Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 1, 2014
arrowPress Releases
October 1, 2014
PR Newswire
View All

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

To feel like the good or bad guy: The role of empathy
by Wai Yen Tang on 07/11/14 11:59:00 am   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.


Matthew Grizzard (University of Buffalo) used “University Press Release” for his videogame study on moral sensitivity that was published in Cyberpsychology, Behavior, and Social Networking. It was super effective, many news media picked up the study. In response, I am blogging on Christian Happ (University of Trier), André Melzer (University of Luxembourg) and Georges Steffgen’s (University of Luxembourg) study on the role of empathy in antisocial and prosocial gaming in Psychology of Popular Media Psychology.


Evidence suggests that violent media influence users’ cognitions, affect, and behavior in a negative way, whereas prosocial media have been shown to increase the probability of prosocial behavior. In the present study, it was tested whether empathy moderates these media effects. In two experiments (N 80 each), inducing empathy by means of a text (Study 1) or a video clip (Study 2) before playing a video game caused differential effects on cognitions and behavior depending on the nature of the subsequent video game: The induction had positive effects on participants’ behavior (i.e., decreasing antisocial and increasing prosocial behavior) after a prosocial game (Study 1), or when participants played a positive hero character in an antisocial game (Study 2). In contrast, empathy increased antisocial behavior and reduced prosocial behavior after playing a mean character in an antisocial game (Study 1 and 2). These findings call attention to the differential effects of empathy depending on game type and game character, thereby questioning the unconditional positive reputation of empathy in the context of video game research.

This post is crossposted at my blog, VG Researcher.

We play videogames to have fun, but that is one of many motivations. We also play because the narrative and the characters within it drives us to continue into its conclusion. It is a reason why we remember and celebrate well-known videogame characters, such as Mario, the Master Chief, Solid Snake, Aerith Gainsborough among others. We are empathetic to many characters, be it the protagonists or antagonists, we are concerned about them that we feel what they felt. Sometimes you might empathize with the characters because others clue you into it, happens when your friends talk how much they feel a character’s emotion. Of course, not all characters overcome the challenges only through antisocial acts, some opted through prosocial acts, or have the option to choose either (via the players’ decisions).

In the research literature, studies found that playing a videogames through antisocial acts, such as violence and criminal acts, increases antisocial behaviours, including aggression. In contrast, playing prosocially in-game, such as helping or supporting other characters, increases prosocial behaviours. The authors noted that empathy was just recently examined in videogames research. Empathy is defined as the ability to understand and share in another’s emotional state or context. The gist is that empathy increases prosocial behaviours, a person with a high level of empathy can sympathize with a victim of violence or feel guilty when committing unjustified acts of violence in a videogame as Tilo Hartmann and colleagues (2010) have found. A question remains is how empathy plays a role when we play as an antisocial versus prosocial videogame character?

Thus, the authors conducted two experiments to examine how playing a prosocial versus antisocial videogame might affect prosocial behaviours and how empathy acts as a potential moderator between them.

Study 1

In this experiment, they examined whether empathy would affect the prosocial effects of prosocial play and whether empathy compensate the antisocial effects of antisocial play.


Participants: 80 undergraduate students, average age is 23.4 years. 55% are women. The average videogame use is 0.9 on a 4 point-scale, a sample inexperienced with videogames.


Mood: The Positive and Negative Affective Scale was used to assess participants’ current mood. It is a 20-item answered on a 5-point agreement scale. This was assessed twice, before and after game play.

Trait Empathy: The Interpersonal Reactivity Index is a popular measure for trait empathy. They used a 12-item version of the index and it is answered on a 5-point scale.

Trait Aggression: The Aggression Questionnaire is a popular measure for trait aggression. They used two subscales of the questionnaire, anger and physical aggression, thus it is a 14-item version and answered on a 5-point scale.

Videogame ratings: Participants rated the videogame they were assigned to play, answered on a 4-point agreement scale.

Videogames used: The antisocial videogame is Manhunt 2 and the prosocial videogame is Trauma Center: New Blood, both are Wii games.


Participants are tested individually and randomly assigned to one of four conditions. The first experimental condition is the induction of empathy. This is achieved by having participants read a newspaper article about videogames positive effects on memory attributed by players’ emotional involvement and empathy. In the non-empathy condition, they were not given a newspaper article to read. The second experimental condition is the videogame, where participants either play the murderous psychopathic Daniel Lamb of Manhunt 2 or the healing touch of Drs. Markus Vaughn or Valerie Blaylock of Trauma Center: New Blood, the article did not disclosed which doctors participants played. These games were picked because both displayed blood and wounds, but differ in terms of goals. Looking over the screenshots of both games, I’m not sure if the presentation of blood is equivalent as I assume that Manhunt 2 would depict blood and gore pretty stylistically whereas Trauma Center would have much less of a focus.

The participants were given a 5 training session, and then play the videogame for 15 minutes. Afterwards, they completed the PANAS a second time. At this point, participants were told they have completed the experiment and the research assistant they can help themselves to take ONE item (pencil or piece of chocolate) as a reward for their participation. They were explicitly told not to take more than one item because there won’t be enough for other participants. The number of items taken more than one would be counted as antisocial behaviour. Participants were also given an envelope that contained a questionnaire for a different study, they were told that it was optional, they can complete and send it back within a week. The return of a completed questionnaire would count as prosocial behaviour.


They examined whether third variables, such as trait aggression, trait empathy, frustration among others, might influence the experimental results, they found none of the variables having any significant effect.

The main effects of videogame type is significant for both antisocial and prosocial behaviours. Participants who played the antisocial videogame stole more items (M= 1.29, SD=1.42) than those who played the prosocial videogame (M=0.77, SD= 0.96) and who were 4.11 times were more likely to complete and return the optional questionnaire.

The authors examined the role of empathy and found that there was no significant effect on antisocial behaviour. However, empathy had some interaction effects for prosocial behaviour. Those who read the empathy article acted differently depending on what game they played. 37% of those who played the prosocial game returned the questionnaire in contrast to the 4.5% of those who played the antisocial game. In comparison to other conditions, those who read the empathy text and played the antisocial videogame were 7.32 less likely to act prosocially.

The authors speculated that raising participants to empathize with videogame characters might have identified with and taken on some of the characters’ attributes: Acting mean just like Daniel Lamb.

Study 2

The authors conducted a second experiment to further examine why playing an antisocial videogame when induce to empathize led to greater antisocial behaviours.


Participants: 80 undergraduate students, average age is 21.8, 69% are women. The average videogame use is 2.1 on a 4-point scale. A sample of fairly inexperienced with videogames.


The same measures used in study 1 with some additions.

Two items regarding participants’ pity and affective concerns towards the opponent character.

Antisocial behavioural intentions: participants read two hypothetical scenarios that are ambiguously aggressive (e.g. someone spilled their drink on you). Participants were asked how their angry their reactions will be on a 4-point scale.

Prosocial behaviours: Researchers told the participants that there was a donation box outside the lab and they were free to donate the money anonymously. However, the donated amount was not normally distributed (i.e. a lot of people donated the same amount), instead they checked whether they donated or not.

Empathy induction: a 2 minute clip from Street Fighter: The Legend of Chun-Li where she witnessed her father being kidnapped as the empathy induction video because it’s from a videogame. Although, they could have used the movie version of Max Payne, but the age of the games might be a potential hindering factor. The non-empathy induction video is 2-minute clip from The Last Emperor.

Videogame used: Streetfighter 4 on the PS3 because Chun Li.

The procedure is the same from study 1, except that the participant either played as Chun-Li who was the victim in the movie or the perpetrator (I have not seen the movie, but it could be Bison, Vega or Balrog). Playtime is 15 minutes.


The authors found effects among participants who watched the empathy video. Those who played the perpetrator reported greater affective concern and pity towards fighting Chun-Li. They found that playing as Chun-Li who was the victim led to less antisocial behavioural intentions whereas the reverse is true when playing as the perpetrator. Regarding prosocial behaviours, there was statistical trend in that 90% of participants who watched the empathy video and played Chun-Li donated money in contrast to all other groups whose percentage average is 71.7%, in essence the former group were 1.26 times more likely to donate.


The take home message is that empathy with videogame playable characters would lead to different behavioural outcomes depending on the characters’ role and in-game behaviours. Empathizing with characters who behaved prosocially or were victim of antisocial behaviours led to greater prosocial behaviours. In contrast, the same empathizing with characters who behaved antisocially led to greater antisocial behaviours. Curiously, but not surprising from prior studies, playing as the perpetrator knowing what they did led to greater emotional concerns with victims.

The authors observed that these results touched on the complexities of morality. Interpreting study 2′s results, participants who played Chun-Li could transfer their feelings felt from the movie Chun-Li since they are the same person. Thus, when fighting for revenge and justice, it becomes morally justified which is rather more satisfying than compensating the victim. In regards to why participants who played as the perpetrator behaving more antisocially and yet felt more pity and emotional concerned with the victim, this may be because of cognitive dissonance. The authors cautioned, however, cognitive dissonance may not be a good explanation due to differences between their experiments. Study 1 can be clearly attributed to cognitive dissonance as those who played Daniel Lamb lacked any justification for aggression, yet those in study 2 participants who played the perpetrator were justified as simply continuing to act violently.

The authors suggested some future research directions. The experiments showed the role of empathy having a differential effect on prosocial and antisocial behaviours, however they did not reveal the underlying process of why. They suggested several psychological mechanisms: cognitive accessibility to prosocial or antisocial thoughts, personality traits of personal distress which might motivate people to be more prosocially engaged, identification with videogame characters. The authors suggested using the “Katie Banks Task” which involves a real person for whom participants can feel empathy.

I don’t have much critiques of the study, but I should cover some related topics to the study. The participants were role taking the characters so that means they go along with what the characters do and think, no moral input from the players. Studies concerning players’ moral choices are covered mostly by Nicholas Bowman (West Virginia University). I don’t have any studies concerning the anti-hero, the closest are studies that investigated bad behaviours done by good guys [Link to my library].


Happ, C., Melzer, A., & Steffgen, G. (2014). Like the good or bad Guy—Empathy in antisocial and prosocial games. Psychology of Popular Media Culture. DOI:10.1037/ppm0000021
Hartmann, T., Toz, E., & Brandon, M. (2010). Just a game? unjustified virtual violence produces guilt in empathetic players. Media Psychology, 13 (4), 339-363. DOI:10.1080/15213269.2010.524912

Related Jobs

Bohemia Interactive Simulations
Bohemia Interactive Simulations — ORLANDO, Florida, United States

Game Designer
Bohemia Interactive Simulations
Bohemia Interactive Simulations — Prague 5, Czech Republic

Game Designer
Raven Software / Activision
Raven Software / Activision — Madison, Wisconsin, United States

UI Lead - Raven
DoubleDown Interactive
DoubleDown Interactive — Seattle, Washington, United States

Game Designer