Reactions to a woman’s voice in an FPS game: The moderating role of status and skill

Michael Kasumovic (University of New South Wales) is an evolutionary biologist, has posted his write-up of a study at The Conversation and even has a comic version. Its data is quite familiar though.

In early 2013, I blogged about a field experiment on players’ reaction to a woman’s voice in an FPS game. Its popular take home message was that a female player received three times as many negative messages as male players do. There was some press attention and I believe the study is mentioned quite often in sexual harassment panels at videogames conventions.

The field experiment’s data was once again explored with a different perspective and a fine-tooth comb on particular variables previously explored in the 2013 study: in-game skill, performance and status. Michael Kasumovic provided an evolutionary perspective of the data and Jeffrey Kuznekoff (Miami University) is the communication scientist who originally conducted the field experiment.

Abstract

This image is still relevant today as it was used in 2013 (Credit: Emmy Cicierega)

So how does the evolutionary perspective relate to videogames? The answer is competition. Long ago in humanity’s evolutionary history, men had to be open about their dominance. So they fought other men to establish a social hierarchy/leaderboard/pecking order for resources and mates. In contrast for women, such fight is less important and quite costly, instead they opted for subtle strategies by reducing the social status of rivals and minimize retaliation. Think about the contrasting strategies between the men and women in Game of Thrones (of course, there are exceptions).

This evolutionary sex difference that established social hierarchies of the past are now at odds with today’s hierarchical demands. Both men and women are competing in areas where sex differences are less relevant. For example, lawyers on a court case or competing on a videogame. The lingering belief of sex difference in establishing hierarchy, such as men’ overt dominance, might explain why women faced a lot of hostility in competitive environments; men are still competing as if it was between men.

The authors reasoned that videogames are good places to test the evolutionary perspective and examine intrasexual competition. Men and women are equally playing videogames; competitions can be taken seriously and there are monetary stakes in professional e-sports; performance is no longer dictated by physical strength, but by response times and cognitive capacity which can be improved through training. On the other hand, the social environment of videogames is still masculine through its sexual stereotyping of characters and the high male-to-female ratio in certain competitive genres.

The authors would not examine videogames from a purely evolutionary perspective, this would miss the importance of social interactions. So, they examined the data with a social constructionist perspective which posited that sexist behaviours were meant to remove women from a male-dominant environment. From the evolutionary perspective, sexist behaviours are responses to a man’s threatened hierarchical status, his ranking is overtaken by a female. This is significantly visible among low-status men whose status is precarious.

Method

They used the data that was published in 2013.

The 2013 study used three Xbox Live accounts for each condition: voiceless, female and male. The female and male voice had pre-recorded phrases they used for the pre-game lobby, during playtime and the post-game lobby. The phrases were innocuous, such as ‘I like this map’, ‘nice shot there’, ‘I had fun playing that game’ (more examples can be found in the Supporting information section under ‘ Language used in audio recordings’)

The videogame used was Halo 3. The game had a matchmaking system grouping players of similar skill level. The skill level, ranging from 1-50, is visible to all players in the lobby. Team Slayer was used as it was the most popular playlist where 25% of all online play occurs in that playlist. The researchers had one experienced player and an inexperienced player.

They recorded 245 matches. 82 in the voiceless condition, 81 in the male voice condition and 82 in the female voice condition. Each game had typically seven other players, which resulted 1711 players who players along with the researchers, 1660 were unique gamers, so quite a small chance of meeting the same player.

The current study

For the current study, they examined the data from the female and male voice matches. The reason is aligned with the authors’ goal of comparing reactions between the male and female voice and the evolutionary perspective.

Second, they focused on individual players’ skill level and performance, this is different from the 2013 study when they examined the average skill level across players. You could say that the 2013 study combed the skill level data with a brush and did not found anything, but the 2015 study combed it with a fine-tooth comb and found something. They examined players’ performance as well as their relative performance to the researcher-player, as indicated by:

They started with 163 matches out of the 245. From those 163 matches, only 102 had any verbal communication and all of the other players’ voices were men, no women were heard during data collection. From these 102 matches, they identified 189 unique players. Of these 189 players, 147 were in the same team as the researcher-player and 42 were on the opposite side. As the numbers are few for the opposite side, the authors decide to examine the teammates’ reactions as they interacted with the male and female voice much longer than the opponents do and the latter interacted with them mostly in the post-game lobby and for less than 3 minutes.

Results

The data can be found online along with their statistical analysis script at github.

The authors hired a professional transcription company to transcribe the verbal interactions. Two coders, one of them is the author, categorized the comments directed towards the researcher-player. These comments were classified as either positive, negative or neutral. I must note the researchers categorized these responses at this level as parsing them out even further gets more subjective (Example comments can be found in the Supporting Information section under ‘Example comments from other players’).

Example positive comments to male voice:

Example positive comments to female voice:

For directed positive, there was a positive correlation with the number of deaths, more deaths resulted in more directed positive comments. However, there is an interaction effect between the voice type and maximum skill level, lower-skilled players uttered less positive comments than higher skilled players when interacting with the female voice. Further examination on relative performance revealed that players with more deaths than the researcher-player uttered more positive comments. Relatively higher skilled players also uttered more positive comments. A near significant interaction was found in that higher skilled relative to the female voice uttered more positive comments whereas they uttered less comments if she was higher skilled than them. Quite interesting, but this finding requires more data for further verification. A negative correlation was found for relative number of kills, in that players with fewer kills than the researcher-player uttered more positive comments. So far, the male and female voice seem to receive positive comments quite fairly, except in the case of skill level.

Example negative comments to male voice:

Example negative to female voice:

For directed negative, the story is very different. First, two players had to be excluded from the analysis because they uttered more than 10 negative comments to the female voice greatly skewing the results. Second, similar from the 2013 study, the female voice received 2.66 (± 0.36) negative comments as compared to the male voice 1.90 (±0.34).

Higher skilled players uttered fewer negative comments. More deaths correlated to more negative comments, however the female voice received more negative comments when the players died less whereas the male voice received more negative comments the players died more often. The number of kills had no relation with negative comments with the male voice, but number of kills do have a relation for the female voice in that fewer kills lead to more negative comments whereas numerous kills had fewer negative comments.

In terms of relative performance and directed negative comments, players who died more than the researcher-player uttered more negative comments as well as players who made fewer kills than the researcher uttered more negative comments. There was no differences between the male and female voice.

Example neutral comments to male voice:

Example neutral comments to female voice:

For directed neutral, players who died more uttered more neutral comments. In terms of relative performance, players with more kills than the researcher uttered more neutral comments. There was no difference between the male and female voice.

Given the high number of negative comments directed towards the female voice, the authors examined for hostile sexism. They found 11 players (13% of 82 players) had uttered hostile sexist comments.

Discussion

The take home message is that men’s negative and positive comments are influenced by their in-game performance and whom they are interacting with, be it male or female. As informed by the evolutionary perspective, low-status and poor performing males responded negatively to female competitors because their hierarchical status is visibly threatened. In other words, men reacted negatively as their worth as a man is outweighed by a higher performing woman. On the other hand, the evolutionary perspective revealed that higher performing men uttered more positively, a switch to a supportive and potentially mate attraction role according to the author.

IMO from another perspective, a man who is outperformed by a woman within a masculine domain, such as first-person shooters, and in view of other men and women can be interpreted as a loss of manhood. This could apply when the man is seen performing equally to a woman. Given how men feel that their manhood had to be earned and how easy to lose it (see Vandello et al., 2008), the man may felt anxious and must alleviate it by dominating her to her “proper” place, an act perceived to be appropriately masculine. Thus, the man’s negative comments are attempts to reassert his precarious manhood in public.

The authors discussed the differential treatment of positive comments between the male and female voice. From an evolutionary perspective, poor performing males uttered more neutral and positive comments as a way of demonstrating submissive behavior to the male voice. Such pattern is way to avoid punishment from their teammates for their failure of cooperation.

The authors continued their evolutionary discussion by relating to dominance. The evolutionary argument posit that dominance rather than attractiveness is what determines male mating success even in today’s society. Those at the bottom of the dominance hierarchy have the lowest chances of mating success as they are superseded by women, the thought of not being worthy for a woman is rather unsettling. As aggression is often used by men in maintaining their social status, the evolutionary perspective gives us the explanation of why female players get demeaning comments that questioned their gamer identity (i.e. “fake girl gamer”), performance (i.e. “girls only casual”), gender identity (i.e. “must be a fake girl”) and their mating potential (i.e. “fat, ugly, or slutty” or “attention whore”). Such demeaning comments act as a way to deny female players’ performance and suppress their disruption in the male hierarchy. Furthermore, the comments resembled negging, questioning the female players’ worth and simultaneously making the male player more worthy. These acts may or may not be deliberate, but they are certainly reactive consistent with the evolutionary perspective.

Here are some of my observations of the findings. Given that the study analysed interactions with team members, this relates to cooperation videogames studies. These studies suggest that cooperation increase prosocial behaviours (see John Velez’ research). In addition, two studies have found this effect to be true when teaming up with an outgroup, say teaming up with a student from a rival university (Velez et al., 2012; Adachi et al., 2014). However, there are caveats from these studies. First, these studies examined interactions between two players, the effect will change when the team size increases. Second, even though it is between different groups, they are between the same sex. The interactions between men and women is very different and few studies have examined gender dynamics in videogames.

The few mixed-sex videogame studies resonate with the current study’s findings. These studies highlight women’s self-perception within the socially masculine environment of videogames. Eden et al. (2010) found that genre rather than skill affect participants’ perception of players’ gender, such that participants assumed players of shooting or fighting games were men. Vermeulen et al. (2014) and Ratan et al. (2015) found that female players felt less skilled and less confident about their own gaming abilities despite being on par with their male counterparts’ abilities through objective measures. This is especially true when they are playing with men and within highly competitive environments. The reasoning for such perception among female players is the social environment’s hostility to women and this is reflected by the current study’s findings: the high proportion of negative comments female players receive (see Bertozzi, 2008 for further reading). As a reminder, competitive videogames is an equal arena for men and women in terms of cognitive and physical abilities, it is rather the cultural environment that favours men over women. The authors suggested that we should teach boys that losing to girls is not socially painful. It’s a game of equals.

References

Adachi, P. J. C., Hodson, G., Willoughby, T., & Zanette, S. (2014). Brothers and sisters in arms: Intergroup cooperation in a violent shooter game can reduce intergroup bias. Psychology of Violence. DOI: 10.1037/a0037407

Bertozzi, E. (2008). ‘you play like a girl!': Cross-Gender competition and the uneven playing field. Convergence: The International Journal of Research into New Media Technologies, 14 (4), 473-487. DOI: 10.1177/1354856508094667

Eden, A., Maloney, E., & Bowman, N. D. (2010). Gender attribution in online video games. Journal of Media Psychology: Theories, Methods, and Applications, 22 (3), 114-124. DOI: 10.1027/1864-1105/a000016

Kasumovic, M. M., & Kuznekoff, J. H. (2015). Insights into sexism: Male status and performance moderates Female-Directed hostile and amicable behaviour. PLoS ONE, 10 (7), e0131613+. DOI: 10.1371/journal.pone.0131613

Kuznekoff, J. H., & Rose, L. M. (2013). Communication in multiplayer gaming: Examining player responses to gender cues. New Media & Society, 15 (4), 541-556. DOI: 10.1177/1461444812458271

Ratan, R. A., Taylor, N., Hogan, J., Kennedy, T., & Williams, D. (2015). Stand by your man: An examination of gender disparity in league of legends. Games and Culture, (pp. 1555412014567228+). DOI: 10.1177/1555412014567228

Vandello, J. A., Bosson, J. K., Cohen, D., Burnaford, R. M., & Weaver, J. R. (2008). Precarious manhood. Journal of Personality and Social Psychology, 95 (6), 1325-1339. DOI: 10.1037/a0012453

Velez, J. A., Mahood, C., Ewoldsen, D. R., & Moyer-Gusé, E. (2012). Ingroup versus outgroup conflict in the context of violent video game play: The effect of cooperation on increased helping and decreased aggression. Communication Research, 41 (5), 607-626. DOI: 10.1177/0093650212456202

Vermeulen, L., Castellar, E. N., & Van Looy, J. (2014). Challenging the other: Exploring the role of opponent gender in digital game competition for female players. Cyberpsychology, Behavior, and Social Networking, 17 (5). DOI: 10.1089/cyber.2013.0331