Cyberostracism: Effects of Being Ignored Over the Internet

Ostracism, the act of ignoring and excluding individuals, is a widespread and powerful social tactic observed across various species and cultures, from primitive societies to modern workplaces. Recent research by Kraut et al. (1998) highlighted a counterintuitive finding that increased internet usage correlated with increased loneliness and depression, suggesting that the lack of face-to-face interaction and the replacement of stronger social ties with weaker ones negatively impacts psychological well-being. This concern is amplified by observations that internet users frequently feel ignored. This study aims to examine the impact of ostracism in a virtual environment, termed "cyberostracism." The researchers hypothesized that even minimal online ostracism would elicit negative psychological reactions similar to those seen in face-to-face settings. The study utilizes two experiments to investigate the effects of cyberostracism on self-reported feelings and conformity behavior, thereby testing Williams's (1997) need threat theory of ostracism, which posits that ostracism threatens four fundamental needs: belonging, self-esteem, control, and meaningful existence.

Existing literature extensively documents ostracism's prevalence and impact. Studies show its use in regulating social behavior in animals and its occurrence across various human contexts, including schools, workplaces, and interpersonal relationships. Research in social psychology underscores the aversive consequences of ostracism, including negative mood, loneliness, anxiety, and feelings of helplessness. These negative reactions are intensified when ostracism is attributed to personal shortcomings or is more blatant. Even imagining or role-playing ostracism elicits negative self-evaluations. Williams's (1997) model provides a unifying framework, defining ostracism as acts of exclusion or ignoring, regardless of intent. This model emphasizes the unique threat ostracism poses to four fundamental needs, leading to initial psychological discomfort and subsequent coping mechanisms aimed at restoring these needs. This study extends this research by investigating the impact of cyberostracism, where the perceived exclusion occurs in virtual environments.

Experiment 1 utilized a novel online paradigm involving a virtual ball-tossing game. 1720 participants initially accessed the website, with 1486 completing the study after removing those who logged off early. Participants, drawn from 62 countries, were randomly assigned to one of four conditions: overinclusion (67% chance of receiving the ball), inclusion (33%), partial ostracism (20%), and complete ostracism (0%). Before the game, participants completed a self-esteem scale. After the game, they completed questionnaires assessing need threat (belonging, self-esteem, control, meaningful existence), aversive impact (mood, intensity of ostracism, group cohesiveness), and their perception of the percentage of throws they received. Experiment 2 modified the paradigm, using a "cyberball" game with increased animation. 231 participants initially accessed the website with 213 completing the experiment after removing those who were suspicious of the study's purpose. Participants were randomly assigned to a 3 (group membership: in-group, out-group, mixed-group) × 2 (ostracism manipulation: inclusion or ostracism) design. After the cyberball game (10 turns), participants performed a perceptual task with responses of five other (ostensible) individuals displayed before their turn to assess conformity. PC and Macintosh users were used as in-group/out-group categories. Post-experimental questionnaires measured perceived throws, belonging, and conformity. Thought bubbles allowed participants to record thoughts during the cyberball game.

Experiment 1 demonstrated that the cyberostracism manipulation was successful, with participants accurately perceiving the percentage of throws they received. Aversive impact significantly increased with higher levels of ostracism. Lower self-esteem correlated with higher aversive impact. Mediation analysis showed that threatened belonging and self-esteem mediated the relationship between ostracism and aversive feelings. Experiment 2 revealed that ostracized participants exhibited higher conformity rates compared to included participants, supporting the hypothesis that ostracism motivates individuals to restore their sense of belonging. However, the interaction between group membership and ostracism on conformity was not significant. The mixed-group condition showed high levels of conformity despite accurate reporting of ostracism and decreased belonging, suggesting that ostracism by a mixed group is particularly aversive.

The findings demonstrate the potency of ostracism, even in a minimal virtual environment. The results offer strong support for Williams's (1997) need threat theory, particularly regarding the impact on belonging and self-esteem. The lack of significant effects on control and meaningful existence may be due to the characteristics of cyberostracism, suggesting that these needs may be less salient in brief virtual interactions or that the methodology did not adequately capture these constructs. The increased conformity in ostracized participants highlights the efforts to re-establish a sense of belonging. The lack of interaction between group membership and conformity suggests that ostracism, regardless of the source, motivates conformity to a new group.

This research demonstrates the profound impact of cyberostracism, highlighting the pervasiveness of the need to belong. The findings support and extend the need threat theory of ostracism. Future research should explore the long-term effects of cyberostracism, the impact on other needs, and the influence of various online social contexts. Further investigation into the moderating effects of different types of online ostracism and individual differences is also warranted.

The study's limitations include the potential for self-selection bias due to high dropout rates, particularly in Experiment 2. The use of computer-generated players may not perfectly replicate the complexity of human interaction. The group membership manipulation in Experiment 2, while effective, may have been stronger than many other minimal group manipulations, which might influence the results. The reliance on self-report measures limits the scope of the conclusions drawn.