Violence in the News: Attachment Styles as Moderators of Priming Effects
Tamyra A. Pierce, Ph.D.
California State University-Fresno
Online Publication Date: February 2, 2005
Journal of Media Psychology, Volume 10, No. 1, Winter, 2005
This study extended priming and weapons effects theories by examining the effects of aggressive stimuli within a newscast with attachment styles as the moderating variable. College students participated in a pretest/posttest experiment. The pretest questionnaire assessed television news viewing patterns and the type of attachment style individuals possessed. In the posttest, participants watched a five-minute simulated newscast and then completed a homonym test to measure primed aggressive thoughts. Results revealed that insecurely attached individuals reported significantly greater primed aggressive thoughts than did securely attached individuals. However, these significant results were only found when insecurely attached individuals were exposed to weapon words alone or when exposed to a combination of weapon words and aggressive words. Aggressive words alone were not a significant priming stimulus.
On October 1, 1997, a sixteen-year-old boy shot and killed his mother and then proceeded to his school where he lethally shot two classmates and wounded seven others. Two months later, on December 1, 1997, a fourteen-year-old boy shot and killed three students and wounded five others (Zucco, 1998). These two shootings were the first of a series of six school killings that occurred during an eight-month period between 1997-1998 (Zucco, 1998). After eight months, these six shootings committed by young males between the ages of 11 and 16, resulted in 15 people killed and more than 40 people wounded. By May 1999, the news had profiled a total of seven school shootings that resulted in over 103 people injured and 30 killed (Pierce-Plank, 2001).
The news media not only covered the school shootings between 1997 and 1999, they also increased the coverage of the amount of violence within the stories (Pierce-Plank, 2001), giving viewers an opportunity to witness the violent actions repeatedly. According to Gerow (1986), "the more often a stimulus is presented, the more likely it is that it will be attended to" (p. 13). Further, priming research suggests that when the stimulus is violent and is seen repetitively, related aggressive thoughts and feelings are cued—increasing the chances for aggressive behaviors (Berkowitz, 1984; Bushman, 1998).
Media profiles of the school shooters revealed that all experienced maladaptive forms of socialization, which could be classified as insecure attachment styles. Attachment styles are biological characteristics that develop in infancy and influence an individual’s ability to interact with others and develop social relationships with others throughout his/her lifetime (Cassidy & Shaver, 1999). Insecure attachment styles are strong predictors of anti-social behaviors (Marcus & Betzer, 1996) and contribute to the likelihood of maladaptive behaviors such as depression, aggression, and/or hostility (Asher & Coie, 1990).
Although television may not prompt aggression in all viewers, other factors may moderate certain effects and increase the chances of aggressive behaviors. This research does not intend to explain the reasons why the students committed the acts of violence in their schools in the late 1990s nor does it intend to determine any causal links between news coverage and school shootings. However, the goal of this research is to examine if and how individual differences interact with aggressive television content to prime aggressive thoughts. More specifically, the purpose of this research is to determine how aggressive stimuli within a newscast contribute to priming aggressive thoughts and how attachment styles may moderate these effects.
Research has found that viewing media violence increases the probability of aggressive behavior; however, the reasons for how media violence increases aggression are diverse. One offered by Berkowitz (1962, 1964) suggests that aggression-evoking cues elicit aggressive responses. In other words, "cues are stimuli in the post-observation situation which have some association with the depicted event or which may be connected with previous aggression-instigating situations" (Berkowitz & Geen, 1966, p. 526). Berkowitz and Geen (1966) found that not all observed aggression leads to overt aggressive behaviors, but these learned behaviors can be exhibited at a later time.
Collins and Quillian (1969) were the first to introduce and popularize the concept of associative network models (Higgins & Kruglanski, 1996). According to Collins and Loftus (1975), "when a concept is processed (or stimulated), activation spreads out along the paths of the network in a decreasing gradient. The decrease is inversely proportional to the accessibility or strength of the links in the path" (p. 411). That is, the more the concept is "continuously processed (either by reading, hearing, or rehearsing)," the longer activation occurs (Collins and Loftus, 1975, p. 411). Some research suggests that repetitive watching of violent stimuli may increase aggressive cognitive scripts and may increase the chances of accessing these links (Berkowitz, 1984; Bushman, 1998). Although only one concept can be processed at a time, the more the concept is continuously activated, the longer the activation process continues. However, the activation decreases over time due to intervening activity. For instance, if the concept of hostility were continuously primed, the activation of related links would continue until either another (related or unrelated) concept was primed or the concept of hostility was re-primed.
Anderson, Anderson, Dill, and Deuser (1998) found that aggressive words, whether read or heard, could cue aggressive thoughts. Words such as "assault," "attack," "explode," "harm," "murder," "shoot," and "stab," (among others) cued related aggressive thoughts in individuals. Buss (1961) found that when subjects were exposed to violent/hostile words, as opposed to more neutral words, there were noticeable changes in galvanic skin responses (GSR). These changes in GSR suggest, "that hostile material evokes a greater emotional reaction than neutral material" (Buss, 1961, p. 119). Therefore, words alone can be enough to evoke physically emotional reactions and this is especially the case for males (Buss, 1961). However, words are not the only stimuli that can evoke aggressive thoughts and behaviors. Weapons have also been found to cue related aggressive thoughts.
Weapons theory is largely based on the associative models, but focuses on the effects that weapons have on priming aggressive thoughts and behaviors. Specifically, the research has examined how actual weapons, pictures of weapons, or weapon words cue aggressive thoughts and behaviors. The explanation as to why this phenomenon occurs is closely linked to the priming processes outlined by the cognitive associative model (discussed earlier). For instance, weapons theory posits that weapon concepts are linked to concepts of aggression and hostility in cognitive structures. Subsequently, Subsequently, weapons prime these associated links (Berkowitz & LePage, 1967).
However, these cognitive connections are dependent on the meanings associated with the memories of the weapon. For instance, the majority of violent television images show guns that are used to threaten or harm others largely in aggressive situations (Anderson, Benjamin, & Bartholow, 1998). Viewing programs such as these would allow a person to store the concept of weapons into memory and link it to a variety of aggressive-related cognitions. Therefore, watching high volumes of violent television not only strengthens the memory concepts, but it also allows for increased activation of other aggressiveaggression-related concepts. Anderson et al. (1998) state that "seeing a gun may increase the accessibility of the associated aggressive thoughts (including scripts)" and this "increased accessibility of hostile or aggressive thoughts may facilitate subsequent aggressive behaviors" (p. 308). There have been some inconsistencies in the testing of the weapons theory but despite this, there is some varying support for the weapons theory (Carlson, Marcus-Newhall, & Miller, 1990; Frodi, 1975).
Even though some research has found support for the weapons effect using the presence of actual weapons as a cue, others have not (Frodi, 1975). However, many researchers have found support of the cue-eliciting effect of pictures of weapons and weapon words (Anderson et al., 1998; Leyens & Parke, 1975). These findings lend themselves to this research because many news programs contain images of weapons within video clips and the inclusion of weapon words within the context of their stories.
Weapon Pictures and Words as Primes. Leyens and Parke (1975) determined that pictorial depictions of guns were sufficient cue-eliciting stimuli to increase aggressive behavior. However, Anderson et al. (1998) found that the weapons effect was greater for weapon- priming words than it was for priming pictures of weapons. Therefore, weapon words significantly increased the accessibility and priming of aggression and related thoughts. Since Anderson et al. (1998) found support for cue-eliciting effects of weapons words, this may suggest that thoughts are not only semantically structured but are also primed accordingly, the idea initially suggested by Collins and Loftus’ (1975) semantic processing model. Because of this advancement in the testing of weapons theory, some suggest that homonyms may be useful tools in accessing semantic priming (Van Orden, 1987).
Bushman (1998) utilized homonyms to test priming effects of exposure to aggressive stimuli. The free associations to homonyms were recorded after the subjects viewed either a violent or non-violent film. The study found that subjects who watched the violent film listed significantly more aggressive associations to the homonyms than did those who watched the non-violent film. This study gives support of Van Orden’s (1987) suggestion that priming of thoughts is a semantic process of phonological recognition. This process then leads to mapping of related meanings and thoughts.
Even though research using weapons theory has had mixed results in the past, the use of homonyms as a tool for accessing semantic priming may prove invaluable. Closely connected, research has also refined the weapons theory by finding support for semantic priming by using weapon words and pictures, as opposed to actual weapons. In addition, moderating factors have been found to contribute to the effects of weapons theory. Specifically, personality traits tend to moderate the weapons effect.
Personality Factors as Moderators of the Weapons Effect. Caprara, Renzi, Amolini, D’Imperio and Travaglia (1984) not only looked at sex differences as a moderator of the weapons effect, but they also examined the personality construct of irritability. Although males were measured to be more biologically aggressive than females, the exposure to weapons did not result in greater aggression. The personality construct of irritability, however, did lead to greater impulsive aggression after being exposed to a weapon cue. Highly irritable subjects were found to administer significantly higher levels of shock when instigated by the weapon cue than were those who were low irritable subjects. Although there are several factors that enhance the likelihood of aggressive thoughts and behaviors, we must can and should examine a seemingly omnipresent source that possibly cues these responses. This source is media violence and more specifically, violent news.
As the research on priming shows, aggressive words and weapons can stimulate aggressive thoughts by linking related concepts in one’s cognitive network. However, as is suggested, other factors may play important roles in determining how concepts are linked. Since individual differences seem to alter the degree to which aggression is developed or cued, it is important to examine specific personality characteristics that may contribute to primed aggression, such as attachment systems.
Bowlby’s (1969/1982; 1973; 1980) attachment theory is based largely on biological factors. Initially, Bowlby (1969/1982) suggested that an infant and its primary caregiver (parent) were tied (attached) because of biological drives such as hunger.; howeverHowever, Bowlby (1980) later discovered that aside from the need for food, a stronger tie that emerged between infant and parent was the biological desire for proximity (closeness). Based on Bowlby’s (1969/1982) theory, the child’s goal is not necessarily the parent him/herself, but rather the state of feeling secure by being close in proximity. Eve
Even though attachment systems in infancy are initially biological, attachment systems evolve through infancy, ultimately predicting attachment relationships later in life (Cassidy & Shaver, 1999).
As the child approaches adolescence, attachment behavior may change. The adolescent may actively disengage the attachment relationship with parents in order to develop autonomy (Cassidy & Shaver, 1999). At this stage, a child transfers attachment behaviors towards peers. "By mid-adolescence, interactions with peers have begun to take on many of the functions that will serve for the remainder of the lifespan—providing important sources of intimacy, feedback about social behavior, social influence and information, and ultimately attachment relationship and lifelong partnerships" (Cassidy & Shaver, 1999, p. 322). Therefore, at this stage, a child transfers attachment behaviors towards peers.
How the attachment with peers develops is predominantly based on the child’s prior attachment relationship with parents (Cassidy & Shaver, 1999). If a child had a secure attachment system with his/her parents, then attachment systems with peers are also likely to be secure. However, insecure attachment systems with parents will likely result in insecure attachment with peers. Therefore, these early developments of attachment are important because they continue throughout an individual’s life-span (Ainsworth, 1982; Bowlby, 1977).
Attachment systems are categorized as either secure or insecure (Bowlby, 1977). A secure attachment system is "characterized in adolescence and adulthood by coherence in talking about attachment-related experiences and affect" and should allow for similar experiences and emotion in peer relationships (Cassidy & Shaver, 1999, p.326). The parent who responds consistently to the expressed needs of an infant will nurture in that child a secure attachment style (Vivona, 2000; Ainsworth, Blehar, Waters, and Wall, 1978). For adolescents who have developed secure attachment systems in parental relationships, the development pathway of attachment with others, including peers, is fairly straightforward and smooth.
Those infants whose needs were not consistently and positively met will likely develop insecure attachment styles. For individuals with insecure attachment systems, the development of relationships with others may be problematic and difficult. Specifically, there is a "defensive exclusion of information about attachment" that "may lead to distorted communications and negative expectations about others, both of which have been linked to problems in social functioning" (Cassidy & Shaver, 1999, p. 326). Insecure attachment systems that include loss and/or rejection have been found to lead to depression, increased aggression, and hostility (Bowlby, 1980). Strong attachment bonds are associated with positive affective responses; however, the perception of loss arouses anxiety and the experience of an actual loss stimulates emotions of sorrow and depression. Research suggests that differing attachment styles are closely linked to emotions and subsequently different attachment styles may predict distinct behavioral characteristics and personality styles (Bowlby, 1979).
Personality Characteristics and Attachment. Insecure attachment is a strong predictor of anti-social behavior (Marcus & Betzer, 1996). Attachment theory suggests that anti-social and aggressive behaviors in insecurely attached individuals are the result of the threat or presence of separation (Bowlby, 1973). For individuals who have insecure attachment, "the world is seen as comfortless and unpredictable; and they respond either by shrinking from it or by doing battle with it" (Bowlby, 1973, p. 208). Further, aggression and anti-social behavior in childhood is a strong predictor for anti-social behaviors in adulthood (West, 1982).
Hirschi (1969) suggested that the lack of attachment to others results in inadequate socialization skills needed to create healthy social relationships with others. Insecure attachment styles have not only been found to increase the development of peer aggression but also increase the likelihood of overall aggressive behaviors toward others (Troy & Sroufe, 1987). Similarly, Kupersmidt and Coie (1990) found that peer rejection was a strong predictor of delinquency (i.e., those individuals who commit criminal actions as indicated by police records). As Asher and Coie (1990) suggest, peer rejection in vulnerable individuals, such as those who have insecure attachment systems, contributes to the likelihood of maladaptive behaviors such as aggression and/or hostility.
Although there may be many factors that contribute to priming aggressive thoughts, one important element may lie within the tenets of attachment theory. Since research suggests that insecurely attached individuals exhibit more violent images and thoughts (Woike, Osier, and Candela, 1996), perhaps insecurely attached individuals are more readily primed by violent words or images. With this in mind, the following hypothesis was proposed:
H1: A news story containing aggressive stimuli (aggressive words and weapon words) will elicit significantly more primed aggressive thought in individuals with insecure attachment systems than those with secure attachment systems.
Although there has been extensive research on the priming effects of aggressive stimuli, research has not examined the priming effects of violence in television newscasts. As a resultIn light of this literature gap, this study investigated the priming effects of violent content within television newscasts. Further, this study extended research on priming and weapons effects by examining the plausible moderating variable of attachments styles.
Participants included 232 college students (102 males and 130 females) from a large Midwest university. Participants’ ages ranged from 18 to 34 (M = 19, SD = 1.7). As a result of 16 participants dropping out of the study after the pretest, a total of 216 individuals completed both the pretest and posttest. Participants received extra credit for participating in both the pretest and posttest.
Design and Procedures
The proposed experiment was a 2 (aggressive word, non-aggressive word) X 2 (weapon word, non-weapon word) pretest-posttest design with the dependent variable being primed aggressive thoughts. The degree to which the dependent variable was altered by the independent variable (version of news story) was predicted to be contingent upon one moderating variable: attachment style.
Participation in the study was voluntary and participants gave oral consent prior to any testing. Because of the proposed design, responses from the pretest had to be associated with responses from the posttest. Participants were asked to include three identification codes on the pretest and then again on the posttest. The identification codes consisted of the last four digits of the participants’ phone number, the last four digits of their social security number, and the month in which they were born. In the pretest phase, participants completed a questionnaire consisting of several scales: television exposure, amount of time spent watching a variety of news programs, and attachment.
After three weeks, the participants returned to the testing site and completed the posttest. This time lapse was included between the pretest and the experimental manipulation to avoid contamination of the priming effects. To match previous tests with the same subjects, all participants were asked to record the same identification codes they included on the pretest questionnaire onto the posttest questionnaire.
Participants were randomly assigned to view one of the eight versions of the 5-minute newscast. After viewing the newscast, participants completed the posttest homonym test.
Television exposure. The researcher developed a television news viewing measure. Participants were asked how often they watched different types of news programs (e.g., local, national, cable news, or news magazine) and how much attention they gave to the coverage. These seven items were assessed using a 7-point Likert-type scale (ranging from 7 = all the time and close attention to 1 = rarely and little attention respectively). Reliability was acceptable for the Likert-type questions (a = .74).
Attachment. The Adult Attachment Scale (Simpson, 1990) identified attachment style as insecure or secure. The 9 attachment questions were measured on a 5-point (1 = strongly disagree to 5 = strongly agree) Likert-type scale. The insecure items of the insecurity/security dimension included: (a) "I’m somewhat uncomfortable being too close to others," (b) "I find it difficult to trust others completely," (c) "I’m nervous whenever anyone gets too close to me," (d) "Others often want me to be more intimate than I feel comfortable being" (Simpson, 1990). Higher scores reflect greater insecurity and lower scores reveal greater security. Security attachment questions included: (a) "I find it relatively easy to get close to others," (b) "I’m not very comfortable having to depend on other people," (c) "I’m comfortable having others depend on me," (d) "I rarely worry about being abandoned by others," and (e) "I don’t like people getting too close to me" (Simpson, 1990, p. 973). Reliability was good (a = .85).
Cue-Eliciting Video. During a pilot test, a sample group of 100 evaluators were drawn from the same population as those who later served as participants in the study. These evaluators rated the degree of aggressiveness of the manipulation words used in the newscasts previously reported to be aggressive (Anderson et al., 1998) (e.g. shooting, massacre, assault, violent, aggression, aggress, explode, harmed, attack, torture(ed), hurt, murderous, deadly). Words such as button, feather, glide, relax, and bird were included as control words and were rated along with the aggressive words. Ratings were made on a
5-point (1 = not at all aggressive to 5 = extremely aggressive) scale.
To create the aggressive word manipulation (e.g., aggressive word news story), 11 of the 31 pre-tested aggressive words were selected. The words were opened fire, tormented, shooting, attack, violent (used twice), explode, blowing up, anger, assault, and violence. Paired-samples t tests were conducted to ensure the aggressive words used in the story were perceived as significantly more aggressive than control words. A mean score for all the control words was computed and tested against the mean score for each aggressive word. To reduce the probability of a Type 1 error, a Bonferroni-type correction was conducted by dividing the significance level (.05) by the number of tests (11). The corrected p-value was .005. All comparisons were statistically significant. These 11 words were excluded from or replaced with a neutral word for the no aggressive word manipulation (e.g., "shooting" was replaced with "incident").
To create the weapon word manipulation, five weapon words were selected from previous research by Anderson et al. (1998). These five words were written into the school shooting script to constitute the weapon word manipulation. The five words were shotgun, automatic rifle, bomb, gun, and club. These words were excluded from the no weapon word manipulation.
Each school shooting manipulation (1 minute in length) was embedded within a 4-minute newscast (total length of the entire newscast being 5 minutes). In addition, the news stories surrounding the experimental condition remained constant, and the experimental story was the only element of the newscast that changed. Further, the news stories surrounding the experimental story were neutral in content and context and did not contain any aggressive words, weapon words, or aggressive pictures (e.g., computer story, flood story, *school crime [1 of 4 manipulations], farming story, and an asthma story).
The newscasts were professionally videotaped in a real news station to ensure the simulation of an actual portion of a newscast. 1 To maintain consistency, two professional news anchors (one male, one female) delivered the stories in an alternating pattern that is normally seen on television news broadcasts. For instance, the male anchor presented the first story in the newscast followed by the female anchor delivering the second story. This sequence continued throughout the 5-minute news segment and remained constant in all of the manipulation videos. In each newscast, the male anchor delivered the experimental story.
Priming. Because of their dual meaning, homonyms were used to assess primed aggression. For example, some homonyms have dual meanings in which one could be aggressive and the other could be non-aggressive (e.g., "punch" can elicit an aggressive association of "hit" or a non-aggressive association of "fruit"), depending on individual interpretations and associations (Bushman, 1998). Given this, to assess priming effects, after viewing the newscast, each participant was given a list of 50 words (26 homonyms and 24 control words). Respondents were instructed to write the first word that came to their mind after reading each word and to not spend a lot of time thinking about the word. There were two versions of the posttest questionnaire that positioned the homonyms differently to allow for a check of order effects. In version one, an aggressive word was positioned first on the homonym list (e.g. "punch"). In version two, a control word was positioned first on the list of homonyms (e.g. bloom).
An aggressive response to a homonym received a score of 1. For instance, a response of "hit" next to the homonym "punch" would receive a 1. Neutral or non-aggressive responses received a score of zero. After each response was coded for aggressiveness, scores were totaled.
Results and Discussion
Testing for Order Effects
Prior to the analysis of the results, a test was conducted to examine the possibility that the order in which the homonyms were presented influenced the primed aggressive thoughts.. Independent samples t-tests were conducted using order of the homonyms as the independent variable (2 levels), and the number of primed aggressive thoughts as the dependent variable. A significant order effect revealed that participants who completed the priming homonym questionnaire (the version of the questionnaire that began the homonym list with a priming homonym that could gain a response of either violent or non-violent) reported more primed aggressive thoughts (M = 4.47, SD = 2.7) than those who completed the control version (the version that started with a control word) (M = 3.76, SD = 2.5); t (214) = -1.99, p < .05. To control for this order effect, the homonym variable was built in as a covariate in the analysis of the hypothesis.
Manipulation checks were performed on the three experimental conditions: weapon word/no weapon word and aggressive word/no aggressive word using aggressive thoughts as the criterion variable. The weapon tape manipulation was significantly different for the two conditions, with individuals who viewed the newscast that included weapon words reporting more aggressive thoughts (M = 4.55, SD = 2.5) than those who viewed the newscast without weapon words (M = 3.63, SD = 2.6), t (214) = 2.60, p < .01 (one-tailed). Similarly, the aggressive word manipulation was significant for the two conditions. Individuals who viewed the tape with the aggressive words reported more aggressive thoughts (M = 4.36, SD = 2.5) than those who viewed the newscast without aggressive words (M = 3.87, SD = 2.7), t (214) = 1.85, p <.05 (one-tailed).
Results of Hypothesis
The hypothesis predicted a significant interaction between newscasts containing aggressive stimuli and attachment on primed aggressive thoughts. It was expected that aggressive stimuli (aggressive words, weapon words) would elicit more aggressive thoughts among insecurely attached individuals rather than among securely attached individuals.
The attachment by weapon words interaction was significant, F (2, 208) = 5.36, p < .01, h 2 = .05. Aggressive thoughts were most frequent in the insecure attachment-weapon word condition (M = 4.80, SE = .34), followed by the secure attachment-weapon word condition (M = 4.30, SE = .32), the insecure attachment-no weapon word condition (M = 3.94, SE = .35), and the secure attachment-no weapon word condition (M = 3.21, SE = .37).
The predicted two-way interaction between attachment and aggressive words was not significant, F (2, 208) = 1.04, p = ns, observed power = .23. Finally, the predicted three-way interaction between attachment and weapon words and aggressive words was significant, F (2, 208) = 6.14, p < .01, h 2 = .05. Aggressive thoughts were most frequent in the insecure-weapon words and aggressive words condition (M = 5.13, SE = .44), followed by the insecure attachment-weapon words and no aggressive words condition (M = 4.88, SE = .49), followed by the secure attachment-weapon words and aggressive words condition (M = 4.73, SE = .46), and finally, the secure attachment-weapon words and no aggressive words condition (M = 4.56, SE = .48).
The goal of this study was to examine the priming effects of aggressive stimuli televised in a newscast with attachment as the moderating variable. It was hypothesized that aggressive stimuli would elicit more aggressive thoughts among individuals with insecure attachment than those with secure attachment styles.
Results of this study found that both manipulations that included weapon words (weapon words alone and the combination of weapon words and aggressive words) cued more primed aggressive thoughts in insecurely attached individuals. These results are to be expected among insecurely attached individuals because individuals with insecure attachments are more prone to aggressive behaviors (Troy & Sroufe, 1987).
In contrast, the non-significant results were not expected. Although there may be any number of plausible reasons for the lack of primed aggression with the aggressive word stimuli, one viable reason may be due to the personality factor of anger. Anger can both contribute to and inhibit aggression, depending on the individual. Some individuals are more aroused/angered by certain cues than others. As Berkowitz (1993) suggested, individuals who are angered are more readily provoked into aggressive actions. Therefore, perhaps the weapon words and the combination of weapon words and aggressive words served to arouse certain individuals more than did the aggressive words alone. In contrast, other individuals may have been inhibited by the aggressive words. Future research should examine the role that anger plays in inhibiting or instigating aggression.
Finally, how individuals interpreted the aggressive stimuli may have altered how they responded to the aggressive stimuli. Specifically, depending on whether the individual associated the aggressive stimulus with consequences, the primed aggression may have been either inhibited or instigated. As Geen and Donnerstein (1998) found, consequences can serve as both reinforcement (instigator) and/or punishment (inhibitor) of a behavior. Therefore, perhaps some of the subjects interpreted the aggressive stimuli as a type of punishment and consequently were inhibited for aggressive thoughts. With this in mind, it is important that future research studies evaluate how the viewer interprets the aggression in order to determine if the aggressive stimulus inhibits or further instigates the primed aggression.
Although no study is without weaknesses, there are a couple of limitations with this study that should be addressed. One limitation of this study coincides with the criticisms concerning the use of college students as participants (Abelman, 1996; Basil, Brown & Bocarnea, 2002). According to the criticisms, scientific studies within the field rely too heavily upon college students for their sample (Sears, 1986). Instead, Sears (1986) suggests that scholars should utilize adults, random samples, and conduct tests in more natural settings. Based on the criticisms, using non-representative samples, such as college students, does not allow for us to make accurate inferences nor can we generalize to other generations (Abelman, 1996).
Another weakness of this study pertains to the criticisms about media effects research and viewer-violence connections. According to Fowles (1999), content sent via television is not the same as the content received by the viewers. "Each viewer ignores, selects, interprets and misinterprets the content according to his or her needs and temperament at the moment" (p. 7). As the vast amount of research shows, not all viewers interpret the content in the same way. In addition, individual differences and personalities may skew the results or at worst, deem the results inconclusive (Vine, 1997). In sum, even though this study found significant support for the weapons effect, other individual factors may have altered these results. As was mentioned previously, future research must examine the individual interpretations of the content as well as individual personality factors.
Any significant findings, no matter how small, are important when investigating the priming effects of aggressive stimuli. Moreover, possible links between violence in the news, primed aggression, and their possible influence on imitative violence among young adults are equally important areas that must continue to be examined. Although this study did not examine any links between primed aggression and the school shootings of the late 1990s, it is very possible that the endless aggressive news coverage of past school shooting may have increased the likelihood of primed aggressive behaviors in some individuals. We may never know the reasons for the thousands of copycat threats that followed the school shooting coverage but with more research we can determine which individuals may have a greater propensity for primed aggression.
1The author contacted the news director at KOMU-Channel 8 (in Columbia, Missouri where the study took place) and explained the study and asked if the station could help with the making of the newscasts. The news editor agreed only if the anchors also agreed to participate—they did. Prior to arriving at the station, the author had gathered the neutral stories from previous KOMU newscasts and also typed the different versions of the school shooting manipulation stories (e.g. weapon words/no weapon word; aggressive words/no aggressive words). Upon arrival, a former student of the author, who was working at the station, typed in the stories into the teleprompter. Tapings occurred between the noon and evening segments and took approximately 20 minutes to complete. After the news stories were taped, the author edited (non-linear) the various manipulation stories and neutral stories into the 4 manipulation tapes that were used in the study. The opening of the story included the normal news opening (logo and music). The author extends deepest thanks to the station and employees for their assistance with this study.
Abelman, R. (1996). Can we generalize from Generation X? Not! Journal of
Broadcasting and Electronic Media, 40, 441-446.
Ainsworth, M. D. S. (1982). Attachment: Retrospect and prospect. In C. M. Parkes & J.
Stevenson-Hinde (Eds.), The place of attachment in human behavior (pp. 3-30).
New York: Basic Books.
Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of
attachment: A psychological study of the strange situation. Hillsdale, NJ:
Anderson, K. B., Anderson, C. A., Dill, K. E., and Deuser, W. E. (1998). The interactive
relations between trait hostility, pain, and aggressive thoughts. Aggressive
Behavior, 24, 161-171.
Anderson, C. A., Benjamin, A. J. Jr., & Bartholow, B. D. (1998). Does the gun pull the
trigger? Automatic priming effects of weapon pictures and weapon names.
Psychological Science, 9, 308-314.
Asher, S. R. & Coie, J. D. (1990). Peer rejection in childhood. New York: Cambridge
Basil, M. D., Brown, W. J., & Bocarnea, M. C. (2002). Differences in univariate values
versus multivariate relationships. Human Communication Research, 28, 501-514.
Begley, S. (1999, May 3). Why the young kill. Newsweek, 32-35.
Berkowitz, L. (1962). Aggression: A social psychological analysis. New York:
Berkowitz, L. (1964). Aggressive cues in aggressive behavior and hostility catharsis.
Psychological Review, 71, 104-122.
Berkowitz, L. (1984). Some effects of thoughts on anti-social and prosocial influences of
media events: A cognitive-neoassociation analysis. Psychological Bulletin, 95,
Berkowitz, L. & Geen, R. G. (1966). Film violence and the cue properties of available
targets. Journal of Personality and Social Psychology, 3, 525-530.
Berkowitz, L. & LePage, A. (1967). Weapons as aggression-eliciting stimuli. Journal of
Personality and Social Psychology, 7, 202-207.
Bowlby, J. (1969/1982). Attachment and loss: Vol. 1. Attachment. New York: Basic
Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation: Anxiety & anger. New
York: Basic Books.
Bowlby, J. (1977). The making and breaking of affectional bonds. British Journal of
Psychiatry, 130, 201-210.
Bowlby, J. (1979). The making and breaking of affectional bonds. London: Tavistock.
Bowlby, J. (1980). Attachment and loss: Vol. 3: Loss. New York: Basic Books.
Bushman, B. J. (1998). Priming effects of media violence on the accessibility of
aggressive constructs in memory. Personality and Social Psychology Bulletin, 24,
Buss, A. H. (1961). The psychology of aggression. New York: Wiley.
Caprara, G. V., Renzi, P., Amolini, P. D’Imperio, G., & Travaglia, G. (1984). The
Eliciting cue value of aggressive slides reconsidered in a personological
perspective: the weapons effect and irritability. European Journal of Social
Psychology, 14, 313-322.
Carlson, M., Marcus-Newhall, A., & Miller, N. (1990). Effects of situational aggression
cues: A quantitative review. Journal of Personality and Social Psychology, 58,
Cassidy, J., & Shaver, P. R. (1999). Handbook of attachment: Theory, research, and
clinical applications. New York: Guilford Press.
Collins, A. M., & Quillian, M. R. (1969). Retrieval time from semantic memory. Journal
of Verbal Learning and Verbal Behavior, 8, 240-247.
Collins, A. & Loftus, E. (1975). A spreading activation theory of semantic memory.
Psychological Review, 82, 407-428.
Fowles, J. (1999). The case for television violence. Thousand Oaks, CA: Sage
Frodi, A. (1975). The effect of exposure to weapons on aggressive behavior from a cross-
cultural perspective. International Journal of Psychology, 10, 283-292.
Geen, R. G. & Donnerstein, E. (1998). Human aggression: Theories, research, and
implications for social policy. San Diego, CA: Academic Press.
Gerow, J. R. (1986). Psychology: An introduction. Glenview, IL: Scott Foresman &
Higgins, E. T., & Kruglanski, A. W. (1996). Social psychology: Handbook of basic
principles. New York: Guilford Press.
Hirschi, T. (1969). Causes of delinquency. Berkeley, CA: University of California Press.
Kupersmidt, J. B. & Coie, J. D. (1990). Preadolescent peer status, aggression, and school
adjustment as predictors of externalizing problems in adolescence. Child
Development, 61, 1350-1362.
Leyens, J. P. & Parke, R. (1975). Aggressive slides can induce a weapons effect.
European Journal of Social Psychology, 5, 229-236.
Marcus, R. F. & Betzer, P. D. S. (1996). Attachment and antisocial behavior in early
adolescence. Journal of Early Adolescence, 16, 229-248.
Pierce-Plank, T. (2001, April). Network news coverage of the school shootings from Fall
1997 through Spring 1999: Sensationalism? Paper presented at Central States
Association Conference, Cincinnati, Ohio.
Sears, D. O. (1986). College sophomores in the laborartory: Influences of a narrow data
base on social psychology’s view of human nature. Journal of Personality and
Social Psychology, 51, 515-530.
Simpson, J. A. (1990). Influence of attachment styles on romantic relationships. Journal
of Personality and Social Psychology, 59, 971-980.
Troy, M., & Sroufe, L. A. (1987). Victimization among preschoolers: Role of attachment
relationship history. Journal of the American Academy of Child and Adolescent
Psychiatry, 26, 166-172.
Van Orden, G. C. (1987). A ROWS is a ROSE: Spelling, sound, and reading. Memory &
Cognition, 15, 181-198.
Vine, I. (1997). The dangerous psycho-logic of media "effects." In Martin Barker &
Julien Petley (Eds.), Ill effects: The media/violence debate (pp. 125-146). London: Routledge.
Vivona, J. M. (2000). Parental attachment styles of late adolescents: Qualities of
Attachment relationships and consequences for adjustment. Journal of Consulting
Psychology, 47, 316-329.
West, D. J. (1982). Delinquency: Its roots, careers, and prospects. London: Heinemann
Woike, B. A., Osier, T. J., & Candela, K. (1996). Attachment styles and violent imagery
in thematic stories about relationships. Personality and Social Psychology Bulletin, 22, 1030-1034.
Zucco, T. (1998, June 14). Shooters in the school yard. Floridian, p. 1F.
Return to home page