Can video games make you sexist?

Abstract

In the last years there has been an intense debate regarding sexist games. Some people have argued that this type of stimuli could have negative effects on users, like promoting discrimination or even violence against women. Correlational studies suggest that there is a connection between game usage and some forms of sexism. Only experimental research could tell us if there is a causal relationship. However, there is only a couple of experiments on the subject. In order to make sense of the limited evidence, we analyzed the results based on theories and concepts from Cognitive and Social Psychology. Our work suggests that, through mechanisms like Spreading Activation, Modeling, the Proteus Effect and Cognitive Dissonance, some forms of sexist game content could make people think in a more sexist manner. Nevertheless, given that researchers only collected measures immediately after exposure, there is no data on how long these effects lasted.

Full text

Can video games make you sexist?
Hugo Aranzaes
B.S. in Psychology
Highlights:
• Surveys show a connection between video game usage and sexism.
• Exposure to sexualized female characters could make sexist ideas more accessible.
• Non-violent sexist in-game behavior seems to be more influential than violent depictions of sexism.
• Enhancing identification with a sexualized female character might lead to more sexist beliefs.
• Feeling bad for harming female characters in games could make people adopt a sexist point of
view.
• There is no data on how much these effects lasted.
In the last years there has been an intense discussion regarding the presence of sexist content in
video games. Critics have pointed out, for example, how there aren’t enough avatars designed
towards women, or how female characters are used to make the male protagonist look strong and
brave, provide sexually arousing content or even satisfy fantasies about overpowering and harming
women. People have opposed this type of content based on the message it entails (for example:
women are weak), as well as the possible intentions behind its design (take revenge on female
figures). A third group, however, has suggested that sexist interactive content could have long term
effects on how the players think and behave. Something that could end up increasing discrimination
and even violence against women. Is there any scientific basis for these claims? Can video games
really make you sexist?
While there isn’t a lot of research on the subject (yet), correlational studies have been able to provide
some interesting results. Surveys have shown no relationship between video game usage and
general sexist beliefs (e.g. “men should be in charge”) (Breuer, Kowert, Festl, & Quandt, 2015). But
they did found a connection with hostile forms of sexist attitudes (“women want to take control away
from men”) as well as irrational beliefs about sexual violence (“women in a revealing dress are asking
for it”) (Fox & Potocki, 2015). On the other hand, playing violent video games has been associated
with negative attitudes toward women (Dill, 2009), a lower tendency to interpret unwanted sexual
advances as harassment (Dill, Brown, & Collins, 2008), and acceptance of myths about rape (“woman
refuse to have sex, even though they want to) (Dill, 2009; Dill, et al., 2008). The evidence suggests
that there is a link between video game usage and both negative attitudes against women and

tolerance towards sexual violence against them. However, this type of studies have their limitations.
While correlations are a great way to find connections, they don’t tell us what type of relationship
there is. Results could indicate that games make people sexist. But it could also mean that sexist
individuals prefer video games over other kinds of media, especially those with violent content. So,
are there any reasons to believe that playing a game can make someone think or behave in a sexist
way?
Psychologists have observed that, when a certain part of our memory is activated, any data closely
related to it becomes easier to access. This principle is called Spreading Activation (Collins, & Loftus,
1975) and based on it we could argue that, if someone is trying to figure out what to think or do,
previous exposure to sexist content should increase their chances of choosing the sexist alternative.
The evidence seems to be at least congruent with this idea. In different experiments, exposure to
content that included sexualized computer-generated female characters, was followed by faster
recognition of words used to objectify women (Yao, Mahood, & Linz, 2010)1, and a lower tendency to
interpret an unwanted sexual approach as harassment (Dill, et al., 2008)2, but not by more sexist
(Fox & Bailenson, 2009)3 or rape supportive attitudes (Dill, et al., 2008; Fox & Bailenson, 2009)4. One
possible explanation is that the sexualized females made any information related to both women and
sex more accessible. This meant that words used when referring to females as sexual objects were
going to be easier to recognize. It might also have made participants more likely to assume that a
woman interacting with a man was actively looking for a sexual relationship. Something that would
have made unsolicited male sexual advances seem less transgressive. On the other hand, showing
more sexist or rape supportive beliefs would have required the activation to spread to irrational ideas
like “females belong in the kitchen” or “women are weak and need male protection”, which are related
to women but not to sex. Something that was less likely to occur. It would appear that exposure to
sexist content can make people report more sexist ideas. But also that the effect will be limited to the
areas were the activation managed to spread (in this case, information related to both women and
sex). It should also be noted that, in each of the studies mentioned, there were other variables besides
the sexualized female characters that could account for the results5. So additional studies would be
needed before we can rule out other causes and explanations.
Character appearance isn’t the only attribute capable of influencing the players. According to the
concept of Modeling (Bandura, 1977), observing someone execute an action and be rewarded for it
is enough for people to start performing this behavior more frequently. Does this mean that games
depicting sexist acts could influence the users’ behavior? In an experiment, playing a game were the
protagonist seduces female characters by faking interest and lying to them, was accompanied by a
higher self-reported tendency to take sexual advantage of women (Yao, et al., 2010)6. However,
watching gameplay of a different title, were the protagonist gets a lap dance in a strip club, picks up
a prostitute on the street, has sex with her, shoots her to get back his money and then escapes the

police, wasn’t followed by more irrational beliefs about sexual assaults (Beck, Boys, Rose & Beck;
2012)7. Why was more sexist and violent content less influential? Well, models tend to work better
when the actions depicted are easy to replicate, and the rewards clear. Talking to women might have
been closer to the everyday life of the participants than going to a strip club or picking up prostitutes.
Also, users might have felt bad for the woman getting shot, or stressed by the idea of having to run
from the police, turning them into undesirable consequences8. What’s interesting about these results
is that they suggest that games which include more subtle and common forms of sexist behavior
could be more persuasive than those depicting blatant acts of discrimination and violence. Still, we
should also consider that our analysis is based on experiments that collected different (although
related) measures (the tendency to take sexual advantage of women and rape myth acceptance). So
using them as an indicator of when sexist models can or cannot influence the players might be a bit
of a stretch.
Another important variable to consider is the game-user interface. For example, it’s not uncommon
for players to see and interact with the environment through the eyes and body of a character. Under
these conditions though, individuals tend to behave as what is expected from their avatar’s
appearance. Something that has become known as the Proteus Effect (Yee & Bailenson, 2007). What
happens then when people play as a gender stereotype, which is a relatively common type of avatar?
Controlling a female character wearing a revealing dress, from a third person perspective, wasn’t
followed by more negative attitudes towards the appearance, occupations or physical abilities of
women (Behm-Morawitz & Mastro, 2009)9. But doing so from a first person perspective, while
watching the avatar replicate their head movements in a virtual mirror, did make a group of women
think more about their bodies (Fox, Bailenson, & Tricase, 2013)10. Having the point of view of an
avatar that reflected their moves probably made it difficult for participants to distinguish some of the
character’s attributes from their own. And given that the revealing dress was its more salient feature,
there was a higher chance that their doubts would revolved around the subject of physical
appearance, explaining why they thought more about their bodies. This result should be enough to
worry, as self-objectification has been associated with different problems (a small list can be found in
Fox, et al. 2013). But the effects didn’t stop there. When the avatar had a face that resembled the
user´s, acceptance of myths about rape increased (Fox, et al., 2013)11. Maybe the added similarity
made participants see themselves as a gender stereotype, at least to a certain extent. Assuming that
they also thought this type of character would endorse irrational ideas about gender roles and sexual
violence, it's not surprising that they supported statements like "women who wear revealing clothes
are asking for it". It seems that embodying gender stereotypes could promote sexist attitudes, at least
when identification with the avatar is technologically enhanced.
In order to make progress in a game, players have to overcome any obstacle they encounter.
Sometimes this involves having to perform actions they would rather avoid, like shooting a dog or

torturing a human being (virtually of course). Going through this type of experiences can make users
feel bad about themselves, and nobody likes to stay that way. What can they do to change this?
Cognitive Dissonance Theory (Festinger, 1957) states that, when people find contradictions between
their beliefs and/or behaviors, they tend to change the way they think. Doing so reduces the feeling
of incongruence and provides a certain level of relief. This usually means that people who find
themselves doing something they don’t approve, tend to reduce their discomfort by adopting beliefs
that support those actions. Something similar seems to happen when people play games that involve
violence against women. While a group of men played titles that required beating or shooting
sexualized female enemies to death12, higher levels of immersion lead to hostile forms of sexism and
low empathy towards the victims of sexual violence (LaCroix, Burrows, & Blanton, 2015)13. The level
of immersion indicates how real, engaging and emotionally arousing a media experience is. This
means that, the more immersed the players felt, the more vivid the characters and the damage
inflicted on them seemed. In this circumstances, having a positive image of their enemies or showing
sympathy towards them would have made participants feel worse about themselves. So, naturally,
they expressed a preference for beliefs that made their enemies look bad. Developers planning to
include gender violence should probably consider how uncomfortable they are going to make the
players feel, and how this can push them towards the adoption of more convenient, although
irrational, beliefs.
At first glance, research on the effects of sexist games seems a bit inconsistent14. Sometimes
exposure to this type of content is followed by more sexist attitudes, and sometimes it doesn’t. But
this isn’t surprising if we consider how each experiment exposed participants to different types of
stimuli, changing the psychological processes involved. As a result, the information affected varied:
Content that included sexist behaviors modified data related to social interactions, while embodying
a gender stereotype altered how users saw themselves15. In other words, the experiments showed
different results because they studied the effects of different things. When researchers performed
more than one experiment on the same subject, the results were more consistent (e.g. LaCroix, et al.
2015).
So, can video games make you sexist? Well, our analysis suggests that some types of sexist game
content can make people think (and potentially, behave) in a more sexist manner. Nevertheless, since
researchers only collected measures immediately after exposure, there is practically no data on how
long these effects lasted16. More importantly, we don’t know whether these processes play a
significant role in the development of a personality capable of committing acts of discrimination,
harassment or violence. Additional research would be needed before we can answer these questions.
In the meantime, the possibility of sexist games having long-term negative effects on players shouldn’t
be discarded nor accepted too easily.

References:
Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice Hall.
Beck, V. S., Boys, S., Rose, C., & Beck, E. (2012). Violence against women in video games: A prequel
or sequel to rape myth acceptance? Journal of Interpersonal Violence, 27(15), 3016–3031. doi:
10.1177/0886260512441078
Behm-Morawitz, E., & Mastro, D. (2009). The effects of the sexualization of female video game
characters on gender stereotyping and female self-concept. Sex Roles, 61(11-12), 808-823. doi:
10.1007/s11199-009-9683-8
Breuer, J., Kowert, R., Festl, R., & Quandt, T. (2015). Sexist Games = Sexist Gamers? A longitudinal
study on the relationship between video game use and sexist attitudes. Cyberpsychology, behavior,
and social networking, 18(4). doi: 10.1089/cyber.2014.0492
Collins, A. M., & Loftus, E. F. (1975). A Spreading-Activation Theory of Semantic Processing.
Psychological Review, 82(6), 407-428.
Dill, K. E. (2009). Violent video games, rape myth acceptance, and negative attitudes towards women.
In: E. Stark and E. S. Buzawa (Eds.), Violence Against Women in Families and Relationships: Volume
4, The Media and Cultural Attitudes (pp. 125-140). Westport, CT: Praeger.
Dill, K. E., Brown, B. P., & Collins, M. A. (2008). Effects of exposure to sex-stereotyped video game
characters on tolerance of sexual harassment. Journal of Experimental Social Psychology. 44(5).
1402-1408. doi: 10.1016/j.jesp.2008.06.002
Festinger, L. (1957). A theory of cognitive dissonance. Stanford: Stanford University Press.
Fox, J., & Bailenson, J. N. (2009). Virtual Virgins and Vamps: The Effects of Exposure to Female
Characters’ Sexualized Appearance and Gaze in an Immersive Virtual Environment. Sex Roles, 61(3-
4). doi: 10.1007/s11199-009-9599-3
Fox, J., Bailenson, J. N., & Tricase, L. (2013). The embodiment of sexualized virtual selves: The
Proteus effect and experiences of self-objectification via avatars. Computers in Human Behavior,
29(3). doi: http://dx.doi.org/10.1016/j.chb.2012.12.027

Fox, J., & Potocki, B. (2015). Lifetime video game consumption, interpersonal aggression, hostile
sexism, and rape myth acceptance: A cultivation perspective. Journal of Interpersonal Violence. DOI:
10.1177/0886260515570747
LaCroix, J. M., Burrows, C. N., & Blanton, H. (February, 2015). Sexualized representations of women
in video games: Psychological immersion predicts subsequent hostility toward women. Poster to be
presented at the 16th Annual meeting of the Society for Personality and Social Psychology, Long
Beach, CA.
Yao, M. Z., Mahood, C., & Linz, D. (2010). Sexual priming, gender stereotyping, and likelihood to
sexually harass: Examining the cognitive effects of playing a sexually-explicit video game. Sex Roles,
62(1-2), 77-88. doi: 10.1007/s11199-009-9695-4
Yee, N. & Bailenson, J.N. (2007). The Proteus Effect: The Effect of Transformed Self-Representation
on Behavior. Human Communication Research, 33, 271-290.
1 Men playing Leisure Suit Larry: Magna Cum Laude (a game about seducing attractive women
through mini-game controlled verbal interactions and were success is rewarded with scenes of the
female characters nude and having sex with the avatar), was followed by faster response times to
terms used to objectify women and a higher self-reported tendency to take sexual advantage of them,
than playing The Sims II (specifically, a stage of the real-life simulator were sex-oriented social
interactions are not possible), and Pac-Man II (which has no human characters).
2 People exposed to static images of sexualized women and physically strong men from video games,
reported a higher tendency to interpret a case of a male professor unexpectedly touching the leg of
a female student as sexual harassment, than those presented with portraits of congressmen and
women.
3 People wearing a virtual reality headset who performed a search task in front of a female agent
wearing a revealing dress, didn’t report more sexist beliefs or rape myth acceptance than those who
had to do it in front of a female character with an outfit that covered most of her skin.
4 It should also be noted that, while performing tasks in front of a sexualized female character didn’t
increase sexist beliefs or rape myth acceptance, when the character wore a revealing dress and
stared at the users, or a conservative outfit while looking at no place in particular, participants showed
more irrational beliefs about sexual violence (Fox & Bailenson, 2009). This could suggest that when
female characters match specific stereotypes, like “sexually aggressive” or “shy and pure”, they
become more effective at stimulating sexist ideas. This might have made statements like “women
wearing short skirts are asking for a sexual assault” or “females are weak and need male protection”
easier to access.
5 Yao, et al. 2010 asked participants to play a game were female characters appeared in revealing
dresses but also nude and having sex. This could have made the players sexually aroused, explaining
their bias towards sex-related information. Dill, et al. 2008, on the other hand, presented people with
static images of both male and female sexualized game characters. So the stereotypical depictions
of men could have also been what made sexist ideas easier to access.
6 Given that word recognition tasks are frequently used to test Priming effects (a phenomena
explained by Spreading Activation), we decided to interpret the differences in the capacity to detect
objectifying terms as a result of this principle. Congruently, we chose to present the effect on the self-
reported tendency to take sexual advantage of women (a variable more related to behaviors) as an

example of Modeling. This and the rest of our interpretations respond to a desire to make sense of
the limited evidence at hand and shouldn’t be considered as the only possible explanation.
7 People exposed to live gameplay of GTA IV were the protagonist gets a lap dance in a strip club,
picks up a prostitute on the street, has sex with her, shoots her to get back his money and then
escapes the police, didn’t report more irrational beliefs about sexual assaults that those presented
with live gameplay of a baseball video game.
8 Additionally, while the Leisure Suit Larry experiment only involved men, the GTA study included
participants from both sexes, making them less likely to interpret seeing attractive women nude and
having sex as a reward.
9 People who played a level of Tomb Raider: Legend, were the avatar had bigger breasts, a smaller
waist and wore a more revealing outfit, didn’t show more negative attitudes towards the appearance,
occupations or physical abilities of women, than those who played a stage were the protagonist had
an outfit that covered most of her skin.
10 Women who wore a virtual reality headset and a saw mirror were a female character in a revealing
dress replicated their head movements in real time, reported more thoughts about their bodies than
those exposed to an avatar with clothes that covered most of her skin.
11 Among the participants in the sexualized condition, watching a character that had a face that
resembled the user´s was followed by higher acceptance of myths about rape, than those presented
with an avatar who had the face of an unknown woman.
12 It should be noted that the enemies were also actively trying to beat or shoot the player’s avatar to
death.
13 For men playing a first person shooter against sexualized female enemies, higher levels of
immersion lead to more hostile sexism. This didn't happen when the enemies were non-sexualized
females or males. For another group of men playing Mortal Kombat: Komplete Edition (a fighting
game which includes clear depictions of body damage, close ups to organs and bones breaking, as
well as gory finales), when the enemies were sexualized women, higher levels of immersion were
accompanied by lower empathy towards rape victims. This didn't happen when the enemies were
strong males.
14 The designs employed in the reviewed experiments aren’t free of limitations, but a clear progress
can be observed. Dill, for example, presented users with static images of characters from commercial
video games. A stimulus that doesn’t represent the fluid audiovisual nature of the medium studied.
Beck, on the other hand, chose to present live footage of a game being played live. This guaranteed
that every participant in a condition was going to be exposed to the exact same content. But it also
meant that they would not have any control over the content (where the character would go or what
he might do), which is an essential component of video games. Yao solved this problem by allowing
individuals to play one of different titles. The only problem was that this introduced other differences
between the experience of each group (like the graphics or gameplay of each game), which, although
unexpectedly, could have contributed to the differences between measures after the experiment. Fox
avoided this limitation by exposing each group to the exact same experience, except for one detail:
The clothes of the female character. This meant that any difference between the groups could be
attributed just to this form of sexist content. Nevertheless, the use of a virtual reality headset and a
particularly unique interactive setting makes it difficult to generalize their results to most commercial
games. Behm-Morawitz design used a different but clever approach. She took advantage of the
options built into an existing game to present participants with more or less sexist versions of the
same title. However, this required users to play different levels, adding differences between the
experiences of each group. Finally, LaCroix applied a mix of this and Fox’s approach. She used game
creating software to change the clothes and gender of the enemies in a First Person Shooter. This
allowed her to present participants with content that was similar to commercial games, while at the
same time having the presence or absence of sexist factors as the only difference among groups. As
a result, it was easy to attribute any post-test differences between the conditions to the sexist variable,
and these findings could be generalized to other existing first person shooters.
15 Additionally, when the type of stimuli changed, so did the variables that made it more or less
influential. Sexist behaviors, for example, were more contagious when they were easy to replicate.
On the other hand, a strong level of identification was necessary before embodying a gender
stereotype was accompanied by more sexist beliefs.

16 One of the surveys did obtain some longitudinal information. Breuer et al. (2015) found no change
among the sexist attitudes of active video game players, two years after their initial survey. This
doesn’t completely rule out long term influences though. Different forms or times of exposure could
lead to different results.