Elsevier

NeuroImage

Volume 102, Part 2, 15 November 2014, Pages 370-380
NeuroImage

Projecting my envy onto you: Neurocognitive mechanisms of an offline emotional egocentricity bias

https://doi.org/10.1016/j.neuroimage.2014.08.007Get rights and content

Highlights

  • Novel evidence for projection of own extent of social emotions onto others.

  • Novel evidence that experience and projection of social emotions rely on shared networks.

  • Novel evidence of individual differences in recruiting shared networks explains self-projection.

Abstract

Humans often project their own beliefs, desires and emotions onto others, indicating an inherent egocentrism. In five studies we investigated the neurocognitive mechanisms underlying emotional egocentricity bias (EEB) and specifically an offline EEB, defined as the projection of one's own tendency to react with a certain emotional response pattern in a given situation onto other people. We used a competitive reaction time game associated with monetary gains and losses that allowed inducing feelings of envy and Schadenfreude. While we found evidence for the first hand experience of envy and Schadenfreude, we also observed an offline bias, that is participants on average projected feelings of envy and Schadenfreude when having to judge others. Importantly the extent of experienced and projected social emotions were highly correlated. This bias was observed when participants were both directly involved and also as an uninvolved party, suggesting the offline bias to be independent of the presently experienced emotion. Under increased time pressure however an online bias emerged whereby participants just projected their presently experienced emotions onto the other. Finally, we show that on the neural level shared neuronal networks underlie the offline EEB at least for envy. Thus, for envy, activity of the same part of anterior insula was sensitive to individual differences both in the experience and the projection of envy. These findings outline the set of circumstances leading to specific types of empathic attribution biases and show that individual differences in the experience of social emotions are predictive of the offline egocentricity bias both on a behavioral as well as a neural level. These data extend present models on the neurocognitive mechanisms of interpersonal understanding in the socio-affective domain.

Introduction

How humans come to understand others poses a perennial quandary to scientists studying this ability. Recent advances in unraveling how one can make predictions of others' mental and emotional states have shown that typically, humans rely on their own experiences to make inferences on what others might think or feel (Gallese and Goldman, 1998, Keysers and Gazzola, 2007, Lamm et al., 2011, Mitchell, 2009, Singer et al., 2004). The set of one's own experiences used to infer the states of others can range from currently experienced physical states (Van Boven and Loewenstein, 2003) to personality traits (Krueger and Clement, 1994, Krueger and Zeiger, 1993) and attitudes (Biernat et al., 1997). Thus, humans are remarkably prone to assuming that others are like them (Gilovich et al., 1983). This assumption is useful when the experience of self and other are matched, but more often than not, it will lead to a bias in judging others egocentrically when they differ. Affective egocentricity bears hidden costs in that it can lead to interpersonal (Thompson and Loewenstein, 1992) as well as group conflict (Chambers et al., 2006). Understanding and isolating the underlying causes of such a socially detrimental proclivity is thus critical in helping to reduce potentially resulting conflict and antisocial behavior.

There is a large body of evidence in the domain of behavioral psychology on the occurrence of egocentric biases in the context of cognitive social judgments (for reviews see Karniol, 2003, Robbins and Krueger, 2005), particularly when targets are assumed to be similar to oneself (Ames, 2004). However, while covering mostly personality traits and attitudes, emotional states as a source of egocentric judgments have been largely ignored (but see O'Brien and Ellsworth, 2012 for studies on the projection of visceral states onto others; Van Boven and Loewenstein, 2003). Further, the neural mechanisms of why such egocentric biases occur are almost missing entirely. So far previous studies have rather focussed on uncovering brain regions associated with overcoming egocentricity rather than what gives rise to this egocentricity in the first place. Thus, recently it was shown that adjusting from the tendency to project one's own preferences onto others recruits brain regions known to be involved in mentalizing, such as medial prefrontal cortex (Tamir and Mitchell, 2010). Further, a recent neuroimaging study on emotional egocentricity bias (EEB) showed using pleasant and unpleasant tactile stimulation that adults' judgments of others' emotional states were strongly influenced by the affective stimulation they were simultaneously experiencing (Silani et al., 2013) and that the recruitment of right supramarginal gyrus (rSMG) was critical in overcoming this. More recently this EEB was shown to be increased in children compared to adults in the context of a paradigm inducing positive and negative emotions by means of monetary rewards and punishments (The Egocentricity Monetary Reward and Punishment Paradigm, in short EMOP; Steinbeis et al., in press). Children judged others to feel more positive when they themselves had just won and judged others to feel negative when they themselves had just lost, which was coupled with reduced recruitment of rSMG. Thus, in two studies, rSMG was shown to be a crucially related to overcoming an EEB that arises out of one's current emotional state (Silani et al., 2013, Steinbeis et al., in press). However, what remains unclear is how such emotional egocentricity occurs in the first place and what the neural mechanisms are that give rise to this.

To study this we used a paradigm capable of inducing higher level social emotions arising out of social comparisons with others, such as envy and Schadenfreude. Thus, there is by now an extensive body of literature on the occurrence of social comparison in adults and the influence that the welfare of others has on one's own well-being (Festinger, 1954, Gibbons and Buunk, 1999). Research has shown that processes of social comparison can lead to feelings of envy and Schadenfreude. Envy can be defined as feeling bad about someone's superiority particularly in something that is personally highly relevant. Schadenfreude in turn entails pleasure at another's suffering or doing worse than oneself (Smith and Kim, 2007, Smith et al., 1996). We were interested in whether those participants who experience more envy or Schadenfreude themselves in social comparison situations also attribute more envy and Schadenfreude onto others when engaging in empathic judgments and seeing others in situations likely to elicit such emotions. Note, that even though we use a paradigm that elicits social emotions to test this, such affective projections are not limited to social emotions only. Thus, it is possible to conceive of the tendency to project basic emotional states such as anger or sadness onto others to the same extent as one experiences these states in comparable situations. Therefore, even though we are using a specific test case of social emotions to test for this hypothesis, we assume that such mechanisms apply to any emotional state that can be projected.

This novel paradigm uses a speeded reaction time game with monetary rewards and punishments to induce positive and negative emotions in participants and an agent who is playing the same game with the participants (the EMOP). Participants simultaneously play a competitive game where in each round both can either win or lose depending on their performance. Full feedback of both player's wins and losses is displayed to both participants and depending on the condition, of interest is either the effect of the other's wins and losses on judging one's own emotional experience or the effect of one's own wins and losses on judging the other's emotional experience. The EMOP has been shown to induce positive and negative feelings in both players and crucially to infer the presence of social emotions such as envy and Schadenfreude in participants (Steinbeis and Singer, 2013, Steinbeis et al., in press). Thus, in the context of this task envy is defined as feeling worse when losing while the competitor wins compared to both participants losing; this is measured as the difference in rating between the condition where participants lose and the competitor wins compared to the condition when both lose. Schadenfreude on the other hand is defined as feeling better when winning while the competitor lost compared to both participants winning; this is measured as the difference in rating between the condition where participants win and the competitor loses compared to the condition when both win.

Importantly, this paradigm was designed to give the possibility to assess the existence of two types of EEB (see Fig. 1). The first type of EEB is indicated by participants having just won or lost money during the game and simply projecting their present pleasant or unpleasant feelings onto the other when having to judge how another feels that just won or lost money during the very same game. Because this type of EEB results from the direct projection of one's current online affective experience we refer to it as an online EEB. Thus in the context of our task a negative bias is said to occur when others are rated as feeling less positive after winning when participants themselves had lost compared to won. A positive bias in turn is said to occur when others are rated as feeling less negative after losing when participants themselves had won compared to lost (see Fig. 1). In contrast, another type of EEB is indicated by participants simulating the extent of their own tendency to feel an emotion when making inferences about another's affective reactions in a comparable situation (Fig. 1). Thus, unlike for the online EEB, for this EEB to arise it is not necessary to currently experience the emotion, but rather to simulate one's own emotional state offline if one were in the situation of the target to be judged. Because this EEB is assumed to arise out of a process of a context-dependent internal and offline simulation and not a simple projection of the actual experienced affective state onto another, we will refer to this as an offline EEB. Thus in the context of our task, the attribution of envy occurs when others (i.e. the competitor) are rated as feeling more negative after participants had won and the competitor lost compared to when both had lost (i.e. attributed envy). Equally, the attribution of Schadenfreude occurs when others are rated as feeling more positive after participants had lost and the competitor won compared to when both had won (attributed Schadenfreude). Crucially, this EEB requires also a correlation between individual differences in experiencing envy and Schadenfreude and individual differences in attributing such tendencies to others (Fig. 1). Importantly, this EEB refers to the projection of the offline simulation of an emotional state and not necessarily to the projection of one's trait (i.e. trait envy or Schadenfreude). While these two may very well be linked, the present investigation focusses solely on projecting simulated emotional states and not traits. The correlation between experienced and attributed emotion is important also in as far as one may assume that participants may have lay theories about others' social emotions in competitive contexts, which may not be linked to the extent that they experience these emotions themselves. Thus, it is only through a correlation between experienced and attributed emotion that an offline bias is fully indicated. The difference between Fig. 1 also illustrates what online and offline EEB should show on average means of opposite directions. This distinction is important for it highlights the myriad ways in which socio-affective judgments can be subject to biases. Thus, for instance online biases may be successfully overcome in that one's present affective experience is discounted, but this could still lead to an offline bias in that one projects one's own general tendency to feel an emotion. Whereas it has been shown that an online EEB can be elicited by means of this paradigm (Steinbeis et al., in press), we were interested if on top of an online EEB it is also possible to elicit an EEB that results from the offline simulation of one's own emotional state if one were in the situation of the target to be judged. Thus, the first major goal of this study was to give first-time evidence for the existence for such an offline EEB (Studies 1a and 1b).

A further goal was to assess the role of contextual factors such as timing and direct emotional involvement influencing when online and offline EEBs are likely to emerge. Previous studies have shown that prolonged periods of time may help in reaching more accurate empathic judgments (van der Heiden et al., 2013) and that egocentric biases increase under time pressure (Epley et al., 2004, Silani et al., 2013), suggesting time to be a crucial factor in the occurrence of egocentric biases. Given that an offline EEB requires first to overcome the tendency to make an online EEB and that to overcome the online bias one's present emotional state has to be discounted, which presumably requires effort and thus a certain amount of time (Silani et al., 2013) we hypothesized that an online bias is more likely to occur than an offline bias when having to make empathic judgments under increased time pressure (Study 2). Further, the online bias requires direct emotional involvement because it is the currently experienced affect that has to be overcome to reach an accurate empathic judgment. For the offline bias we hypothesize that such first-hand experience of a social emotion is not required given that it is assumed to result from active processes of internally simulating what another person is likely to feel in a similar situation based on the emotional tendency one usually displays in a comparable social comparison situation. Thus, the offline bias should also occur even without any direct emotional involvement of the person who is doing empathic judgments, that is when one is not currently experiencing a particular affective state. We therefore hypothesize that the offline bias should also occur in conditions when one merely has to rate the affective experience of another without being directly emotionally implicated in the task itself but being just a neutral passive observer of two other people engaging in this task (Study 3).

One third major goal of this study was to uncover the neural mechanisms underlying the occurrence of the offline EEB (Study 4). In the affective domain, a shared representation account has been proposed for the experience and the understanding of others' primary emotions across several domains (e.g. affective touch, pain and reward; Bernhardt and Singer, 2012, Jabbi et al., 2008, Keysers et al., 2004, Lamm et al., 2011, Mobbs et al., 2009, Singer, 2012, Singer et al., 2004, Wicker et al., 2003). According to this “shared network” hypothesis, the same brain regions are recruited when having a first-hand experience of an emotion or just vicariously observing another person experiencing these emotions. Despite abundant evidence for shared networks in social neuroscience underlying the first-hand and vicarious affective experience of primary emotions and sensations such as touch, disgust, taste and pain, so far these studies have not tested, whether individual differences in the degree to which these emotions are experienced first-hand are predictive of individual differences in the degree to which one attributes such experiences and emotions to others. Shared networks hypotheses however should predict such a correlation based on individual differences in affective experience, whereby the same brain regions underlying the experience of social emotions such as envy and Schadenfreude should also be recruited when attributing these states to others.

Imaging studies suggest that feelings of envy are linked with activation in the dorsal anterior cingulate cortex (Takahashi et al., 2009) and feelings of Schadenfreude with activation in the ventral striatum (Bault et al., 2011, Dvash et al., 2010, Singer et al., 2006, Takahashi et al., 2009), as well as the medial prefrontal cortex (Bault et al., 2011). In addition and as a result of the heightened emotional engagement of our task we expected to see involvement of brain regions involved in processing highly arousing negative emotions, such as the anterior insula, for the experience of envy, a region argued to play a crucial role in aversive social emotions (for a review see Lamm and Singer, 2010). Thus, we would expect differential shared networks underlying the first-hand and vicarious experience of positively valenced social emotions of Schadenfreude and negatively valence emotions of envy. Crucially, however, if shared neural representations of social emotions are related to the offline EEB we would expect the degree of neural activity observed during the first-hand experience of social emotions to be predictive of the extent of this activity observed in the same areas during the attribution of these very social emotions onto the other playing agent. Importantly, because rSMG has been previously been reported to play a crucial role in overcoming the EEB and given that we are currently interested in what gives rise to an EEB in the first place as opposed to the neural mechanisms that help to overcome it, we did not expect any involvement of rSMG in the context of our task. Further, given the affective nature of our task, we specifically predicted the involvement of brain regions involved in processing the affect of oneself as well as others, such as the anterior insula, as opposed to brain regions typically implicated in cognitive perspective taking, such as cortical midline structures (Amodio and Frith, 2006).

To test our hypotheses, we conducted five studies, four behavioral and one imaging study. In the first two studies (Studies 1a and 1b), we established the effectiveness of our EMOP in eliciting social emotions of envy and Schadenfreude in adults as well as an offline EEB. In a third study (Study 2), we show that when increasing the time pressure, we elicit an online EEB, while in a fourth study (Study 3), we show that the offline EEB also occurs when participants were merely passively watching and thus not emotionally engaged in the task at all. A final study (Study 4) using functional magnetic resonance imaging (fMRI) was performed to gain an understanding of the neural mechanisms underlying the offline EEB.

Section snippets

Materials and methods

All studies were approved by the ethics committees of Zuerich University and of the Canton of Zurich (E68/2008) as well as of the University of Leipzig (E029-11-24012011) and all participants gave informed consent to participate in the studies. With the exception of Study 1a, in which subjects were Swiss, all subjects in the remaining studies were German.

Studies 1a and 1b (establishing and replicating the EEB)

Given the a priori hypotheses we had on the separable effects of envy and Schadenfreude and their respective associations with being attributed to others, we computed separate models for experienced emotions and another for attributed emotions by means of two repeated measures ANOVA with one factor of two levels each (Experienced envySelf Judgment: Self Loss/Other Win  Self Loss/Other Loss; Experienced SchadenfreudeSelf Judgment: Self Win/Other Loss  Self Win/Other Win; Attributed envyOther

Discussion

The present set of behavioural and neuroimaging studies sought to examine the conditions for the occurrence of online and more importantly an offline EEB, as well as the neuro-cognitive mechanisms underlying offline bias. More specifically, we employed a speeded reaction time task, the EMOP, associated with winning and losing money while playing with another agent. This task is capable of eliciting strong social emotions arising out of social comparisons, namely envy and Schadenfreude.

Conclusion

In conclusion the present set of studies introduces the existence of an offline emotional egocentricity bias in social judgments by showing that healthy human adults indeed tend to attribute their own tendencies to display social emotions such as envy and Schadenfreude in social comparison situation to others. This offline EEB however is dependent of time available for such judgments. Thus, we show that a slight temporal modification of our paradigm is capable of transforming an offline EEB

References (50)

  • L. Thompson et al.

    Egocentric Interpretations of Fairness and Interpersonal Conflict

    Organ. Behav. Hum. Decis. Process.

    (1992)
  • L. van der Heiden et al.

    Inter-individual differences in successful perspective taking during pain perception mediates emotional responsiveness in self and others: an fMRI study

    NeuroImage

    (2013)
  • B. Wicker et al.

    Both of us disgusted in My insula: the common neural basis of seeing and feeling disgust

    Neuron

    (2003)
  • J. Zaki et al.

    Overlapping activity in anterior insula during interoception and emotional experience

    NeuroImage

    (2012)
  • D.R. Ames

    Strategies for social inference: A similarity contingency model of projection and stereotyping in attribute prevalence estimates

    J. Pers. Soc. Psychol.

    (2004)
  • D.M. Amodio et al.

    Meeting of minds: the medial frontal cortex and social cognition

    Nat. Rev. Neurosci.

    (2006)
  • N. Bault et al.

    Medial prefrontal cortex and striatum mediate the influence of social comparison on the decision process

    Proc. Natl. Acad. Sci. U. S. A.

    (2011)
  • B.C. Bernhardt et al.

    The Neural Basis of Empathy

    Annu. Rev. Neurosci.

    (2012)
  • M. Biernat et al.

    Simultaneous assimilation and contrast effects in judgments of self and others

    J. Pers. Soc. Psychol.

    (1997)
  • J.R. Chambers et al.

    Misperceptions in intergroup conflict - Disagreeing about what we disagree about

    Psychol. Sci.

    (2006)
  • J. Dvash et al.

    The envious brain: the neural basis of social comparison

    Hum. Brain Mapp.

    (2010)
  • N. Epley et al.

    Perspective taking as egocentric anchoring and adjustment

    J. Pers. Soc. Psychol.

    (2004)
  • A. Etkin et al.

    Functional neuroimaging of anxiety: A meta-analysis of emotional processing in PTSD, social anxiety disorder, and specific phobia

    Am. J. Psychiatr.

    (2007)
  • L. Festinger

    A Theory of Social Comparison Processes

    Hum. Relat.

    (1954)
  • K.J. Friston et al.

    Statistical parametric maps in functional imaging: A general linear approach

    Hum. Brain Mapp.

    (1994)
  • Cited by (28)

    • Happy for Us not Them: Differences in neural activation in a vicarious reward task between family and strangers during adolescent development

      2021, Developmental Cognitive Neuroscience
      Citation Excerpt :

      Primarily, future research should include a broader spectrum of social emotions. To elucidate the relationship between participants and targets as well as their emotional valence we suggest including conditions of compassion, envy, and schadenfreude (Steinbeis and Singer, 2014). To realize a more detailed distinction between social targets, the inclusion of peers or friends, especially during adolescent development, might prove to be of value in future studies.

    • When differences matter: rTMS/fMRI reveals how differences in dispositional empathy translate to distinct neural underpinnings of self-other distinction in empathy

      2020, Cortex
      Citation Excerpt :

      This reduction in rated intensity of the other person's emotion reflects the extent our empathic judgements are biased (“drawn towards”) one's own emotion (see Fig. 1). The presence of this egocentric bias in empathy has been corroborated by additional studies, including ones that had induced emotions in other ways than via visuo-tactile stimulation (Hoffmann, Singer, & Steinbeis, 2015; Steinbeis et al., 2015; Steinbeis & Singer, 2014; Tomova et al., 2014; von Mohr et al., 2019). In terms of the neural correlates, the study of Silani et al. (2013) also identified across two fMRI experiments the right supramarginal gyrus (rSMG) to be more active when empathically judging emotions that are incongruent with the participant's self-experienced emotion.

    • The neuroscience of social comparison and competition

      2023, Cognitive, Affective and Behavioral Neuroscience
    View all citing articles on Scopus
    View full text