https://orcid.org/0000-0003-0543-277X
![Sample our Behavioral Sciences journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/110801/TOC-Subject-Sample-2021-Behavioral-Sciences.jpg"})
ABSTRACT
Victims often have to decide whether or not they want to forgive the offender after a transgression. Forgiveness has in general been shown to have positive consequences (e.g. for the victim–transgressor relationship), but recent theorizing suggests that these may be conditional on the extent to which a forgiveness response is attributed to benevolent (vs. malevolent) motives. Here, we investigate how the presence of an attentive audience influences offenders’ motive attributions and post-transgression behavior. Specifically, we hypothesize that offenders attribute a forgiveness reaction to malevolent rather than benevolent motives if it occurs in front of an audience vs. in private, and that this leads to withdrawal from the victim rather than reconciliation. While two preliminary vignette studies (N = 396) provided initial support for these assumptions, a preregistered virtual reality experiment (N = 156) yielded more mixed results. Specifically, in line with our predictions, receiving a forgiveness response from the victim in front of an attentive audience made participants keep a greater distance to the victim than receiving a forgiveness response in private. This effect was, however, found in only one out of three approach–avoidance tasks. We discuss possible explanations for these results and future research avenues.
Disclosure statement
There is no actual or potential conflict of interest including any financial, personal, or other relationships with other people or organizations whatsoever.
Correction Statement
This article has been republished with minor changes. These changes do not impact the academic content of the article.
Notes
1. In more severe transgression situations, such as maltreatment or criminal offenses, attributional dynamics may depend more on the forgiver’s actual behavior or the victim–offender relationship rather than minor contextual cues.
2. This may be even exacerbated if the audience has not witnessed the initial transgression but learns about it only through the forgiveness response. However, our theorizing also applies to situations in which the audience already knows about the transgression.
5. In fact, the motive-attribution framework proposes similar psychological phenomena for forgiveness and punishment. Consequently, Pre-Study 1 was initially designed to examine audience effects on both responses to an offense. However, after conducting Pre-Study 1, we realized that empirical studies investigating offenders’ attribution of forgiveness require a different methodological approach than studies investigating offenders’ attribution of punishment (e.g. regarding the severity of the transgression, see above). We therefore focus on the attribution of forgiveness only in this manuscript but emphasize that future research may examine audience effects on the attribution of punishment along the lines of the present research.
6. Although, theoretically, individualistic and competitive motive attributions are considered distinct dimensions (Gollwitzer & Okimoto, Citation2021), we treat them as one mediator in our power analysis, given that we assume them to be highly correlated.
7. This questionnaire was followed by four items measuring participants’ perception of the victim’s response that are not directly derived from the motive-attribution framework (see materials on the OSF). However, these items are not pertinent for the present manuscript and will, thus, not be considered any further.
8. Note that we also measured participants’ perceived value consensus with the victim (three items, e.g. Okimoto & Wenzel, Citation2010). However, this measure is not pertinent to the present research (i.e. we did not include it in Pre-Study 2). Consequently, we will not discuss this any further.
9. Note that we initially conducted and reported these mediation analyses without bootstrapped standard errors for the indirect effects. Conducting this analysis with bootstrapping (5,000 iterations) yielded similar results with only few minor differences. We added this analysis to our analysis script on the OSF and report whenever bootstrapping yielded different results throughout the manuscript.
10. The indirect effects through competitive motives were significant when using bootstrapping (5,000 iterations), B = −0.23, SE(B) = 0.11, z = −2.02, p = .043.
12. Indirect effects on justice-related satisfaction via prosocial motives (B = −0.11, SE(B) = 0.06, z = −1.91, p = .056) were nonsignificant when using bootstrapping (5,000 iterations).
13. Indirect effects on avoidance intentions via prosocial motives (B = 0.04, SE(B) = 0.02, z = 1.72, p = .086) and competitive motives (B = 0.05, SE(B) = 0.02, z = 1.94, p = .053) were nonsignificant when using bootstrapping.
14. Bombari et al. (Citation2015, p. 2) define a virtual human as “a computer generated three-dimensional digital representation that looks and acts like a real human.” In previous literature, virtual humans were differentiated between avatars that were controlled by real humans and agents that were controlled by computer-algorithms. Given that the interaction partners in our study are virtual humans controlled by computer algorithms, we refer to them as agents in the following.