The Role of Burden and Deviation in Ostracizing Others

ABSTRACT

Ostracism (being excluded and ignored) is a painful experience, so why do individuals ostracize others? Previous research suggests individuals often ostracize those who are deviate, but not always. We posit that there may be two types of deviation, burdensome and non-burdensome, and the former is most likely to be ostracized. Study 1 manipulated burdensome deviation by programming a group member to perform more slowly (8 or 16 sec.) than others (4 sec.) in a virtual ball-toss game. Participants perceived slower players as more burdensome and deviate than normal speed players. Additionally, participants ostracized (gave fewer ball tosses to) the slowest player. Study 2 examined participant responses to both burdensome deviation (8- and 16-sec. players) and non-burdensome deviation (goth appearance). Participants again perceived the slower players to be burdensome and deviate, and ostracized them. They perceived the goth player to be deviate but not burdensome and did not ostracize this player.

Keywords:

• burden
• deviate
• deviation
• social control
• social influence
• sources of ostracism

ACKNOWLEDGMENTS

We thank Jeremy Rohrer for modifying Cyberball to include manipulating confederate player speed. We also thank Katie Wirth and Amanda Wesselmann for being our deviation models.

Notes

1. The literature on negative reactions to deviate group members sometimes uses the terms deviant or deviance. Deviance implies a negative evaluation but deviation is not always viewed negatively (e.g., Moscovici, 1976). It is unclear if laypeople make these conceptual distinctions, but for consistency and theoretical clarity we use the neutral terms deviate and deviation (e.g., Schachter, 1951).

2. We had other measures besides the one reported in this manuscript; some were filler items/personality measures to bolster the cover story and others were pilot measures (e.g., emotions toward the players; self-reported motives for ball-tossing). These measures were not germane to our hypotheses so we did not analyze them and do not included them here. These data can be found at osf.io/nrf72. Additionally, these data come from a larger data set in which we were piloting instruction manipulations. The instructions for the study reported told participants that the game will last 45 tosses; the other condition instructions specified 5 minutes. This time/tosses instruction manipulation had no significant effects on our dependent variables; therefore we excluded participants in the conditions specifying that the game would last 5 minutes. These data and measures are available from the first author by request.

3. We created the deviate appearance stimuli by taking two sets of pictures using two female confederates. We photographed each confederate in normal appearance and in goth appearance. We asked a pilot sample (N = 71) to rate each of these pictures on a 7-point scale ranging from 1 (not at all deviant) to 7 (extremely deviant). We informed participants: “Deviate means to differ from some commonly accepted standard of appearance, beliefs, or behaviors. In the United States, how much does this individual deviate from the norm?” We neglected to change the scale anchors from “deviant” to “deviate,” but as previously noted it is unclear if laypeople make conceptual distinctions between these two terms. In both pairs, participants rated the goth appearance picture as more deviate than the non-goth appearance picture (ts > 3.50, p < .01, ds > 0.38). Our stimuli successfully manipulated deviation, but the effect was small (Cohen, 1988) and the photograph means were both below the midpoint (2.21, SD = 1.41 and 3.38, SD = 1.62) so we decided to bolster the images by adding heaver makeup to our confederates, resulting in the images we used in the current study.

Additional information

Funding

This material is based upon work supported by the National Science Foundation under Grant No. 0519209 award to KDW.

Notes on contributors

Eric D. Wesselmann

Eric D. Wesselmann is affiliated with Illinois State University.

James H. Wirth

James H. Wirth is affiliated with The Ohio State University at Newark.

John B. Pryor

John B. Pryor and Glenn D. Reeder are affiliated with Illinois State University.

Kipling D. Williams

Kipling D. Williams is affiliated with Purdue University.