Content uploaded by Bruce D Bartholow
Author content
All content in this area was uploaded by Bruce D Bartholow on Dec 11, 2014
Content may be subject to copyright.
Consensus on Media Violence Effects
1
Consensus on Media Violence Effects: Comment on Bushman, Gollwitzer, and Cruz
Craig A. Anderson
Luca Andrighetto, University of Genova, Italy luca.andrighetto@unige.it
Bruce D. Bartholow, University of Missouri, U.S. bartholowb@missouri.edu
Laurent Begue, University of Grenoble, France lbegue@msh-alpes.fr
Paul Boxer, Rutgers University, U.S. pboxer@psychology.rutgers.edu
Jeanne Funk Brockmyer, University of Toledo, U.S. jeanne.brockmyer@gmail.com
Melinda C. R. Burgess, Southwestern Oklahoma State University, U.S.
melinda.burgess@swosu.edu
Esther Calvete, Universidad de Deusto, Spain esther.calvete@deusto.es
Joanne Cantor, University of Wisconsin, U.S. jrcantor@wisc.edu
Sarah M. Coyne, Brigham Young University, U.S. smcoyne@byu.edu
Karen Dill-Shackleford, The Fielding Institute, U.S. kshack@fielding.edu
Ed Donnerstein, University of Arizona, U.S. edonners@email.arizona.edu
Alessandro Gabbiadini, University of Milano Bicocca, Italy ale.gabbiadini@gmail.com
Bryan Gibson, Central Michigan University, U.S. gibso1b@cmich.edu
Youssef Hasan, Qatar University, Qatar youssef.hasan@qu.edu.qa
Adam K. Lueke, Central Michigan University, U.S. lueke1a@cmich.edu
Izaskun Orue, Universidad de Deusto, Spain izaskun.orue@deusto.es
Paolo Riva, University of Milano-Bicocca, Italy paolo.riva1@unimib.it
Victor C. Strasburger, University of New Mexico, U.S. VStrasburger@salud.unm.edu
Chiara Volpato, University of Milano-Bicocca, Italy chiara.volpato@unimib.it
Wayne Warburton, Macquarie University, Australia wayne.warburton@mq.edu.au
Author notes: Authorship is alphabetical. Address correspondence to Craig A. Anderson,
Department of Psychology, Iowa State University, Ames, IA 50011-3180 email: caa@iastate.edu
Keywords: Media violence; denialism; aggression; video game violence
Title page including all author information
Consensus on Media Violence Effects
1
Consensus on Media Violence Effects: Comment on Bushman, Gollwitzer, and Cruz
Keywords: Media violence; denialism; aggression; video game violence
Masked Manuscript without author information
Consensus on Media Violence Effects
2
Consensus on Media Violence Effects: Comment on Bushman et al.
Abstract
We summarize the main findings of Bushman, Gollwitzer, and Cruz (this issue), highlights its
empirical contributions, and notes interesting patterns and implications for future research. We
also introduces the denialism perspective to the consensus among true media violence scientists
and pediatricians versus claims of nonconsensus by those highly motivated to deny evidence of
harm. The results demonstrate the invalidity of the common denialist claim that no consensus
exists among experts on the reality of harmful media violence effects on children and adolescents.
We note the likely bias against finding consensus of harm among the communication and media
samples, given the prevalence of non-experts among their memberships. This article also presents
a new breakdown of the Bushman et al. findings, highlighting the high consensus for causal
screen media violence effects on aggression, which fairly closely mirrors findings from that
voluminous research literature, and compares this to the lack of consensus on the harmful effects
of print media violence, which corresponds to a quite small research literature. We conclude with
brief discussions of factors underlying denialism, research domains relevant to the study of
denialism, implications of widespread gamer denialism among children and adolescents for
conducting research on media violence effects, and a call for research on denialism and on
remedies for it in a world that needs accurate beliefs to effectively deal with harmful products
such as screen violence, greenhouse gases, and viruses.
Consensus on Media Violence Effects
3
Bushman, Gollwitzer, and Cruz (this issue) report a very interesting survey study of the
beliefs held by several key groups who are (or at least ought to be) concerned about potentially
harmful effects of violent media on children. Their well-conducted survey provides convincing
evidence that there is considerable consensus among members of media and communication
societies, pediatricians, and even parents that exposure to media violence increases hurtful
behavior by children. These consensus beliefs mirror the largely consistent research findings of
actual negative effects of violent media on thoughts, feelings and behaviors (e.g., Anderson et al.,
2003, 2010; Anderson & Bushman, 2002; Bushman & Huesmann, 2006; Greitemeyer, & Mügge,
2014).
1
This commentary highlights the importance of these findings, notes why the media and
communication samples may be biased against believing in harmful media effects (relative to a
sample of true media violence experts), and discusses several issues involving vocal critics of
mainstream research findings.
Much like other well-known cases in which powerful profitable industries have waged
disinformation campaigns against specific scientists and general fields of scientists whose
research suggests that their products cause harm, the television, film, and video game industries
and their apologists spend considerable time, effort, and money sowing the seeds of doubt about
the science in this area.
2
As so aptly noted by Nijhuis (2008), the industries (and their apologists)
don't have to prove anything in order to win; all they have to do is sow the seeds of doubt. They
"win" if enough doubt is sown to convince the public and public policy makers "to reject the case
for taking action to tackle threats to health." (Diethelm & McKee, 2009, p. 2). In short, if they can
prevent a strong consensus from emerging, the denialists win.
Denialism
Consensus on Media Violence Effects
4
Denialism, the act of denying factual information, comes in many forms, arises from a
range of motivations, and is supported by many psychological processes. Some of the most
blatant historical cases do not directly involve science, but instead deny historical events, such as
the Holocaust. But many cases do involve science "facts." We put the word "facts" in quotes to
highlight one reason why denialism is so successful when its tactics are used in scientific domains.
Specifically, scientists are trained to not use such absolute words, because scientific conclusions
should be seen as subject to change if the data and the scientific field changes. That is, the good
scientist role includes being extremely cautious about drawing firm conclusions, and teaches us to
avoid the F-word ("fact"). We are supposed to be skeptical, and perpetually open to criticism and
alternative perspectives.
Of course, nonscientists (including the general public, journalists, politicians…) don't
understand this reluctance of scientists to use the F-word or its scientific cousin "cause" (the C-
word). Denialism plays on this reluctance in numerous ways, not the least of which is to suggest a
lack of consensus among scientists about whether a particular product causes harm, and to couch
their criticisms of the consensus science (and scientists) in terms of normal scientific skepticism
(Anderson & Gentile, 2008; Anderson, Gentile & Buckley, 2007; Jack, 2011). The tactics used by
denialists are well known, and according to Diethelm and McKee (2009) include: (1)
identification of conspiracies; (2) use of fake experts; (3) selectivity; (4) creation of impossible
expectations of what research can do; and (5) use of misrepresentation and of various logical
fallacies. It is outside the scope of the present comment to identify specific instances of use of
each of these tactics by media violence denialists; we nonetheless invite readers to keep these
tactics in mind as they read articles and news reports about the media violence "debate." We bring
Consensus on Media Violence Effects
5
up the issue of denialism in this comment because the Bushman et al. article identifies and
debunks one denialistic tactic.
Bushman et al. Findings
In Figure 1, we highlight the primary screen media (television, movies, video games) and
print media (comic books, literature) results to facilitate our discussion. One notable aspect of
Figure 1 is the high degree of consensus that screen media violence is a causal risk factor for
aggressive behavior. This result shows that the screen media industries' and their apologists'
claims of a lack of consensus are greatly overstated. There is considerable consensus among
members of media and communication societies, pediatricians, and parents. The whopping
differences between the Causal and the Not Causal columns practically leap from the page. In
short, these results debunk one use of the denialist tactic of misrepresentation.
Also obvious from a quick look at Figure 1 is that there is relatively little consensus about
print media violence effects on aggressive behavior. This comparison of screen and print media
consensus mirrors the research literature in at least one interesting way. Specifically, the research
literature on screen media violence effects is much larger and more compelling than the literature
on print media violence effects. We are not saying that print media violence effects don't exist
(e.g., Bushman, Ridge, Das, Key, & Busath, 2007; Coyne, Ridge, Stevens, Callister, & Stockdale,
2012). But, there is much less research on print media effects, and there is little (if any) evidence
of long term effects of print media violence on aggression; relevant longitudinal and cross-
sectional studies are virtually absent. Furthermore, there are theoretical reasons to believe that on
average, print media violence effects are likely to be considerably weaker. One needs only to
consider the amount and vividness of violence encountered per hour while reading the Lord of the
Rings trilogy versus watching the associated movies, versus playing the associated video games,
Consensus on Media Violence Effects
6
and consider the context (or lack thereof) between typical screen versus print media violence, to
get an idea of the vast differences between the psychological processes engaged by different
media types. And, consider the reading skills (and therefore age) required to read and enjoy that
book series, versus the very young age at which one can comprehend and participate in the
violence displayed in the movies and video games.
An additional point of interest concerning screen media violence consensus arises from a
close inspection of differences between the four samples. Specifically, there is less consensus
among some of the sampled groups than there logically should be. Logically, people who aren't
themselves true experts in a scientific domain should base their beliefs about that domain on the
statements made by the true experts. For example, our beliefs about the reality of global warming
are based on the statements of panels of global climate scientists. Somewhat closer to home, even
though all of the present authors have expertise in the broad domain of human aggression, none of
us are experts on the behavioral effects of testosterone on aggression in mice. Therefore, we must
rely on true experts in that domain for any beliefs we might hold about testosterone and mouse
aggression. It would be foolish and silly for us to denigrate the findings of the mouse-
testosterone-aggression experts because our personal experiences with mice seem different; it
would be unethical for us to proclaim to the world that our views of mouse testosterone-
aggression findings were as valid as those who actually study the phenomenon simply because we
have done research on the broad topic of aggression.
People who are not true experts on media violence research (see Anderson & Gentile,
2008, p. 287, for suggested criteria) should either have no firm opinion about the issue or should
rely on reports by the true experts to form their opinion. And, every major scientific expert panel
that has reviewed the research on screen media violence effects has come to the same conclusion,
Consensus on Media Violence Effects
7
that media violence (usually meaning screen violence) is a causal risk factor for increased
aggressive behavior. This includes expert panels created by the American Academy of Pediatrics,
the American Psychological Association, the American Academy of Child & Adolescent
Psychiatry, the American Medical Association, the American Academy of Family Physicians, the
American Psychiatric Association, the U.S. Surgeon General, the International Society for
Research on Aggression, the U. S. National Institutes of Health, and most recently the Society for
the Psychological Study of Social Issues (SPSSI, 2014), and among others. Scientific panels in
other countries have reached the same conclusion. Nonexperts should either admit (to themselves
and others) that they really don't have a belief about media violence effects (if they are unaware of
what the true experts have concluded), or should adopt the position of the true experts.
Interestingly, there is one notable group missing from the Bushman et al. consensus study:
the small group of true scientific experts on media violence. It seems safe to assume that this
group, were it to be identified and sampled, would yield even higher levels of consensus, given
that subsamples of such true experts have repeatedly written reports finding such evidence (e.g.,
the U.S. Surgeon General's panel that published the Anderson et al. 2003 report).
So, why wasn't consensus about the causal effects of screen violence even higher in
Bushman et al.'s study? Of particular interest (and some concern) is the fact that members of the
sampled media and communication societies didn't show substantially greater accuracy about
screen violence effects than did the parents. There are likely several factors involved in the
differences among the four groups. The very high consensus found in the pediatrician sample may
result from the fact that the American Academy of Pediatrics has done a good job of
communicating the true experts' findings to their members, something that neither of the other
Consensus on Media Violence Effects
8
sampled societies (media, communication) has done. And of course, pediatricians see large
numbers of children, many of whom have behavioral problems that the parents discuss with them.
The other two sampled societies both have substantial proportions of members who are
neither behavioral science experts nor researchers in media violence, but rather are people
interested in learning about effective ways of using media in the real world. Some do not have a
research doctorate in an appropriate behavioral science, some are employed by media industries,
and some of them could reasonably be characterized as media violence denialists. In other words,
we cannot assume that all members of these two societies are "researchers" or experts in media
violence effects research. Indeed, many of the top media violence experts are not members of
these societies. To be sure, many members are true experts, but many are not. Many likely are
strong supporters of freedom of speech rights as embodied in the U.S. First Amendment (as are
we), many are great fans and consumers of violent media (as are some of us, including the first
author), and some may feel threatened by the possibility of admitting the harmful effects of
violent media (see Laurin, Kay, & Fitzsimons, 2012; Nauroth, Gollwitzer, Bender, & Rothmund,
2014).
3
It would be interesting to see what level of consensus would emerge from membership of
other organizations that vary in terms of their relevant expertise and their own expert panel
statements concerning media violence effects, such as the International Society for Research on
Aggression (ISRA) and the Society for the Psychological Study of Social Issues (SPSSI).
The Bushman et al. survey did not include an option allowing respondents to indicate that
they either had no opinion or did not believe that they had sufficient knowledge to have an
opinion. A substantial portion of those who checked the "neither agree nor disagree" option may
have been of this type. This is not necessarily a weakness of the study for its intended purpose,
but would be a question of interest for future research.
Consensus on Media Violence Effects
9
Another point of interest is that the somewhat lower level of screen media consensus
reported by parents (relative to pediatricians) can be viewed as the proverbial glass that is either
half full or half empty. The pessimistic view is that it demonstrates just how effective media
violence denialism has been; despite the consensus among true experts in this domain over
several decades, despite public statements by various expert panels (again, over several decades),
and despite work by AAP and other parent/child education and advocacy groups, many parents
are still ignorant about the true facts of screen media violence effects. This ignorance also helps
explain why so few parents take an active role in regulating their children's use of violent media,
which has been documented in many studies.
The optimistic view of the parent results is that despite the major efforts of media violence
denialists, and despite the failure of news journalism in general to accurately portray the state of
the science (Bushman & Anderson, 2001; Martins, Weaver, Yeshua-Katz, Lewis, Tyree & Jensen,
2013), most parents do have the factually correct belief that screen violence is a causal risk factor
for aggression. The extent to which that belief comes from personal observation of their own
children; from news reports they have read; from the educational efforts of pediatricians,
parent/child support and advocacy groups; from schools; or from other sources is unknown, but
would be worth further study.
Future Studies
In the history of research on the smoking/lung cancer link, one fascinating finding was
that the first group of physicians to quit smoking was thoracic surgeons, those who most directly
saw the ravages of smoking on the lungs. It would be interesting to know whether a similar
phenomenon is occurring among psychologists. Specifically, which groups of psychologists are
most likely to closely monitor and control their children's exposure to media violence? Do
Consensus on Media Violence Effects
10
members of societies with greater expertise in media violence (e.g., ISRA) or greater interest in
applying psychological science to real world issues (e.g., SPSSI) show greater or lesser consensus
on media violence effects than the groups studied by Bushman et al.?
Another set of important questions in need of research concerns the extent to which self-
image, self- identification, or self-involvement with media violence drives denial of the scientific
findings. Nauroth et al. (2014) showed that gamers feel stigmatized by and are angry about
research findings that demonstrate negative effects of violent games. Bender, Rothmund, and
Gollwitzer (2013) demonstrated empirically that as research participants, gamers will sabotage
studies of violent video game effects on aggression, and do so even when a cover story is present.
In general, several large research domains (e.g., attitudes, decision under uncertainty,
motivated cognition, self-identity) are relevant to questions about how people deal with
information that is discrepant with prior beliefs or important values. Generally, they show that
people will go to great lengths to defend importantly held beliefs and values, including selective
searches for information that supports their position, engaging in biased information processing
and perception, and selectively attending to and remembering biased information. Classic studies
of this type includes Lord's work on capital punishment beliefs (Lord, Ross, & Lepper, 1979).
Similarly, Anderson's work on social theory formation, perseverance, and change found that even
trivial beliefs formed on the basis of weak or even hypothetical data can survive logically
compelling challenges (e.g., Anderson, & Lindsay, 1998; Anderson & Sechler, 1986).
Basically, we would expect that people who strongly identify with violent games—e.g.,
gamers, producers or sellers of games— are most likely to deny any harmful effects, because such
effects threaten either the self ("I've played violent games all my life, I'm not an aggressive person,
so your claim of harmful effects can't be true"), some important self-related aspect of one's life ("I
Consensus on Media Violence Effects
11
sell video games to children, I'm a loving parent and a good citizen, so your claim of harmful
effects can't be true"), or even their job. As research participants, such people also are the most
likely to intentionally sabotage studies in which they believe the link between violent games and
aggression is being assessed, a sort of "reverse" demand characteristics effect (Bender,
Rothmund, & Gollwitzer, 2013). Sabotage can easily be done in most (but not all) studies, by
intentionally behaving very non-aggressively in standard laboratory aggression paradigms, by
reporting low levels of past aggression in survey studies, or by under-reporting one's own amount
of exposure to violent media.
There are numerous theoretical reasons for this denial and these behavioral reactions,
including cognitive dissonance, self-esteem maintenance, and other motivated cognition processes.
Indeed, the fact that a very few researchers consistently fail to replicate well-established findings
may be the result of their using research methods that fail to adequately disguise the violent
media/aggression aspect of their studies (see the comparison of different research groups by
Greitemeyer, & Mügge, 2014). All it would take would be a revealing study name on a sign-up
sheet, a weak cover story on a consent form or in the instructions, or even the reputation of the lab
as being one that conducts research on media and aggression.
Even a cursory inspection of gaming sites reveals that even children and adolescents are
well aware of the media violence and aggression issue. On one hand, this makes conducting
media violence research in the modern era much more difficult than in past eras. It also increases
the need for researchers to: (a) more fully disclose their study names (on sign up sheets and/or
consent forms), recruitment procedures, cover stories, and instructions; and (b) more carefully
assess and report participant suspicion. On the other hand, this also provides an opportunity to
investigate denial and perseverance processes in the context of highly motivated beliefs and
Consensus on Media Violence Effects
12
values, as interesting research topics in their own right. Similar research can (and should) be done
on global warming denialism (for instance), and on discovering procedures that reputable
scientific and public policy groups could use to help the general public to accurately understand
scientific facts that are of relevance to them personally and to the welfare of larger society (e.g.,
that vaccinations do not cause autism). Continuing to wallow in artificially created doubt is
nonproductive, regardless of whether that doubt is created by a profitable business or by more
individually motivated denialists.
Consensus on Media Violence Effects
13
References
Anderson, C.A., Berkowitz, L., Donnerstein, E., Huesmann, L.R., Johnson, J., Linz, D.,
Malamuth, N., & Wartella, E. (2003). The influence of media violence on youth.
Psychological Science in the Public Interest, 4, 81-110.
Anderson, C.A., & Bushman, B.J. (2002). The effects of media violence on society. Science, 295,
2377-2378.
Anderson, C.A., Gentile, D.A., & Buckley, K.E. (2007). Violent Video Game Effects on Children
and Adolescents: Theory, Research, and Public Policy. New York: Oxford University Press.
Anderson, C.A., & Gentile, D.A. (2008). Media violence, aggression, and public policy. In E.
Borgida & S. Fiske (Eds.), Beyond Common Sense: Psychological Science in the Courtroom
(pp. 281-300). Malden, MA: Blackwell.
Anderson, C.A., & Lindsay, J. J. (1998). The development, perseverance, and change of naive
theories. Social Cognition, 16, 8-30.
Anderson, C.A., & Sechler, E.S. (1986). Effects of explanation and counterexplanation on the
development and use of social theories. Journal of Personality and Social Psychology, 50,
24-34.
Anderson, C. A., Shibuya, A., Ihori, N., Swing, E. L., Bushman, B.J., Sakamoto, A., Rothstein,
H.R., & Saleem, M. (2010). Violent video game effects on aggression, empathy, and
prosocial behavior in Eastern and Western countries. Psychological Bulletin, 136, 151-173.
Bender, J., Rothmund, T., & Gollwitzer, M. (2013). Biased estimation of violent video game
effects on aggression: Contributing factors and boundary conditions. Societies, 3, 383-398.
doi:10.3390/soc3040383
Consensus on Media Violence Effects
14
Bushman, B. J., Gollwitzer, M., & Cruz, C. (in press). There is broad consensus: Media
researchers agree that violent media increase aggression in children, and pediatricians and
parents concur. Psychology of Popular Media Culture,
Bushman, B.J., & Huesmann, L.R. (2006). Short-term and long-term effects of violent media on
aggression in children and adults. Archives of Pediatric & Adolescent Medicine, 160, 348-
352.
Bushman, B. J., Ridge, R. D., Das, E., Key, C. W., & Busath, G. L. (2007). When God sanctions
killing; Effect of scriptural violence on aggression. Psychological Science, 18(3), 204-207.
DOI: 10.1111/j.1467-9280.2007.01873.x
Coyne, S. M., Ridge, R., Stevens, M., Callister, M., & Stockdale, L. (2012). Backbiting and
bloodshed in books: Short term effects of reading physical and relational aggression in
literature. British Journal of Social Psychology, 51, 188-196.
Diethelm, P., & McKee, M., (2009). Denialism: What is it and how should scientists respond?
European Journal of Public Health, 19, 2-4.
Greitemeyer, T., & Mügge, D. O. (2014). Video games do affect social outcomes: A meta-
analytic review of the effects of violent and prosocial video game play. Personality and
Social Psychology Bulletin, 40, 578 –589.
Jack, A. (2011). Reclaiming skepticism: Scientists fight back against climate change deniers.
Slate, June 26, 2011. Downloaded on 6/26/11 from: http://www.slate.com/id/2297599.
Laurin, K., Kay, A. C., & Fitzsimons, G. J. (2012). Reactance versus rationalization: Divergent
responses to policies that constrain freedom. Psychological Science, 23, 205-209.
Consensus on Media Violence Effects
15
Lord, C. G., Ross, L. & Lepper, M. R. (1979). Biased assimilation and attitude polarization: The
effects of prior theories on subsequently considered evidence. Journal of Personality &
Social Psychology, 37, 2098-2109.
Martins, N., Weaver, A. J., Yeshua-Katz, D., Lewis, N. H., Tyree, N. E., & Jensen, J. D. (2013).
A content analysis of print news coverage of media violence and aggression research.
Journal of Communication, 63, 1070-1087. doi: 10.1111/jcom.12052
Nauroth, P., Gollwitzer, M., Bender, J. & Rothmund, T. (2014). Gamers against science: The case
of the violent video games debate. European Journal of Social Psychology, 44(2), 104-116.
doi: 10.1002/ejsp.1998
Nijhuis, M. (2008). The doubt makers. Miller-McCune, 1(2), 26-35.
SPSSI, (2014). SPSSI Research Summary on Media Violence. Report released in April, 2014.
Downloaded on April 4, 2014, from:
http://www.spssi.org/index.cfm?fuseaction=page.viewPage&pageID=1899&nodeID=1
Consensus on Media Violence Effects
16
Figure 1. Beliefs about the causal effects of screen (television, movies, & video games) and print
media (comic books & literature) violence on aggressive behavior by members of a scholarly
media society, a scholarly communication society, pediatricians, and parents. Note: Not
Causal is the combined total of strongly disagree and disagree categories, Don't Know is the
neither agree nor disagree category, and Causal is the combined total of strongly agree and
agree categories. Data Source: Bushman, B. J., Gollwitzer, M., & Cruz, C. (this issue).
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Not
Causal
Don't
Know
Causal Not
Causal
Don't
Know
Causal
Screen Media Print Media
Beliefs-Media Violence Effects on Aggression
Media Society
Communication
Society
Pediatricians
Parents
Consensus on Media Violence Effects
17
Footnotes
1
Only comprehensive reviews are included among the examples. There are additional instances
of highly selective (and frequently biased) reviews, but they are less relevant because the
comprehensive reviews cited here include considerably more relevant studies. There also are
many excellent older comprehensive reviews; see Anderson et al. (2003) for citations to many of
them.
2
Scientific denialism examples include tobacco effects on cancer and heart disease, asbestos
effects on cancer, mercury poisoning, lead poisoning, Dioxin poisoning, acid rain, evolution,
global warming, HIV as a cause of AIDS, false claims about vaccines and autism, and real
findings on football/brain injury effects.
3
Of course, scientific consensus that violent media exposure is a causal risk factor for later
aggressive and violent behavior does not (and should not) directly translate into public policy that
restricts the production, dissemination, or use of such media by anyone; other factors play major
roles in public policy (Anderson & Gentile, 2008).
Figure
