7  General Discussion

In my dissertation studies (total N = 2142), I examined whether playing a single-player video game alone replenishes belonging after social rejection, a potential disengaged–prosocial response to social rejection in the bi-dimensional rejection taxonomy (Sunami, Nadzan, and Jaremka 2020). In Study 1, I validated a single-item measure of belonging, the Heart Manikin, used as the primary outcome for the subsequent studies. In Study 2, I examined whether recalling one’s time playing a social surrogate video game vs. a non-social surrogate video game can replenish belonging after social rejection. Lastly in Study 3, I examined whether rejected people who play a video game with varying degrees of parasocial relationship and social world contents can replenish their belonging. The results of the studies did not support the social surrogacy hypothesis. I discuss the implications of the current nulls results below.

7.1 Impact on the Social Rejection and Video Games Literature

7.1.1 Heart Manikin as a Quick Measure of Belonging

In Study 1, I attempted to validate the Heart Manikin. Results indicated a strong evidence for the convergent validity of the heart manikin with belonging-related measures, including a sense of belonging, self-esteem, control, and meaningful existence. The Heart Manikin scores also converged consistently with measures of positive affect. On the other hand, I found moderate evidence for the discriminant validity of the Heart Manikin. The Heart Manikin scores did not correlate with unrelated constructs, such as interpersonal reactivity, paradoxical mindset, sleep quality, abuse perpetration, food craving, body image, and subjective socioeconomic status. However, the measure did not show discriminant validity against measures of arousal and dominance. Overall, I suggest that the Heart Manikin has a strong convergent validity with measures of belonging and positive valence. Future studies in social rejection research could use this measure to efficiently measure state belonging. Given the moderate discriminant validity of this measure, I recommend researchers to use other concurrent measure if they wish to measure belonging that is independent from arousal, dominance, and possibly subjective socioeconomic status.

I also tested whether the Heart Manikin scores were sensitive to the laboratory manipulation of social rejection. Across Studies 1c, 1d, and 1e, I observed that participants in the rejected condition reported lower Heart Manikin scores than those in the non-rejected condition, supporting the sensitivity of this measure. I suggest that the Heart Manikin can be a useful, quick tool to check an effectiveness of a social rejection manipulation.

7.1.2 Possible Explanations for the Null Results and Implications to Social Surrogacy Hypothesis

The current results are not consistent with the social surrogacy hypothesis on surface (Gabriel and Valenti 2017). In two studies, I observed that rejected participants did not replenish their belonging after writing about a video game with social surrogates (compared with writing about a video game without them), and after playing a novel video game with higher parasocial relationships and social world content (vs. low parasocial relationships and social world content). Below I speculate why I observed null results.

Ineffective Manipulations on Social Surrogacy. The manipulations for the social surrogates used in Studies 2 and 3 were new, and thus they have never been validated to manipulate social surrogacy. That being said, I expected that these manipulations were reasonable to induce social surrogacy for the following reasons.

In Study 2, I used a role-playing game for social surrogacy essay since role-playing games often present strong relatable characters and immersive stories and social world. I contrasted these games with non role-playing games that usually do not have these components. In Study 3, I developed a novel role-playing game to independently manipulate parasocial relationships and social worlds, Since the game was new, I was able to avoid any influence from participants’ previous exposure to the game’s characters or stories.

Yet, I did not observe results consistent with the social surrogacy hypothesis. To explore the effectiveness of the manipulations, I used exploratory manipulation checks. Results for the manipulation checks were inconsistent across Studies 2 and 3. In Study 2, participants reported higher parasocial relationships and social worlds in the social surrogacy condition, compared with the non-social surrogacy condition. In Study 3, participants reported similar levels of parasocial interactions and social worlds, regardless of the type of the video game they played. Overall, I do not conclude about the effectiveness of the manipulations given that the manipulation check items were never used to validate manipulations. Future studies should investigate how we effectively induce social surrogacy (parasocial relationships and social worlds) and how we can measure these constructs in a validated manner.

Type II Error. An absence of an effect does not mean that the true effect is absent—it can mean Type II error, missing a true effect. But, I suggest that Type II error is unlikely since (a) I ensured that all studies were powered to detect an effect of (d = 0.35), and (b) the null findings are consistent across studies. That said, the current studies could only capture an effect size that is larger than (d = 0.35). If the true effect of social surrogacy on belonging was smaller than this hypothesized effect size, the current studies could not detect the effect.

Towards Refining the Theory. Another possibility is that the social surrogacy hypothesis may not be robust in its current form, and the theory needs to identify boundary conditions and expected effect sizes of social surrogates on belonging. Empirical evidence supporting the social surrogacy hypothesis has mainly come from studies on books and TV programs, but not video games. One possibility is that the social surrgacy hypothesis operates better in reading books and watching TV programs, but apply less to playing video games. Future studies should examine these possibilities.

7.1.3 Implications to Video Game Studies

In the video games literature, accumulating theoretical work and qualitative evidence suggest that video game players can feel being connected with characters in the game, and thus satisfying relatedness needs (Bopp et al. 2019; Burgess and Jones 2020; Poretski et al. 2019; Tyack and Wyeth 2017). However, no experimental studies have tested this possibility, and the current studies offered initial experimental tests of this possibility.

The current null results did not find that participants satisfied belonging (relatedness needs) by writing about video games (Study 2) or by playing one (Study 3). However, I did find that participants reported forming more parasocial relationships with non-player characters, more immersion, more engagement with the narrative, more social world, and more enjoyment for a social surrogacy game (e.g., a role-playing game), compared with a non-social surrogacy game in Study 2. Moreover, people who enjoyed their video game more reported feeling more belonged and happier, forming more parasocial relationships with characters, engaged more with the narrative, and immersed more into the story (see the bivariate correlation analysis in Study 3). These results at minimum suggest that belonging, paraoscial relationships, social worlds, and enjoyment are interrelated in video games.

7.2 Possible Impact on Society

All humans have a fundamental need to belong, and when this need is threatened, people experience adverse mental and physical health outcomes (Cacioppo et al. 2006; Hawkley et al. 2010; Jaremka et al. 2013). People experience threats to belonging in everyday life (Nezlek et al. 2012). Identifying an effective strategy to replenish belonging after social rejection will help efforts to develop interventions to protect belonging, and ultimately improve mental and physical well-being. One step for identifying such intervention is to measure belonging in a quick and effective way. The Heart Manikin validated in my dissertation can be an ideal tool for a large-scale research that requires less cost per participant.

The current null results for the social surrogacy hypothesis do not offer clear strategies to reduce threats to belonging. However, I did find that participants who enjoyed a video game reported higher belonging compared with those who did not enjoy across studies. Future studies could explore whether playing an enjoyable video game has a positive impact on belonging vs playing an unenjoyable video game. Such evidence could add to the broader conversation about the benefits of playing a video game (Granic, Lobel, and Engels 2014).

7.3 Constraints on Generality and Future Directions

I discuss the constraints on generality of the present findings (Simons, Shoda, and Lindsay 2017) to highlight any design or sample characteristics that can impose constraints on interpretation of the results and future directions in this section.

7.3.1 Social Surrogates in Non-Rejected People

Across the current studies (Studies 2 and 3), all participants experienced acute social rejection before seeing social surrogates. Thus, the current studies did not test whether non-rejected people can increase belonging, or whether people with chronic feelings of social rejection (e.g., loneliness) could replenish belonging via social surrogacy in video games, both important directions for future investigations.

7.3.2 Negative Parasocial Relationships and Social Worlds

The current studies only examined positive parasocial relationships and social worlds. In Study 2, participants recalled their parasocial relationships and social worlds in their favorite game. In Study 3, participants experienced a friendly parasocial target and a positive social world. The results of the current study may not generalize to other video games where people have a negative parasocial relationship with characters, or negative experience being immersed in a social world. People can hate characters in TV programs (Chory 2013; Jennings and Alper 2016)—likewise, people can hate non-player characters and form a negative parasocial relationship. People can also immerse themselves in negative social worlds—social worlds that are immoral or ethically unjust, such as ones described in many horror films (e.g., the Saw Franchise, the Texas Chain Saw Massacre, etc.). Future research should carefully consider the nature of the parasocial relationships and the social worlds in video games, and whether they can replenish or even hurt belonging.

7.3.3 Another Type of Social Surrogacy: Reminders of Others

The social surrogacy hypothesis identifies three types of social surrogates: parasocial relationships, social worlds, and reminders of others (Gabriel and Valenti 2017). In my dissertation, I focused on parasocial relationships and social worlds but not reminders of others—remnants of real social relationships, such as photographs of close others, comfort foods prepared by loved ones. I did not focus on the remainder of others because the current definition of reminders of others requires a real preexisting social relationship that is absent in single-player video games.

A new avenue for research may be to examine if the definition of reminders of others includes the parasocial relationship and social worlds. People can play video games to remind themselves of past parasocial relationships and social worlds experienced previously in video games—especially those multiple releases over time. For example, long-time players of the Animal Crossing series can play a newly released Animal Crossing: New Horizons in 2020 (E. P. D. Nintendo 2020), and remember about the parasocial relationships they formed with the older game such as Animal Crossing: New Leaf in 2012 (E. A. D. Nintendo 2012). Similarly, playing Witcher 3 in 2015 (CD Projekt Red 2015) can remind the player of their time immersed in the social world in the first Witcher in 2007 (CD Projekt Red 2007). Future studies can examine whether people can replenish belonging via remainders of the parasocial relationships or social worlds.

As mentioned in Study 2, I speculate that people may have replenished belonging by remembering their memories of playing a single-player video game in the presence of a close other. For example, people can feel loved by simply remembering their time playing Tetris in front of their friends. This way, people replenish belonging, not because of the content of the game, but because of the time shared with their friends. Future studies that focus on single-player video game can ask participants to report how much they spend playing a video game in front of others (e.g., passing controllers to each other, or simply letting someone watch the game). A novel hypothesis is that participants can replenish their belonging by remembering a video game that they shared playing with close others, similar to replenishing belonging via comfort food (Troisi and Gabriel 2011; Troisi et al. 2015).

7.3.4 Character Identification

In the present dissertation, I focused on parasocial relationships with the non-player characters in video games. However, existing studies suggest that players can be emotionally attached to the player characters they control—such as Commander Sheperd in Mass Effect and Geralt of Rivia in Witcher (Bopp et al. 2019). According to the current definition of a parasocial relationship, the relationship between the player and the player character may not be considered as parasocial since the relationship can be two-sided: the player can control and influence the player character’s behavior, which in turn influences the player’s behavior (Banks 2015; Banks and Bowman 2016; Cohen 2014). Thus, the relationship players form with the player character may not fall under the concept of social surrogates. However, players can be emotionally attached to the player character and thus may replenish their sense of belonging. Indeed, theoretical discussions suggest that players can form parasocial relationships with player characters with established backgrounds (e.g., Lara Croft from the Tomb Raider series), but not with the avatars that they create themselves [e.g., the player character in Skyrim; Kavli (2012); Lewis, Weber, and Bowman (2008)].

I explored the role of player character identification in Study 3. In general, identification with the player character was unrelated to belonging. However, among participants who played a game without a parasocial relationship target, those identifying more with the player character reported higher belonging than those identifying less with the player character. This association was absent among participants who played a game with a parasocial relationship target. One possible explanation for these results is that participants identifying with the player character (Higra) were more likely to experience higher belonging in an absence of the parasocial target (Sashu) since the player can focus more on the player character. On the other hand, in the higher parasocial relationship condition, players’ attention was divided between the player character and the parasocial relationship target. Future studies can investigate whether players benefit form from certain player characters to form parasocial relationships and immerse themselves in the social worlds.

7.3.5 Multiplayer Gameplay

The current dissertation focused on solo gameplay because of its focus on social surrogates—non-human entities that can satisfy the belonging need (Gabriel and Valenti 2017). Accordingly, the current results do not generalize to multiplayer gameplay. One unexamined avenue for future research is to understand the impact of a common social surrogate shared by two real players. For example, two players can form a parasocial relationship with the same non-player character or immerse in the same social worlds (Gabriel, Valenti, and Young 2016). For example, two players of the Massively-Online Multiplayer Game, Final Fantasy XIV (Enix 2010) can simultaneously form a parasocial relationship with Gigi or become members of the same guild. Similarly, these players also share knowledge about the stories of Final Fantasy. Having shared parasocial relationships or social worlds may benefit real social relationships. Indeed, couples who consume media together tend to have better relationship quality, and thus higher belonging (Gomillion et al. 2017). Taken together, I expect that people who experience social surrogates together with close others will report higher belonging than those who experience them alone.

7.3.6 Mechanisms

In the proposal, I planned to speculate on possible mechanisms if I find that rejected people replenished their belonging via social surrogates in video games. I speculated that a video game player can experience positive emotions, which can facilitate replenishing belonging (Williams 2009). Or, they can experience a sense of confidence and self-esteem in playing a video game, and this increase in self-esteem could increase belonging consistent with the sociometer hypothesis (Leary et al. 1995). Rejected people can also play a single-player video game to simply distract themselves, and distraction can replenish belonging (Hales, Wesselmann, and Williams 2016; Nadzan, Jaremka, and Sunami 2019; Wesselmann et al. 2013). However, I did not find that rejected people replenished belonging by social surrogates in the current studies in the first place. Future research should investigate whether social surrogates in video games can replenish belonging first before investigating mechanisms.

7.4 Conclusion

My dissertation examined whether people can replenish their belonging following social rejection by playing a single-player video game with social surrogates. The results did not support the social surrogacy hypothesis. I do not have a strong evidence that can explain the current null results. Possibilities of the null results include ineffective manipulations of social surrogacy, Type II error, and unexplained boundary conditions.

I note that many past studies in social psychology focused on investigating negative effects of playing video games (for discussion, see Anderson et al. 2010; Hilgard, Engelhardt, and Rouder 2017). I took a different perspective and focused on possible positive influence of the gameplay (see Adachi and Willoughby 2013; Granic, Lobel, and Engels 2014 for similar perspectives). I hope my dissertation contributes towards a more nuanced understanding of video games and how they influence social well-being.

Despite the technological advances to connect us better, social rejection continues to be an everyday experience of modern human life. I hope my dissertation contributes the way for future efforts to better understand the role of video game in belonging, and eventually contributes to developing ways to mitigate the detrimental effects of social rejection.