Too Good to be Liked? When and How Prosocial Others are Disliked

Too Good to be Liked? When and How Prosocial Others are Disliked. Lucia L.-A. Boileau, David J. Grüning and Herbert Bless. Front. Psychol., August 19 2021. https://doi.org/10.3389/fpsyg.2021.701689

Abstract: Outstandingly prosocial individuals may not always be valued and admired, but sometimes depreciated and rejected. While prior research has mainly focused on devaluation of highly competent or successful individuals, comparable research in the domain of prosociality is scarce. The present research suggests two mechanisms why devaluation of extreme prosocial individuals may occur: they may (a) constitute very high comparison standards for observers, and may (b) be perceived as communal narcissists. Two experiments test these assumptions. We confronted participants with an extreme prosocial or an ordinary control target and manipulated comparative aspects of the situation (salient vs. non-salient comparison, Experiment 1), and narcissistic aspects of the target (showing off vs. being modest, Experiment 2). Consistent with our assumptions, the extreme prosocial target was liked less than the control target, and even more so when the comparison situation was salient (Experiment 1), and when the target showed off with her good deeds (Experiment 2). Implications that prosociality does not always breed more liking are discussed.

General Discussion

The present research demonstrates that individuals who perform an outstanding degree of prosocial behaviors may be devaluated—due to their prosocial behaviors. Specifically, across two experiments, the prosocial target was liked less than the control target. This consistent pattern is unlikely to be due to participants' perception that the displayed behaviors did not unambiguously reflect prosocial behavior: When explicitly evaluating prosociality, the prosocial target was clearly perceived as prosocial (and more so than the control target). The finding that prosocial behaviors may decrease rather than increase liking seems rather surprising at first glance. Past research suggests that liking and perceptions of prosociality in others are in fact very highly correlated (Imhoff and Koch, 2017). However, the observed devaluation is in line with prior empirical research suggesting that superior prosocial others are indeed sometimes devaluated through rejection and dislike (Fisher et al., 1982; Herrmann et al., 2008; Parks and Stone, 2010; Pleasant and Barclay, 2018).

The present research goes beyond prior research that has similarly demonstrated a possible disliking of prosocial targets by suggesting and investigating two possible underlying processes. Thus, it responds to the call that mediating mechanisms for the dislike of very prosocial targets are yet to be investigated (Parks et al., 2013).

First, the reduced liking of the prosocial target was more pronounced when comparisons between the target and the observers were induced by the information that observers would first evaluate the target and then themselves on the very same items. Eliciting such a comparison expectation increased disliking of the prosocial target. Presumably, in this situation, the extremely prosocial target constituted a very high comparison standard, and this high standard would have negative consequences for participants' evaluations of themselves (Mussweiler, 2003; Bless and Schwarz, 2010; Morina, 2021). This conclusion extends indirect evidence by Parks and Stone (2010) by providing an experimental manipulation of the assumed comparison component.

Second, as predicted, the dislike of the prosocial target was increased when perceptions of communal narcissism (Gebauer et al., 2012; Nehrlich et al., 2019) were elicited by informing participants that the target actively sought to let others know about her prosocial behaviors. This finding suggests that a target's prosocial behavior will not turn into more liking but backfire when that target is perceived as someone who exerts “excessive self-enhancement” in the domain of prosociality and who is showing off with her good deeds (Rentzsch and Gebauer, 2019; p. 1373).

Interestingly, the two proposed accounts (comparison processes, and communal narcissism) may in fact be related. The perception that an individual competes for higher status on the communal dimension (Gebauer et al., 2012) might elicit social comparison processes just as a situation in which comparison is made salient. Communal narcissists do show-off because they want to outperform others—which inevitably entails a comparison. In case the comparison is not only given on the side of the prosocial target but also on the side of the perceiver, one might speculate that we manipulated comparative aspects of the situation in Experiment 1 and comparative aspects of the target in Experiment 2. Such speculation may explain why assessed communal narcissism in Experiment 1 was strongly related to disliking the prosocial target. In turn, eliciting social comparison via the situation might increase perceptions of communal narcissism as a form of target devaluation (cf. Monin, 2007). Note, however, that we did not find a significant interaction effect of target type and comparison salience on perceptions of communal narcissism in Experiment 1.

In combination, the present research provides first evidence on two potential mechanisms to explain the devaluation of very prosocial others. The findings thus provide a first, yet important step, for investigating the processes that lead to the devaluation of prosocial targets.

Open Issues and Caveats

Although the present research offers new insights into the potential devaluation of prosocial others, it is important to address several open issues and caveats. First, one may speculate about the degree of prosocial behavior that is necessary to elicit a dislike of the target (in combination with the comparison component). Of course, prosocial behavior does not necessarily lead to disliking. In this respect, it is important to point out that our target was very outstanding with respect to prosocial behavior. Thus, to elicit devaluation, the target might have to be perceived as “too good” or “too perfect” (note that self-devaluation tends to be elicited when the outstanding other is perceived as unreachable; Lockwood and Kunda, 1997). Evidence in this respect has been reported for the ability domain demonstrating that a superior target was liked more than an average person—if the superior target displayed some imperfections in other domains (Aronson et al., 1966). Further research is needed to address this issue systematically in the prosocial domain.

Second, our conclusion on the crucial role of comparison processes for the dislike of outstanding prosocial targets matches with research on the devaluation of targets that outperform others in the domain of performance and abilities (Pleban and Tesser, 1981; Salovey and Rodin, 1984; Alicke et al., 1997; Alicke, 2000; Lassiter and Munhall, 2001). The present findings thus suggest that similar processes may cause devaluation of both oustandingly capable, and outstandingly prosocial individuals (for a discussion of different, yet overlapping conceptualizations of ability vs. prosociality, see, e.g., Fiske et al., 2007; Gebauer et al., 2014; Abele et al., 2016). Similarly, these processes may also cause devaluation of outstandingly moral individuals (e.g., Monin et al., 2008; Minson and Monin, 2012). Importantly, despite some overlaps, prosociality and morality are not the same. Morality comprises being loyal, fair, law-abiding, and pure (Graham et al., 2013)—aspects that are at least partly independent of prosociality. However, given the similar patterns of dislike observed for superior targets in these various domains, it seems worthwhile to investigate communalities and differences between prosociality and morality in their underlying mechanisms.

Third, we did not find complementary patterns for perceived communal narcissism vs. perceived modesty in Experiment 2. The obtained findings do not allow for an answer to this issue. With respect to the concept of narcissism it might be interesting to investigate whether narcissism and modesty are located on different sides of the same dimension or whether the two concepts are at least partly unrelated to each other (for a discussion of the humility and grandiose narcissism dimension, see Miller et al., 2012; Gebauer and Sedikides, 2019). This relation might also depend on whether narcissism and modesty are measured via self-reports or other-reports. Interestingly, while self-rated communal narcissism has been conceptualized and investigated (Gebauer et al., 2012; Nehrlich et al., 2019), the perception of communal narcissism in others has so far received little systematic investigation (Rentzsch and Gebauer, 2019).

Fourth, the crucial interaction of target type and communal narcissism (i.e., show-off vs. modest condition) in Experiment 2 did not reach the conventional level of significance when tested two-sided. We readily acknowledge this aspect. Note, however, that we pre-registered our study so that one-sided tests statistics could potentially be applied. Due to the unexpected drop out (due to the attention check, see above) the conducted analyses were presumably underpowered, which constitutes a common problem in the field of psychological research (Maxwell, 2004).

Fifth, the reduced liking of the outstandingly prosocial target reflects a contrast effect. General models on context effects in social judgment (cf. Bless and Schwarz, 2010) hold that—under specified conditions—contrast effects may turn in assimilation effects. We readily subscribe to this possibility. One condition that might apply to the present research could rest in the perceived similarity between target and perceiver. In case perceivers assume a high overlap between themselves and the target, they may derive positive implications for themselves rather than devaluating the target (e.g., basking in the reflected glory, Cialdini and DeNicholas, 1989; see also Brown et al., 1992; for an overview on assimilation vs. contrast effects, see Bless and Schwarz, 2010).

Sixth, our sample predominantly consisted of females. As our target was female as well, this might have influenced our results (e.g., see Espinosa and Kovárík, 2015), for gender differences in prosocial behavior). To address this issue, further research needs to test potential gender differences in the evaluation of outstandingly prosocial others.

Finally, it needs to be pointed out that we minimized the potential ambiguity of the prosocial behavior. One could speculate that other devaluation mechanisms (i.e., denying prosociality, ascribing lower competence, etc.) might be at work when the prosocial behavior is more ambiguous and more open to interpretations.

Outlook

While readily acknowledging the open issues addressed above, we strongly believe that the present research addresses important issues. We consistently demonstrated that performing prosocial behaviors may lead to reduced liking. Moreover, we showed that the reduced liking of prosocial individuals is triggered by both comparison processes and perceptions of communal narcissism. The present set of studies therefore show, to our knowledge, the first direct evidence on underlying mechanisms in the devaluation of very prosocial others and provide a basis for future research.

Leaving the laboratory situation, the obtained findings suggest that performing prosocial behaviors is no guarantee to be liked. In fact, in some instances, individuals may be disliked because of their prosocial behaviors. This observation is in line with the ambiguously connoted term “do-gooder,” which on the one hand describes the target's “desire and effort to help people” but on the other hand, points out to potential evaluations of the target's behavior as “wrong,” or “annoying” (see Merriam-Webster, n.d.). We assume that the ambivalence of the term “do-gooder” is rather widespread. If so, research needs to pay more attention to the devaluation of prosocial others, as it might constitute a potential obstacle to individuals' motivation for prosocial behavior.