Ostracizing for a Reason: A Novel Source Paradigm for Examining the Nature and Consequences of Motivated Ostracism

ABSTRACT

Ostracism, a complex social phenomenon, involves both targets (ostracized individuals) and sources (ostracizers). The current experiment redressed a gap in the ostracism literature by devising a novel, three-phase paradigm to investigate motivated ostracizing. In the current study, 83 females were assigned to one of four conditions during a Cyberball game: motivated sources chose to ostracize an obnoxious fellow player, induced sources ostracized a fellow player at the behest of the experimenter, targets were ostracized, and included participants received the ball proportionately. Analysis of participants’ primary needs, emotions, ratings of their co-players, and behavior toward their co-players indicated that being the target of ostracism was a robustly aversive experience. Both motivated and induced sources reported fortified control. Moreover, a motive for ostracizing influenced source experience: induced sources experienced greater levels of negative moral emotion and behaved more prosocially toward their target than motivated sources. The flexibility and demonstrated impact of this novel paradigm adds to the toolkit available to researchers interested in expanding insight into the psychological processes underlying, and the motivational and behavioral outcomes of being, a source of ostracism.

Keywords:

• motivated sources
• ostracism
• primary needs
• source paradigm

Notes

1. After the taste-test task, but before reporting demographics, participants completed a delayed Primary Needs Questionnaire (PNQ) interspersed with filler items. This questionnaire consisted of the same items used in the immediately following Cyberball; however, the instructions asked them to answer the questions according to how they felt “right now.” Analysis of self-esteem, belonging, meaningful existence, and control need levels as a function of condition revealed main effects of need type, F(2.85,176.40) = 117.87, p < .001, η_p² = .66, and condition, F(3,62) = 6.92, p < .001, η_p² = .25. Bonferroni-adjusted post-hoc tests revealed that levels of primary needs all differed significantly from one another (ps < .001, ds ≥ 0.50), with meaningful existence needs more satisfied than belonging needs, which were more satisfied than self-esteem needs, which in turn were more satisfied than control needs. Further, total need satisfaction (collapsing across need type) was lower among targets compared to participants in all other conditions (ps ≤ .02, ds ≥ 0.77), who did not differ from one another (ps > .99, ds ≥ 0.27). The interaction between primary need type and condition was not significant, F(8.54,176.40) = 0.58, p = .80, η_p² = .03.

2. In the current study, two participants in the motivated-source condition received a slight variation of the paradigm (i.e., Student 2’s speech was less derisive, but still obnoxious). Both of these participants engaged in complete ostracism of Student 2. We have included their data in the analyses reported here. Exclusion of their data from reported analyses resulted in no substantive changes to the nature of the reported findings, barring one post-hoc comparison: participants in the motivated-source condition reported only marginally higher levels of felt inclusion than participants in the inclusion condition (p = .06).

3. Mauchly’s test of sphericity revealed that sphericity was violated for primary needs ratings, χ²(5) = 18.10, p = .002. Hence, degrees of freedom were corrected using the Huynh-Feldt estimate of sphericity (ϵ = .91).

4. Chow et al. (2008) documented that increased anger among targets causally explained increased antisociality in the taste-test task. In our data, exploratory analyses of just the self-reported anger variable revealed that, whereas anger was elevated among targets compared included participants, t(31) = 2.86, p = .008, d = 0.98, anger did not correlate with antisocial behavior toward Students 1 or 2 in these conditions, Student 1: r(33) = −.22, p = .22; Student 2: r(33) = −.11, p = .55. There are many situational constraints on anger-induced aggression (e.g., Denson, 2013), and it remains a question for future research to determine the circumstances in which ostracism-induced anger guides antisocial behavior toward sources.

Additional information

Funding

This project was supported by an Australian Research Council Discovery Grant (DP110105195).

Notes on contributors

Sarah L. Gooley

Sarah L. Gooley and Lisa Zadro are affiliated with the University of Sydney.

Lisa Zadro

Sarah L. Gooley and Lisa Zadro are affiliated with the University of Sydney.

Lisa A. Williams

Lisa A. Williams is affiliated with UNSW Australia.

Elena Svetieva

Elena Svetieva is affiliated with Católica-Lisbon School of Business and Economics.

Karen Gonsalkorale

Karen Gonsalkorale is affiliated with the University of Sydney.