111=4.70; p=0.032; η²_p=0.04) (Figure 2A). Follow-up of these interactions showed that even correct comparisons (“You correct-Other correct”) did not trigger different emotional responses between groups, while downward comparison (“You correct-Other wrong”) led to MD-SA participants reporting more nervousness (t_86.98=-4.15, p<0.001), guilt (t_61.55=-3.58, p<0.001) and shame (t₅₃=-3.80, p<0.001) than controls. During upward comparison (“You wrong-Other correct”), MD-SA participants reported more nervousness (t_94.93=-4.54, p<0.001), guilt (t_75.61=-2.88, p<0.001), shame (t_80.20=-5.45, p<0.001) and disappointment (t₁₁₁=-3.48, p<0.001) than controls; while during even wrong comparisons (“You wrong-Other wrong”) MD-SA participants reported more shame (t_54.96=-4.39, p<0.001) than controls (with this between-group difference not being as strong as for the “You wrong-Other correct” outcome (t_221.97=-3.21, p=0.002)).

MD-SA participants reported less happiness (t_105.56=2.92; p=0.004; δ=0.55), more nervousness (t_90.80=-7.48; p<0.001; δ=-1.40), shame (t_79.92=-6.23; p<0.001; δ=- 1.17), anger (t_56.78=-2.05; p=0.045; δ=-0.38) and sadness (t_60.16=-2.19; p=0.032; δ=-0.41) than controls about anticipating meeting again with the co-player.

All participants rated their performance as less accurate than the co-player (F_1,110=32.98; p<0.001). MD-SA participants evaluated their performance as less accurate than controls (F_1,110=4.58; p=0.035). A significant interaction between the group and the subject object of the evaluation was found (F_1,110=7.60; p=0.007), with groups not differing in their perception of the co-player’s performance (p=0.092), but MD-SA participants rating their own performance as less accurate than controls (p=0.003) (Figure 2B).

Additional analyses splitting the group factor in three levels (Controls, MD, SA) (see Supplementary Materials) showed that both the MD and SA subgroups contributed to the between-group differences in emotions between the Control and MD-SA groups (Supplementary Figure 1 and Supplementary Analyses).

Event-Related Potential results

MFN

Significant main effects of self-outcome (F_1,14105.98=5.37; p=0.02) and other-outcome (F_1,14105.15=12.5 2; p<0.001) were found, with larger negativity for negative vs. positive feedback. A significant self-outcome*other-outcome*group (Control, MD-SA) interaction (F_1,14105.58=5.19; p=0.023) was found. In both groups, downward comparisons (“You correct-Other wrong”) elicited a larger negativity than the even correct comparison (“You correct-Other correct”). However, only for controls, upward comparison (“You wrong-Other correct”) elicited larger negativity than the even wrong outcome (“You wrong-Other wrong”) (p=0.05), while these two outcomes did not show significant differences in MD-SA participants (Figure 3 A2).

Interestingly, the MLM with the group factor divided into three levels (Control, MD, SA) showed a significant self-outcome*other-outcome*group interaction (F_2,13282.26=4.58; p=0.010). In all groups, the MFN was more negative for "You correct-Other wrong" compared to "You correct-Other correct" (p<0.002). However, while in both controls (p=0.05) and MD (p=0.02) participants, the MFN was more negative for "You wrong-Other correct" compared to "You wrong-Other wrong", in the SA group, the opposite effect was observed (i.e. the MFN was more positive for "You wrong-Other correct" compared to "You wrong-Other wrong") (p=0.05). This difference was significant between the SA and the other two groups (vs. MD: p=0.002, vs. Control: p=0.006) (Figure 3 A3). See Table 2 and Supplementary Materials for additional results on the MFN.

P300

A significant self-outcome*other-outcome interaction was found (F_1,14105.02=60.45; p<0.001) with even conditions eliciting larger amplitudes than unequal comparisons. This is, the even correct outcome (“You correct-Other correct”) elicited a larger P300 than the downward comparison (“You correct-Other wrong”) (p<0.001), and the even wrong outcome (“You wrong-Other wrong”) elicited a larger P300 than the upward comparison (“You wrong-Other correct”) (p<0.001) (Figure 3 B2).

A significant trial*self-outcome*other-outcome was found (F_1,14104.95=14.845; p<0.001). When the participant was correct, the mean amplitude of P300 increased over trials (“You correct-Other correct” and “You correct-Other wrong”) (t_14108.92=2.48; p=0.013; β=0.001; t_14108.86=5.95; p<0.001; β=0.002 respectively). However, when the participant was wrong, the mean amplitude of P300 increased only when the co-player was also wrong (“You wrong-Other wrong”) (t_14109.40=-2.71; p<0.001; β=-0.001) but not when the co-player was correct (“You wrong-Other correct”) (t_14109.33=1.35; p=0.18; β=0.000), with these slopes being significantly different (t_14108.94=-2.80; p<0.001; β=-0.001) (Figure 3 B4).

In addition, a significant trial*group (Control, MD-SA) interaction was found (F_1,14105.86=7.70; p=0.006), with P300 increasing over time for controls (t_14.111.87=4.49; p<0.001; β=0.001) but not for MD-SA participants (t_14111.71=0.58; p=0.56; β=0.000). Since trial 120, the mean amplitude was larger for controls than for MD-SA participants (p<0.05) (Figure 3 B3). Both MD and SA subgroups contributed to this effect (Supplementary Figure 2 and Supplementary Analysis). See Table 2 and Supplementary Materials for additional results on the P300.

LPP

Significant main effects of self-outcome (F_1,14106.13=35.84; p<0.001) and other-outcome (F_1,14105.20=9.87; p=0.002) were found, with larger amplitudes for negative vs. positive feedback. Additionally, these factors interacted (F_1,14105.69=11.065; p=0.001), with unequal comparisons showing larger amplitudes than even outcomes. Specifically, downward comparison (“You correct-Other wrong”) elicited a larger amplitude than the even correct condition (“You correct-Other correct”) (p<0.001), and upward comparison (“You wrong-Other correct”) elicited a larger amplitude than the even wrong condition (“You wrong-Other wrong”) (p<0.001) (Figure 3 C2). There were no significant group effects with the group factor opened into two (Control, MD-SA) or three levels (Control, MD, SA) (see Table 2 and Supplementary Materials).

This study examined emotional and neural responses to social comparison in participants with symptoms of depression and/or social anxiety and healthy controls. The task was successful in eliciting social comparison processes. When participants were correct, they reported more negative feelings for downward comparison than for the even outcome. When participants were wrong, they reported more negative emotions for upward comparison than for the even outcome. These results are in agreement with the notion that upward comparison can lead to negative affect being a threat to self-esteem, and with the proposal that (depending on the context) downward comparisons can also elicit negative feelings ^3,66.

Crucially, the MD-SA group reported more negative feelings for downward and upward comparisons than healthy controls. This is consistent with evidence showing that depressed and socially anxious individuals are more sensitive to social comparison ^5,6, and with the theory that social comparison can act as a trigger for negative cognitive biases (in particular negative self-evaluations) in these populations ^4,67.

MFN

The MFN was more negative when participants were wrong than when they were correct, which is consistent with previous findings ^15,68–70. Interestingly, across participants, downward comparison elicited a more negative MFN than the even correct outcome. This MFN modulation for downward comparison is consistent with a previous study and may relate to the emotional processes involved in downward comparison ¹⁶. Interestingly, MFN modulations have been associated with empathy ^71,72, which could be acting during downward comparisons. In addition, the MFN modulation doe downward comparison could also reflect a warning signal linked to a potential social threat since outperforming others can trigger negative feelings and actions from those others ⁷³.

When the participant was wrong, it was observed that upward comparison elicited a more negative MFN than the even wrong outcome in controls and depressed participants, while the opposite was observed in the SA group.

The MFN modulation observed in controls and depressed participants for upward comparison is consistent with previous studies ^10,14,16, and it may reflect the negative emotional/motivational impact of upward comparison, as well as prediction error signals related to inequality. Interestingly, socially anxious individuals showed the opposite MFN modulation for upward comparison. It could be thought that upward comparison did not generate a negative emotional response in this group. However, this is not consistent with the emotional reports. Alternatively, it is possible that in social anxiety the upward comparison does not elicit a prediction error signal, due to socially anxious subjects not being as surprised for being worse than others, given their typical pessimistic expectations about themselves. The MFN effect on social anxiety is in line with a previous study showing a reduced MFN in high vs. low trait anxiety individuals in a monetary gambling task ⁷⁴. Nevertheless, replication of this finding is important since MFN studies in social anxiety are very scarce and have reported both a trend for an enhanced MFN linked to an accentuated negative response ²⁷, as well as an altered MFN related to abnormal expectations ⁷⁵.

P300

The P300 signal increased over time for positive but not for negative outcomes. During most of the task, the P300 was larger for self-correct than for self-wrong, consistent with previous social comparison studies ^13,14,16. This may relate to the higher motivational significance of positive vs. negative feedback.

In addition, the P300 was modulated by social comparison. Inequitable outcomes (both downward and upward comparisons) led to decreases in P300 compared to even outcomes, consistent with a previous study ¹⁶. The decreased P300 for inequitable outcomes may relate to the conflict, difficulty, and affective/cognitive load that these situations entail. It has been observed that the P300 decreases in conflict situations that engage cognitive processes that compete for resources ^21,24. In our task, both downward and upward comparisons are likely to imply higher levels of social conflict compared to the even outcomes, as they face individuals with social inequality. Interestingly, P300 amplitudes increased over time for all outcomes except for upward comparison, suggesting that, especially in this situation, it may be difficult for subjects to learn how to deal with conflict and liberate cognitive resources for the P300.

Crucially, the mean amplitude of the P300 increased over time in controls but not in MD-SA participants. This result suggests that in controls, as time progresses, there is a decrease in the level of conflict and cognitive/affective load that social comparison information elicits, allowing for a decrease in competitive processes for the P300. In contrast, this learning/facilitation effect would be blunted in the MD-SA group. Depression and social anxiety are characterized by a cognitive style marked by negative cognitive biases, rumination, and worries ^76,77. This cognitive style could be contributing to MD-SA participants lingering on the negative aspects of social comparison, precluding them from liberating resources for the processes behind the P300. In line with this, rumination and worries are thought to underlie impairments in attention, concentration, and memory in depression and anxiety, taking away resources that are needed for the performance of tasks ^41–43. However, it should be noted that the group factor did not interact with the task outcomes on the P300, suggesting that the MD-SA cognitive style similarly affected the processing of all task conditions at the P300. It would be worth it to further explore in future work whether psychopathology has a more specific effect on the P300 processing of social comparison outcomes.

LPP

In agreement with previous studies, inequitable outcomes (downward and upward comparisons) elicited a larger LPP than even outcomes ^11,13. Since the LPP is thought to reflect extended processing of the affective value of outcomes, these findings suggest higher emotional intensity and arousal of inequitable vs. even outcomes, which is in line with the emotional reports.

We did not observe group effects on the LPP. Both reduced and enhanced LPPs have been observed in depression ^46,49 and social anxiety ^51,53, suggesting that the effects of these disorders on the LPP are complex and likely dependent on specific task designs. Further research on the effect of these disorders on the LPP during social comparison is relevant, given our findings of increased emotional responses in MD-SA participants and the observed modulation of the LPP with social comparison.

Limitations

The sample was composed of university students between 18 and 35 years, primarily women, limiting the generalizability of results. In particular, it would be interesting to extend this work to the study of more severe clinical populations derived from the health system.

This study investigated for the first time neural responses to social comparison in depression and social anxiety. Individuals with these disorders showed enhanced negative emotions towards downward and upward comparisons. Interestingly, controls and depressed volunteers differentiated from socially anxious participants at the MFN, possibly due to socially anxious individuals not being as surprised for being worse than others. At the P300, downward and upward comparisons led to decreases compared to even outcomes, which may relate to the higher levels of conflict that social inequality implies. The MD-SA group showed a blunted P300 increase over time in response to the task outcomes, which may relate to difficulties in this group in dealing with social comparison information, precluding individuals from liberating resources that could be allocated for attending incoming social information. Findings may contribute to the understanding of the neural substrates of social comparison in people with depression and social anxiety.

Acknowledgments

We thank all the volunteers who took part in this study. We thank Camila Zugarramurdi and Emilia Fló for helping with the EEG analysis. We also thank Santiago Garat, Oscar Nievas, Octavio Santos, and Laura Uriarte for their help at the beginning of this project.

Author contributions

VP: data curation, funding acquisition, formal analysis, investigation, methodology, project administration, software, visualization, writing original draft, writing review and editing.

ENS: data curation, funding acquisition, investigation, methodology, project administration, writing review and editing

GFT: investigation, methodology, writing review and editing

AP: investigation, software, implementation, writing review and editing

FCC: methodology, writing review and editing

EMM: investigation, methodology, writing review and editing

DK: investigation, methodology, supervision, validation, writing review and editing

ÁC: investigation, methodology, supervision, writing review and editing

VBG: conceptualization, funding acquisition, investigation, methodology, project administration, resources, supervision, writing original draft, writing review and editing

Data Availability Statement

The data used in this study can be made available from the corresponding author on reasonable request.

Code Availability Statement

The code generated during this study can be made available from the corresponding author on reasonable request.

Funding

This work was supported by the Comisión Sectorial de Investigación Científica (CSIC), Universidad de la República (UdelaR), Uruguay, Agencia Nacional de Investigación e Innovación (ANII), Uruguay [AC and VBG, grant number POS_NAC_2015_1_109714 to VP]; Programa de Desarrollo de las Ciencias Básicas (PEDECIBA), Uruguay [AC and VBG], and Ministerio de Ciencia, Innovación y Universidades, Spain [grant number PID2021-124420NB-100 to DK].

Additional Information

The authors declare no conflicts of interest.

Festinger, L. A Theory of Social Comparison Processes. Hum. Relat. 7, 117–140 (1954).
Kedia, G., Mussweiler, T. & Linden, D. E. J. Brain mechanisms of social comparison and their influence on the reward system. NeuroReport 25, 1255–1265 (2014).
Buunk, A. P. & Gibbons, F. X. Social comparison: The end of a theory and the emergence of a field. Organ. Behav. Hum. Decis. Process. 102, 3–21 (2007).
Swallow, S. R. & Kuiper, N. A. Social comparison and negative self-evaluations: An application to depression. Clin. Psychol. Rev. 8, 55–76 (1988).
Buunk, B. P. & Brenninkmeyer, V. B. Social comparison processes among depressed individuals: Evidence for the evolutionary perspective on involuntary subordinate strategies? in Subordination and defeat: An evolutionary approach to mood disorders and their therapy 147–164 (Lawrence Erlbaum Associates Publishers, 2000).
McCarthy, P. A. & Morina, N. Exploring the association of social comparison with depression and anxiety: A systematic review and meta-analysis. Clin. Psychol. Psychother. 27, 640–671 (2020).
Antony, M. M., Rowa, K., Liss, A., Swallow, S. R. & Swinson, R. P. Social comparison processes in social phobia. Behav. Ther. 36, 65–75 (2005).
Fernández-Theoduloz, G. et al. Social avoidance in depression: A study using a social decision-making task. J. Abnorm. Psychol. 128, 234–244 (2019).
Uriarte-Gaspari, L. et al. Who do I want in my team: Social avoidance of high qualified partners in depression and social anxiety. J. Affect. Disord. Rep. 10, 100402 (2022).
Boksem, M. A. S., Kostermans, E. & De Cremer, D. Failing where others have succeeded: Medial Frontal Negativity tracks failure in a social context: MFN and failure in a social context. Psychophysiology 48, 973–979 (2011).
Luo, Y. et al. Social Comparison Manifests in Event-related Potentials. Sci. Rep. 5, 12127 (2015).
Qiu, J. et al. The impact of social comparison on the neural substrates of reward processing: An event-related potential study. NeuroImage 49, 956–962 (2010).
Wu, Y., Zhang, D., Elieson, B. & Zhou, X. Brain potentials in outcome evaluation: When social comparison takes effect. Int. J. Psychophysiol. 85, 145–152 (2012).
Zhang, X. et al. Neural correlates of acceptance and rejection in online speed dating: An electroencephalography study. Cogn. Affect. Behav. Neurosci. 22, 145–159 (2021).
Glazer, J. E., Kelley, N. J., Pornpattananangkul, N., Mittal, V. A. & Nusslock, R. Beyond the FRN: Broadening the time-course of EEG and ERP components implicated in reward processing. Int. J. Psychophysiol. 132, 184–202 (2018).
Qi, Y., Wu, H., Raiha, S. & Liu, X. Social value orientation modulates context-based social comparison preference in the outcome evaluation: An ERP study. Neuropsychologia 112, 135–144 (2018).
Zhang, H. et al. Context-based interpersonal relationship modulates social comparison between outcomes: an event-related potential study. Soc. Cogn. Affect. Neurosci. 16, 439–452 (2021).
San Martín, R. Event-related potential studies of outcome processing and feedback-guided learning. Front. Hum. Neurosci. 6, 304 (2012).
Kok, A. On the utility of P3 amplitude as a measure of processing capacity. Psychophysiology 38, 557–577 (2001).
Polich, J. Updating P300: An integrative theory of P3a and P3b. Clin. Neurophysiol. 118, 2128–2148 (2007).
Hu, X., Pornpattananangkul, N. & Nusslock, R. Executive control- and reward-related neural processes associated with the opportunity to engage in voluntary dishonest moral decision making. Cogn. Affect. Behav. Neurosci. 15, 475–491 (2015).
Johnson, R., Henkell, H., Simon, E. & Zhu, J. The self in conflict: The role of executive processes during truthful and deceptive responses about attitudes. NeuroImage 39, 469–482 (2008).
Johnson, R., Barnhardt, J. & Zhu, J. The deceptive response: effects of response conflict and strategic monitoring on the late positive component and episodic memory-related brain activity. Biol. Psychol. 64, 217–253 (2003).
Fabre, E. F., Causse, M., Pesciarelli, F. & Cacciari, C. Sex and the money- How gender stereotypes modulate economic decision-making: An ERP study. Neuropsychologia 75, 221–232 (2015).
Hajcak, G., Dunning, J. P. & Foti, D. Motivated and controlled attention to emotion: Time-course of the late positive potential. Clin. Neurophysiol. 120, 505–510 (2009).
McLoughlin, G., Makeig, S. & Tsuang, M. T. In search of biomarkers in psychiatry: EEG-based measures of brain function. Am. J. Med. Genet. B Neuropsychiatr. Genet. 165, 111–121 (2014).
Judah, M. R. et al. Electrocortical Evidence of Enhanced Performance Monitoring in Social Anxiety. Behav. Ther. 47, 274–285 (2016).
Mueller, E. M., Pechtel, P., Cohen, A. L., Douglas, S. R. & Pizzagalli, D. A. Potentiated processing of negative feedback in depression is attenuated by anhedonia. Depress. Anxiety 32, 296–305 (2015).
Tucker, D. M., Luu, P., Frishkoff, G., Quiring, J. & Poulsen, C. Frontolimbic response to negative feedback in clinical depression. J. Abnorm. Psychol. 112, 667–678 (2003).
Foti, D. & Hajcak, G. Depression and reduced sensitivity to non-rewards versus rewards: Evidence from event-related potentials. Biol. Psychol. 81, 1–8 (2009).
Gu, R., Huang, Y. X. & Luo, Y. J. Anxiety and feedback negativity. Psychophysiology 47, 961–967 (2010).
Brush, C. J., Ehmann, P. J., Hajcak, G., Selby, E. A. & Alderman, B. L. Using Multilevel Modeling to Examine Blunted Neural Responses to Reward in Major Depression. Biol. Psychiatry Cogn. Neurosci. Neuroimaging 3, 1032–1039 (2018).
Liu, W. hua et al. The influence of anhedonia on feedback negativity in major depressive disorder. Neuropsychologia 53, 213–220 (2014).
Gangadhar, B. N., Ancy, J., Janakiranaiah, N. & Umapathy, C. P300 amplitude in non-bipolar, melancholic depression. J. Affect. Disord. 28, 57–60 (1993).
Nan, C. et al. The P300 component decreases in a bimodal oddball task in individuals with depression: An event-related potentials study. Clin. Neurophysiol. 129, 2525–2533 (2018).
Sachs, G. et al. P300 event-related potentials and cognitive function in social phobia. Psychiatry Res. - Neuroimaging 131, 249–261 (2004).
Zhou, L. et al. Abnormalities in P300 components in depression: an ERP-sLORETA study. Nord. J. Psychiatry 73, 1–8 (2019).
Klawohn, J., Santopetro, N., … A. M.- & 2020, undefined. Reduced P300 in depression: Evidence from a flanker task and impact on ERN, CRN, and Pe. Wiley Online Libr. 57, (2020).
Santopetro, N. J., Brush, C. J., Bruchnak, A., Klawohn, J. & Hajcak, G. A reduced P300 prospectively predicts increased depressive severity in adults with clinical depression. Psychophysiology 58, (2021).
Santopetro, N. J., Brush, C. J., Burani, K., Bruchnak, A. & Hajcak, G. Doors P300 moderates the relationship between reward positivity and current depression status in adults. J. Affect. Disord. 294, 776–785 (2021).
Hirsch, C. R. & Mathews, A. A cognitive model of pathological worry. Behav. Res. Ther. 50, 636–646 (2012).
Keller, A. S., Leikauf, J. E., Holt-Gosselin, B., Staveland, B. R. & Williams, L. M. Paying attention to attention in depression. Transl. Psychiatry 9, 279 (2019).
van Vugt, M. K., van der Velde, M., & ESM-MERGE Investigators. How Does Rumination Impact Cognition? A First Mechanistic Model. Top. Cogn. Sci. 10, 175–191 (2018).
Klawohn, J., Burani, K., Bruchnak, A., Santopetro, N. & Hajcak, G. Reduced neural response to reward and pleasant pictures independently relate to depression. Psychol. Med. 51, 741–749 (2021).
Weinberg, A., Perlman, G., Kotov, R. & Hajcak, G. Depression and reduced neural response to emotional images: Distinction from anxiety, and importance of symptom dimensions and age of onset. J. Abnorm. Psychol. 125, 26–39 (2016).
Bauer, E. A. & MacNamara, A. Comorbid anxiety and depression: Opposing effects on the electrocortical processing of negative imagery in a focal fear sample. Depress. Anxiety 38, 690–699 (2021).
Foti, D., Olvet, D. M., Klein, D. N. & Hajcak, G. Reduced electrocortical response to threatening faces in major depressive disorder. Depress. Anxiety 27, 813–820 (2010).
Granros, M., MacNamara, A., Klumpp, H., Phan, K. L. & Burkhouse, K. L. Neural reactivity to affective stimuli and internalizing symptom dimensions in a transdiagnostic patient sample. Depress. Anxiety 39, 770–779 (2022).
Benau, E. M. et al. Increased neural sensitivity to self-relevant stimuli in major depressive disorder. Psychophysiology 56, e13345 (2019).
Hagemann, J., Straube, T. & Schulz, C. Too bad: Bias for angry faces in social anxiety interferes with identity processing. Neuropsychologia 84, 136–149 (2016).
Kinney, K. L., Burkhouse, K. L. & Klumpp, H. Self-report and neurophysiological indicators of emotion processing and regulation in social anxiety disorder. Biol. Psychol. 142, 126–131 (2019).
Wieser, M. J. & Moscovitch, D. A. The Effect of Affective Context on Visuocortical Processing of Neutral Faces in Social Anxiety. Front. Psychol. 6, (2015).
Kraft, J. D. et al. Documenting the late positive potential towards self-imagery within social anxiety. Psychiatry Res. Neuroimaging 321, 111457 (2022).
Beck, A. T. An Inventory for Measuring Depression. Arch. Gen. Psychiatry 4, 561 (1961).
Sanz, J., Perdigón, A. L. & Vázquez, C. Adaptación española del Inventario para la Depresión de Beck-II (BDI-II): 2. Propiedades psicométricas en población general. CLÍNICA SALUD 34 (2003).
Bobes, J. et al. Validation of the Spanish version of the Liebowitz social anxiety scale, social anxiety and distress scale and Sheehan disability inventory for the evaluation of social phobia. Med. Clin. (Barc.) 112, 530–538 (1999).
Liebowitz, M. R. Social Phobia. in Modern Trends in Pharmacopsychiatry (ed. Klein, D. F.) vol. 22 141–173 (S. Karger AG, 1987).
Sheehan, D. V. The Mini-International Neuropsychiatric Interview (M.I.N.I.): The Development and Validation of a Structured Diagnostic Psychiatric Interview for DSM-IV and ICD-10. J Clin Psychiatry 12 (1998).
Ferrando, L., Bobes, J., Gibert, J., Soto, M. & Soto, O. MINI Entrevista Neuropsiquiátrica Internacional. (2000).
Arditte Hall, H. A., Quinn, M. E., Vanderlind, W. M. & Joormann, J. Comparing cognitive styles in social anxiety and major depressive disorders: An examination of rumination, worry, and reappraisal. Br. J. Clin. Psychol. 58, 231–244 (2019).
Kashdan, T. B. The neglected relationship between social interaction anxiety and hedonic deficits: Differentiation from depressive symptoms. J. Anxiety Disord. 18, 719–730 (2004).
Jasper, H. The ten-twenty electrode system of the International Federation. Clin Neurophysiol 10, 370–375 (1958).
Oostenveld, R., Fries, P., Maris, E. & Schoffelen, J.-M. FieldTrip: Open Source Software for Advanced Analysis of MEG, EEG, and Invasive Electrophysiological Data. Comput. Intell. Neurosci. 2011, 1–9 (2011).
Hu, X., Xu, Z. & Mai, X. Social value orientation modulates the processing of outcome evaluation involving others. Soc. Cogn. Affect. Neurosci. 12, 1730–1739 (2017).
Volpert-Esmond, H. I., Merkle, E. C., Levsen, M. P., Ito, T. A. & Bartholow, B. D. Using trial-level data and multilevel modeling to investigate within-task change in event-related potentials. Psychophysiology 55, e13044 (2018).
Luo, Y., Eickhoff, S. B., Hétu, S. & Feng, C. Social comparison in the brain: A coordinate-based meta-analysis of functional brain imaging studies on the downward and upward comparisons: Neural Signatures of Downward and Upward Comparisons. Hum. Brain Mapp. 39, 440–458 (2018).
Antony, M. M., Rowa, K., Liss, A., Swallow, S. R. & Swinson, R. P. Social comparison processes in social phobia. Behav. Ther. 36, 65–75 (2005).
Gehring, W. J. & Willoughby, A. R. The Medial Frontal Cortex and the Rapid Processing of Monetary Gains and Losses. Science 295, 2279–2282 (2002).
Hajcak, G., Moser, J. S., Holroyd, C. B. & Simons, R. F. The feedback-related negativity reflects the binary evaluation of good versus bad outcomes. Biol. Psychol. 71, 148–154 (2006).
Sambrook, T. D. & Goslin, J. A Neural reward prediction error revealed by a meta-analysis of ERPs using great grand averages. Psychol. Bull. 141, 213–235 (2015).
Fukushima, H. & Hiraki, K. Whose loss is it? Human electrophysiological correlates of non-self reward processing. Soc. Neurosci. 4, 261–275 (2009).
Thoma, P. & Bellebaum, C. Your Error’s Got me Feeling – How Empathy Relates to the Electrophysiological Correlates of Performance Monitoring. Front. Hum. Neurosci. 6, (2012).
Boksem, M. A. S., Kostermans, E., Milivojevic, B. & De Cremer, D. Social status determines how we monitor and evaluate our performance. Soc. Cogn. Affect. Neurosci. 7, 304–313 (2012).
Gu, R., Ge, Y., Jiang, Y. & Luo, Y. Anxiety and outcome evaluation: The good, the bad and the ambiguous. Biol. Psychol. 85, 200–206 (2010).
Cao, J., Gu, R., Bi, X., Zhu, X. & Wu, H. Unexpected Acceptance? Patients with Social Anxiety Disorder Manifest their Social Expectancy in ERPs During Social Feedback Processing. Front. Psychol. 6, (2015).
LeMoult, J. & Gotlib, I. H. Depression: A cognitive perspective. Clin. Psychol. Rev. 69, 51–66 (2019).
Wells, A. et al. Social phobia: The role of in-situation safety behaviors in maintaining anxiety and negative beliefs. Behav. Ther. 26, 153–161 (1995).

Table 1. Sample characteristics

	Control	MD-SA	p-value
n	72	63
Sex
Female	67	57	0.58 ns
Male	5	6	0.58 ns
Age	22.5 ± 3.8	23.5 ± 4.4	0.17 ns
Years of education
Undertaken	15.8 ± 2.8	15.9 ± 2.9	0.89 ns
Complete	14.2 ± 2.6	13.5 ± 2.2	0.07 ns
Nicotine consumption
No	62	53	0.75 ns
Yes	10	10	0.75 ns
Skilful hand
Left-handed	12	7	0.35 ns
Right-handed	60	56	0.35 ns
Discipline of study
Health science	45	39	0.61 ns
Social science and arts	12	14
Technology and science	15	10
BDI-II	1.4 ± 1.9	22.7 ± 10.9	<.001
SHAPS	0.1 ± 0.4	2.7 ± 3.0	<.001
SAD	2.4 ± 2.6	21.8 ± 5.7	<.001
FNE	7.1 ± 5.4	23.9 ± 5.7	<.001
LSAS
Total	13.1 ± 9.8	74.9 ± 23.6	<.001
Fear/anxiety	7.1 ± 6.1	37.4 ± 12.7	<.001
Avoidance	6.0 ± 4.8	37.6 ± 12.4	<.001
STAI
Trait	35.3 ± 5.3	59.3 ± 7.4	<.001
State	28.1 ± 4.5	46.0 ± 10.7	<.001
INCOM	30.1 ± 7.6	37.8 ± 9.2	<.001
PANAS
Positive affect	34.7 ± 5.6	19.8 ± 4.4	<.001
Negative affect	16.3 ± 2.7	26.5 ± 6.1	<.001
RSES	25.6 ± 3.7	12.4 ± 5.6	<.001
ACIPS	90.2 ± 7.5	70.5 ± 14.3	<.001
IIP
Total	51.1 ± 21.3	107.1 ± 24.8	<.001
Domineering/controlling	6.2 ± 3.8	8.4 ± 4.1	<.001
Vindictive/self-centered	4.4 ± 3.0	7.5 ± 3.9	<.001
Cold/distant	3.7 ± 3.0	12.0 ± 5.1	<.001
Intrusive/needy	5.9 ± 4.0	8.6 ± 5.3	<.001
Socially inhibited	4.6 ± 4.0	18.4 ± 7.2	<.001
Nonassertive	6.6 ± 4.9	19.1 ± 6.0	<.001
Overly accommodating	8.7 ± 5.1	16.2 ± 6.0	<.001
Self-sacrificing	11.0 ± 4.6	16.8 ± 4.9	<.001

Values are given as mean ± standard deviation. p values are based on the independent-samples t-test. ns: no significant difference between groups. BDI-II: Beck Depression Inventory-II (scores from the experimental session); SHAPS: Snaith-Hamilton Pleasure Scale; SAD: Social Avoidance and Distress; FNE: Fear of Negative Evaluation; LSAS: Liebowitz Social Anxiety Scale (scores from the experimental session); STAI: State-Trait Anxiety Inventory; INCOM: Iowa-Netherlands Comparison Orientation Scale; PANAS: Positive And Negative Affect Schedule; RSES: Rosenberg Self-Esteem Scale; ACIPS: Anticipatory and Consummatory Interpersonal Pleasure Scale; IIP: Inventory of Interpersonal Problems.

Table 2. Results of ERPs

	Control	MD-SA
MFN
“You correct-Other correct”	11.1 ± 13.0	10.2 ± 12.8
“You correct-Other wrong”	8.0 ± 12.5	8.1 ± 12.6
“You wrong-Other wrong”	7.2 ± 12.1	7.0 ± 12.9
“You wrong-Other correct”	6.4 ± 12.5	7.1 ± 12.5
P300
“You correct-Other correct”	23.9 ± 13.2	21.4 ± 13.9
“You correct-Other wrong”	22.6 ± 13.7	19.9 ± 13.9
“You wrong-Other wrong”	21.7 ± 13.3	20.0 ± 15.1
“You wrong-Other correct”	19.3 ± 14.3	18.0 ± 14.7
LPP
“You correct-Other correct”	13.8 ± 12.8	13.7 ± 14.1
“You correct-Other wrong”	16.7 ± 12.9	15.9 ± 14.1
“You wrong-Other wrong”	15.0 ± 13.2	14.4 ± 15.3
“You wrong-Other correct”	16.3 ± 14.6	16.2 ± 15.3

Values are given as mean ± standard deviation.

No competing interests reported.

SUPPLEMENTARYMATERIALEEG01.odt

Download PDF

Version 1

posted

You are reading this latest preprint version

Event-related potentials of social comparisons in depression and social anxiety

Status:

Version 1

Abstract

Figures

Introduction

Methods

Results

Discussion

Conclusion

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1