Bradford L. Yates*, Michelle Ballard, Mary Ann Ferguson, Kirk Filer,
Ann Villanueva, Alison Knott, and Tracy Cristal
University of Florida
*College of Journalism and Communications
P.O. Box 118400
Gainesville, FL 32611-8400
(352) 392-7966 (o)
Paper presented to the Communication Theory and Methodology Division
Association for Education in Journalism and Mass Communication
Baltimore, Maryland August 5-8, 1998
Video Violence: Desensitization and Excitation Effects on Learning
An experiment tested desensitization and excitation effects of video violence on learning. In accordance with excitation theory, it was hypothesized that viewers exposed to violence will have poorer recall of events compared to viewers not exposed. Desensitization theory led to the proposition that viewers desensitized to violence and exposed to a violent stimulus will have better recall of events compared to non-desensitized viewers. Results indicate support of desensitization theory for information presented after violent stimuli.
Video Violence: Desensitization and Excitation Effects on Learning
Television is one of the most influential forces in people's lives. Since no one can acquire all their knowledge from direct experience, people often turn to the media to provide them with information about events, people, or even products. It is in this manner that people learn to orient themselves in the world in which they live (Bandura, 1994). While the primary goal of certain media, such as newscasts and commercials, is to inform and educate the public, past studies have shown that people recall very little of what they see on television (Katz, Adoni & Parness, 1977). Broadcasters of informative programming need to be aware of factors that may enhance or inhibit the acquisition of knowledge by the audience. The purpose of the present study is to investigate the impact of one type of media content, television violence (i.e., severe aggression; that is, acts that could potentially cause physical injury or death to an individual (Royal Commission on Violence, 1976)), on one's ability to remember the information viewed.
There is a general belief that television violence has an impact on its audience (Bandura, Ross & Ross, 1963; Berkowitz & Rawlings, 1963; U.S. Surgeon General's Scientific Advisory Committee, 1972; Wharton & Mandell, 1985; Gadow & Sprafkin, 1989; Singer, 1989; Gerbner, 1992; Comstock & Strasburger, 1993; Gerbner, Morgan, & Signorielli, 1994; Committee on Communications, 1995). Much of the viewing audience is made up of children. They are constantly inundated with violence every time they turn on the television or see a movie. Over 1,000 studies have demonstrated a causal connection between media violence and aggressive behavior in children (U.S. Surgeon General's Scientific Advisory Committee, 1972; U.S. Department of Health and Human Services 1982; Comstock, 1989; Committee on Communications, 1995). In addition, exposure to violent content can give rise to hostile or violent thoughts (Jo & Berkowitz, 1994). With the pervasiveness of violence in society there is cause for continual reexamination of its effects on individuals. The present study focuses on another possible result of television violence: the effects of viewing violent content on viewers' recall and recognition of factual information presented before and after violent scenes.
The Process of Memory Recall
To facilitate understanding of recall, an outline of how information is stored in memory and then recalled is useful. Theories suggest there are three steps that must be followed to recall previously presented information: encoding, storage, and retrieval. Encoding occurs when a person views information that is then processed and understood in terms of conceptual representations in the short-term memory. Since people cannot take in everything they see at once, the stimuli to which the person pays attention will be primarily encoded (Kellermann, 1985).
Storage occurs when information accumulated in short-term memory is transferred into long-term memory. Again, certain factors affect what information will be stored in long-term memory. For example, if a person mentally rehearses a stimulus to which they have been exposed, it is more likely that information will be transferred to long-term memory (Kellermann, 1985).
Finally, retrieval is the process of accessing information that has been stored. However, just because information has been encoded and stored does not mean that it can be retrieved on demand. The broader the category of mental information to be searched, the more difficult it will be for the person to retrieve, and therefore recall, specific information (Kellermann, 1985).
In understanding the effect of violence on viewers' memories, the concept of recall can be divided into two subcategories: aided and unaided. Unaided recall is the ability to recall information without any outside suggestion. Unaided recall is frequently operationalized by asking respondents to write down everything they remember about a stimulus (Kellermann, 1985; Scott & Goff, 1988). On the other hand, aided or cued recall is assisted by outside suggestion. It is often operationalized by fill-in-the-blank and multiple choice questions (Loftus & Burns, 1982; Scott & Goff, 1988). While both questions cue the respondents memory, multiple choice queries actually measure recognition rather than recall because the respondent is forced to choose the correct answer from a list of possible choices (Loftus & Burns, 1982; Mundorf, Drew, Zillmann, & Weaver, 1990).
Warrington & Weiskrantz (1970) suggested that aided recall might be easier than unaided recall. Unaided recall requires a much larger category of information to be searched, basically the list of all possible answers. Aided recall, on the other hand, narrows significantly the category of mental information to be searched. Moreover, other researchers (Loftus & Burns, 1982; Prasad & Smith, 1994) have suggested that recognition may be easier than either type of recall. With memory recognition measures the answer is one of the multiple choices; the individual need only direct his/her cognitive facilities to those specific possible answers. Consequently, most researchers (Warrington & Weiskrantz, 1970; Loftus & Burns, 1982; Kellermann, 1985; Prasad & Smith, 1994) agree that unaided recall is more difficult than aided recall and both types of recall are more difficult than recognition.
One of the primary theories that has been used to predict the effects of media violence on recall is excitation theory (Scott & Goff, 1988; Newhagen & Reeves 1992; Prasad & Smith, 1994). Excitation theory states that when viewers view a compelling video, such as a violent film clip, they become physiologically aroused. When the compelling video terminates, the internal state of arousal remains. The remaining arousal is then carried over into subsequent situations experienced by viewers (Cantor, Mody, & Zillmann, 1974).
It is argued that the viewers' state of arousal leads to enhanced cognitive performance because arousal tends to activate a "fight or flight" mechanism. Even though viewers may know that what they see on television is not an actual threat, the subconscious is unable to distinguish it as such for a few moments. Consequently, the channels for acquiring information are opened to deal with this threat. As a result, viewers' recall or recognition of factual details, after being presented a violent video, will be enhanced compared to recall or recognition of information presented before the violence.
Scott and Goff (1988) demonstrated that recall of information presented immediately after a violent image is significantly poorer than information presented two minutes after the violence. Scott and Goff hypothesized that viewers aroused by a violent video may be unable for a short time to attend to the information presented after the video. Viewers will still be focused on the violence just witnessed, and therefore, distracted from the new information; however, after the two-minute interval, viewers will still be aroused, but no longer distracted by the violent stimulus. Consequently, recall for factual detail will be enhanced until the subjects' arousal eventually wears off. In addition, subsequently presented information will be recalled or recognized better among aroused viewers than among those not physiologically aroused (Newhagen & Reeves, 1992).
Under excitation theory, exposure to a violent video also decreases recall for factual details occurring before the video. Newhagen and Reeves (1992) argue that arousal, which results from the presentation of a violent video, interferes with the rehearsal necessary to store previously viewed factual information in the viewers' memory. Consequently, a subject who has been exposed to a violent stimulus will have significantly poorer recall of details occurring before the stimulus compared to a subject who has not been exposed to the stimulus.
Moreover, Loftus and Burns (1982) demonstrated that for events occurring immediately (i.e., within four seconds) prior to the violent incident, memory was drastically poorer for subjects who had been exposed to a violent stimulus compared to subjects not exposed to the stimulus. In the Loftus and Burns study, viewers saw a tape of a bank robbery. In the experimental film (i.e., with violent stimuli), the bank robbers shot a child in the face. In the control film (i.e., no violent stimuli), the film cuts back to the interior of the bank. In both films, approximately four seconds before the critical moment (either the child is shot or the film cuts to the bank interior) a boy wearing a football jersey with a number on the back runs into the scene. The shirt is visible for approximately two seconds. When subjects in both groups were later asked, via cued recall, the number on the boy's shirt, only 4% of the subjects exposed to the violent film could recall the number, while 27.9% of subjects exposed to the nonviolent film were able to give the correct number.
A second experiment was conducted to see if similar results were obtained using a forced-choice recognition design (i.e., multiple choice). While more subjects exposed to the violent stimulus were able to choose the correct number (28%), this percent was still significantly lower than the number of subjects in the nonviolent condition who chose the correct number (55%). Thus, the results obtained in these studies, where recall and recognition was inhibited for factual detail occurring before a violent stimulus, support excitation theory.
Nevertheless, other researchers have obtained results that appear contrary to excitation theory. Mundorf and his colleagues (1990) found that subjects exposed to violent content had significantly poorer recognition for factual detail occurring after the stimulus compared to subjects not exposed to violent stimulus. To explain these results, Mundorf et al. suggest that when subjects are exposed to violent stimuli, they become preoccupied with their aroused emotional state. Focusing on this state and seeking ways to reduce their arousal, subjects pay little attention to subsequent factual detail. As with excitation theory, the physiological arousal eventually wears off and subjects' ability to acquire information returns to normal. In fact, excitation theory and the proposal of Mundorf et al. appear quite similar except Mundorf's proposal does not have a period of enhanced recall after the initial impairment, which the excitation theory does. Rather, under the preoccupation proposal, impairment remains throughout the aroused state.
Desensitization theory states that individuals who watch large amounts of violence become less sensitive to future violent content than individuals who watch less violence (Comstock, 1989). Psychologists have demonstrated that people gradually become less physiologically and emotionally aroused as they view more violence. For example, Cline, Croft, and Courrier (1973) showed a violent television portrayal to children who were heavy television viewers and those who were not heavy viewers. Cline and colleagues found that children who watched a lot of television (arguably a violent medium) became less physiologically aroused when shown the violent clip compared to the children who were not heavy viewers.
In another study, Thomas, Horton, Lippincott, and Drabman (1977) showed different film clips to two groups of 8-10 year-old children. One clip was of a violent police drama. The other was of a volleyball game. Each group was then shown a television clip of real-life violence. Thomas et. al found that the children who had viewed the violent police drama were significantly less aroused by the subsequent violent clip compared to the children who had viewed the non-violent volleyball game. Thomas and her colleagues repeated this experiment with college students and obtained similar results.
Many studies have demonstrated that viewing violence on television desensitizes viewers to subsequent violent images. However, it has not been established whether this desensitization occurs only for violent portrayals on television or in movies, or if the desensitization also occurs toward violence in the real world. This is one area of future investigation in desensitization theory (Comstock, 1989).
The present study seeks to examine the effects of violence on aided and unaided recall for factual details occurring before and after the violent video under conditions where viewers are desensitized because they are regular violence viewers versus those viewers who are not. In accordance with excitation theory and desensitization theory, the following hypotheses are proposed:
H1: Those exposed to video violence will recall less about information presented before the violence than will those who did not see the violence.
H2: Those who are desensitized to violence and exposed to the video violence will recall more about information presented before the violence than will those not desensitized.
H3: For those viewing violent stimuli, recall for information presented after the stimuli will be greater than for those who did not see the stimuli.
H4: Those who are desensitized to video violence and exposed to the violence will recall more after the violence than will those not desensitized.
A pool of 77 subjects from an undergraduate advertising course at a large university participated in a post-test only control group experiment. Each subject was randomly assigned to the experimental (violence) or control group (no violence) using a color code. All participating subjects were asked to fill out a preliminary questionnaire. After the subjects completed the form, all consenting participants who had been randomly selected to view the non-violent video described below were asked to follow members of the research team to an auditorium-style room in an adjacent building. The subjects remaining in the original auditorium who viewed the violent version of the movie constitute the experimental group, while those who moved are the control group.
Both groups were shown a 5:23 video clip of the movie, "Henry: Portrait of a Serial Killer." The clip was projected onto a large screen in both auditorium-style rooms. The experimental group saw the entire 5:23 as it appears in the original movie. In this clip there is a sequence of 48 seconds depicting a murder scene in which two men repeatedly stabbed another man with an electric prodder on parts of his upper body. The victim was choked with an electrical extension cord and his head was smashed with a television set. The television was eventually plugged into an outlet sending electrical charges throughout the body and thus causing an excessive amount of bleeding. On the other hand, those in the control group did not see the 48 seconds of violence contained in the original clip. Those 48 seconds were replaced with a nonviolent clip taken from another part of the movie. Therefore, the control group viewed the same material as the experimental group for 2:40 before the violent scene and 1:55 after the violent scene, but was not exposed to the violent stimulus (Figure 1).
Figure 1 Schematic representation of stimuli used in experiment
Experimental Group Total = 5:23
2:40 :48 1:55
Control Group Total = 5:23
2:40 :48 1:55
After the subjects viewed the clip they were asked to complete a 19-item questionnaire that tested their recall of facts from the movie clip. Seven of the items on the questionnaire tested recall of factual details occurring before the violent stimulus, and seven questions tested recall of factual details after the stimulus. Four of these fourteen questions were randomly assigned in unaided form to one version of the questionnaire and randomly assigned in aided form to a second version of the questionnaire. Thus, one-half of the subjects in the experimental group and in the control group received the unaided version of these four questions, while the other half of each group received the aided version of these four questions. These four questions were designed to measure recall of details that occurred 1-4 seconds before the stimulus, 1-4 seconds after the stimulus, one minute before the stimulus, and one minute after the stimulus. In addition, other items measured the overall appeal of the movie clip and the type of movies subjects generally watched.
Although both aided and unaided recall questions are used to assure that recall was adequately measured, the responses of both types of questions will be used in a general recall index unless we find evidence for differences for two types of recall measures.
Of the 77 subjects who participated in the study, two were disqualified because they had seen the movie used as the stimulus. Thus, the total number of subjects was 75. Of the 75 respondents 36 were male and 39 female. Although the subjects ranged in age from 19 to 32 years, 75% were 20 or 21 years old. Nearly all of the respondents (97.4%) were upperclassmen (i.e., juniors or seniors), and over 95% were advertising majors.
The Recall Measures
Each answer to the questions that dealt with factual information about the film clip were coded as a 1 for correct and a -1 for incorrect. Another question asked how certain or uncertain the respondent was about their answer. The subjects' certainty of the correctness of their answer was multiplied by the -1 or +1. Thus, respondents who gave a correct response and were certain of the response received a value of +5 for the answer while the incorrect response with high certainty would earn a -5. This equation provided the opportunity to distinguish between respondents who actually recalled the information and those who by chance provided the correct response.
An index was created from the pre-violence (Table 1) and post-violence (Table 2) recall measures. As mentioned above, the response to each question that dealt with information before the violent stimulus was multiplied by subjects' level of certainty of their response. These newly created scores were summed and averaged to create an index of recall of information presented before the stimulus. Scores ranged from 35 to -35. The same procedure was conducted for questions dealing with events after the stimulus, which resulted in an index of recall of information presented after the stimulus.
The Desensitization Measures
The questionnaire included several indicators of desensitization for tendency toward violence viewing. Participants indicated from a list of 47 movies those movies they had seen. The goal was to identify any subjects who had seen the movie used as the stimulus in this study, and to obtain a measure of general violent movie viewing habits.
The list of movies included many genres. After coding the movies as either seen (1) or not seen (0), we conducted a factor analysis to determine whether there was a violent films factor. A Principal axis factoring, Varimax solution indicated a 3-factor solution. Movies loading on the factor that appeared to represent violent movies included the 13 movies in Table 3.
After scoring (1=seen and 0=not seen), a summed index of the 13 movies were correlated with four other questions that measured violent viewing habits to validate the movie index as an indicator of desensitization. The correlation coefficients indicated a moderate correlation between the two measures that dealt directly with the violent stimulus (r = .33, p £ .002; r = .37, p £ .001) and a very strong correlation between the two measures that asked about violence viewing in general (r = .67, p £ .001; r = .68, p £ .001). (See Table 4.) Based on these correlations, we were relatively confident about using the movie index as an indicator of desensitization.
The movie index was used to categorize subjects in terms of their violent viewing habits. Based on a median split, two levels (high and low) of sensitization to violence were created. Those who saw between 9 and 13 of the violent films (54.7 %) were classified as highly desensitized to violence. Subjects who saw 0 to 8 of the movies (45.3%) were categorized as subjects with low desensitization to violence. The high and low categories of desensitization were used to test for a multiple analysis of variance on pre-violence viewing recall (Table 1), and post-violence viewing recall (Table 2).
A t-test or an ANOVA, as appropriate, was run to test our hypotheses for pre- and post-violence recall. We first hypothesized that those exposed to video violence would recall less about information presented before the violence than would those who did not see the violence, but pre-violence viewing found no main effect for the viewing [F = (1, 73) = .47, p £ .50]. The second hypothesis proposed that those who are desensitized to violence and exposed to the video violence would recall more about information presented before the violence than would those not desensitized. However, we found no interaction between the stimulus condition and level of desensitization [M = 1.3, SD = 1.2; M = 1.2, SD = 1.4 (t = .17 (37), p £ .44)]. (See Table 5 and Figure 2.)
Additional analyses for recall before the violence indicated no significance between groups in terms of low desensitization level [M = 1.3, SD = 1.2; M = 1.4, SD = 1.3 (t = .21 (32), p £ .42)]. Similar findings resulted when the means were compared for both groups based on high desensitization [M = 1.2, SD = 1.4; M = 1.5, SD = 1.1 (t = .74 (39), p £ .24)]. Comparison of means within the control group in terms of desensitization level also yielded no significance [M = 1.4, SD = 1.3; M = 1.5, SD = 1.1 (t = .32 (34), p £ .38)]. (See Table 5 and Figure 2.)
Our third hypothesis stated that for those viewing violent stimuli, recall for information presented after the stimuli would be greater than for those who did not see the stimuli. Analysis of variance found no main effect for post-violence viewing or for desensitization [F (1, 73) = .34, p £ .56].
The fourth hypothesis suggested that those who are desensitized to video violence and exposed to the violence would recall more after the violence than would those not desensitized. There was no significance within the experimental group in terms of desensitization level [M = .4, SD = .7; M = .6, SD = .7 (t = 1.13 (37), p £ .14)]. (See Table 5 and Figure 3.)
However, additional analyses for the recall means after the violence indicated significance for two of the three comparison groups. Significance was found between groups among highly desensitized members [M = .6, SD = .7; M = .12, SD = .7 (t = 1.93 (39), p £ .03)] as well as within the control group when comparisons were based on level of desensitization [M = .7, SD = 1.1; M = .2, SD = .7 (t = 1.72 (34), p £ .05)]. No significance was found for the recall measure after the violence between groups among subjects with low levels of desensitization [M = .4, SD = .7; M = .7, SD = 1.1 (t = 1.10 (32), p £ .14)]. (See Table 5 and Figure 3.)
This study presented threats to both internal and external validity. One such threat to internal validity was instrumentation. First, subjects who viewed the non-violent clip were taken to another building on campus. The 3-minute walk could have elevated physiological arousal levels in the control group before viewing the film clip, while the experimental group did not receive such pre-experiment arousal. The different levels of physiological arousal between the two groups of subjects could have affected subjects' subsequent recall of film details.
Second, subjects were administered the study in two locations and the film clips were played on different screens using different video projection units. Moreover, the images projected by the two video machines varied slightly in color. This could have affected items in the survey that contained references to color.
Third, a couple of questions were answered correctly by nearly all subjects in both conditions, while one was not answered correctly by any subjects in either condition. Sufficient variance on all questions might have yielded more significant findings.
An external validity issue present in this study is generalizability. Replication of the study is needed before we can confidently generalize our findings. One key element that should be examined is the interaction effect found for post-violence viewing recall. More data is needed to help explain the finding that shows non-desensitized subjects who do not view a violent stimulus have better recall than those desensitized to violence but who are not presented the violent stimulus.
As the results indicate, a main effect was not found for recall of information before violence viewing, nor was there an interaction with desensitization. In addition, recall of information after the stimulus produced no main effect. However, there was an interaction between violence viewing and desensitization for post-violence recall measures.
Three of the post-violence viewing cells mirror our hypotheses. We expected subjects who were desensitized to violence to have greater recall than those not desensitized. We also expected those who were not desensitized to violence and who were exposed to the violent stimulus to score lower on the recall measure. However, those subjects not exposed to the violent stimulus and not desensitized to violence scored highest on the recall measure.
One explanation for such a result might be that those subjects who are not desensitized and who did not receive the violent stimulus are not regular movie watchers. Perhaps these subjects were aroused by the movie clip, despite its lack of violence, which enhanced their ability to rehearse the information. Such an occurrence is in accord with excitation theory. When subjects are exposed to arousing material it enhances their ability to process information because their senses are heightened. Since the material was non-threatening there was little chance the state of arousal would have reached a point at which the subjects' mental processes would have ceased to function properly. Support for this explanation was found when means for desensitized and non-desensitized subjects in the control group were compared (M = .5, SD = .2; M = .6, SD = .1 (t = 1.94 (34), p £ .03). Furthermore, subjects in the non-desensitized group on the average watch less movies (M = 21.1, SD = 5.9) than those who are desensitized (M = 29.5, SD = 4.4 (t= 7.09 (73), p £ .001). This result offers additional support to the notion that the non-violent film clip was arousing to the non-desensitized subjects because they do not see as many movies as those desensitized.
Another possible explanation might be that these subjects are lower thrill seekers than those desensitized to violence, and they might generally tend to process information more efficiently than their counterparts in the experimental group.
Perhaps the desensitized nature of the subjects requires a significant level of arousal that was absent in the control stimulus. As mentioned above, an arousing stimulus can heighten the senses, which in turn may produce greater recall. However, when a stimulus is not arousing to an individual, such as the non-arousing stimuli presented in the control group, desensitized individuals may have even more trouble recalling information because they do not achieve a state of arousal that heightens their senses and mental processes necessary to encode information in their memory. The implications of this finding may have a dramatic effect on the learning process. Educators may have to resort to scaring their students to get them to learn something.
SUGGESTIONS FOR FUTURE RESEARCH
This study leads us to ask several questions. How do we increase learning for high violence viewers? Do we present these thrill seekers with information/stimuli that increases their fear and arousal levels prior to presenting the information we want recalled? As for non-thrill seekers, if we need to arouse these individuals, how can we do so without frightening and violent stimuli?
These questions also raise important social concerns. How can we expect to teach young people in a society where violence in the movies/media prevails? As noted earlier, will we have to scare the living daylights out of today's 10-year old raised on video games and media to get him/her to recall anything? Should learners be segmented by thrill seekers and non-thrill seekers? If our hypotheses hold true, we could be in for a "frightening" future as the level of fright needed for arousal and learning increases.
There are several possibilities for future research in this area. First, replication of this study could control for internal and external threats to validity more effectively and possibly provide more accurate results. Both experimental and control groups could be tested in the same room, using the same screen and video machine. In addition, recall questions with little or no variance should be removed and new questions added.
Second, this study only addressed movie (i.e., fictional) violence. Future research could focus on real-world violence. Such an investigation could provide information as to whether the perception of violence as real or fictional affects viewer recall.
Bandura, A., Ross, D., & Ross, S. A. (1963). Imitation of film-mediated aggressive models. Journal of Abnormal and Social Psychology, 66, 3-11.
Bandura, A. (1994). Social cognitive theory of mass communication. In J. Bryant & D. Zillmann (Eds.), Media Effects: Advances in Theory and Research (pp. 61-90). Hillsdale, NJ: Lawrence Erlbaum Associates.
Berkowitz, L. & Rawlings, E. (1963). Effects of film violence on inhibitions against subsequent aggression. Journal of Abnormal and Social Psychology, 66, 405-412.
Cantor, J., Mody, B., & Zillmann, D. (1974). Residual emotional arousal as a distractor in persuasion. Journal of Social Psychology, 92, 231-244.
Cline, V. B., Croft, R. G., & Courrier, S. (1973). Desensitization of children to television violence. Journal of Personality and Social Psychology, 27, 360-365.
Committee on Communications, 1994 to 1995. (1995). Media violence. Pediatrics, 95, 949-951.
Comstock, G. (1989). The evolution of American television. Newbury Park, CA: Sage Publications.
Comstock, G. & Strasburger, V. (1993). Media violence: Q & A. Adolescent Medicine: State of the Art Review, 4, 495-509.
Jo, E., & Berkowitz, L. (1994). A priming effect analysis of media influences: An update. In J. Bryant & D. Zillmann (Eds.), Media Effects: Advances in Theory and Research (pp. 43-60). Hillsdale, NJ: Lawrence Erlbaum Associates.
Gadow, K., & Sprafkin, J. (1989). Field experiments of television violence with children: Evidence for an environmental hazard? Pediatrics, 83, 399-405.
Gerbner, G. (1992). Violence in cable-oriented television programs: A report to the national cable television association. Philadelphia, PA: Annenberg School for Communication.
Gerbner, G., Morgan, M., & Signorielli, N. (1994). Television violence profile no. 16: The turning point. Philadelphia, PA: Annenberg School for Communication.
Katz, E., Adoni, H., & Parness, P. (1977). Remembering the news: What the picture adds to recall. Journalism Quarterly, 54 (2), 231-239.
Kellermann, K. (1985). Memory processes in media effects. Communication Research: An International Quarterly, 12 (1), 83-131.
Loftus, E., & Burns, T. (1982). Mental shock can produce retrograde amnesia. Memory & Cognition, 10 (4), 318-323.
Mundorf, N., Drew, D., Zillmann, D., & Weaver, J. (1990). Effects of disturbing news on recall of subsequently presented news. Communication Research, 17 (5), 601-615.
Newhagen, J., & Reeves, B. (1992). The evening's bad news: Effects of compelling negative television news images on memory. Journal of Communication, 42 (2), 25-42.
Prasad, V., & Smith, L. (1994). Television commercials in violent programming: An experimental evaluation of their effects on children. Journal of the Academy of Marketing Science, 22 (4), 340-351.
Scott, R. K., & Goff, D. H. (1988). How excitation from prior programming affects television news recall. Journalism Quarterly, 65 (2), 615-620.
Shapiro, M. A., & Lang, A. (1991). Making television reality: Unconscious processes in the construction of social reality. Communication Research, 18 (5), 685-705.
Singer, D. (1989). Children, adolescents, and television-1989: I. Television violence: A critique. Pediatrics, 83, 445-446.
Thomas, M. H., Horton, R. W., Lippincott, E. C., & Drabman, R. S. (1977). Desensitization to portrayals of real life aggression as a function of exposure to television violence. Journal of Personality and Social Psychology, 35, 450-458.
U.S. Department of Health and Human Services: Public Health Service, Alcohol, Drug Abuse, and Mental Health Administration: National Institute of Mental Health. (1982). Television and behavior: Ten years of scientific progress and implications for the eighties. Washington, DC: U.S. Government Printing Office. U.S. Department of Health and Human Services publication ADM 82-1195, vol. 1.
U.S. Surgeon General's Scientific Advisory Committee on Television and Social Behavior. (1972). Television and Growing Up: The Impact of Televised Violence: Report to the Surgeon General, U.S. Public Health Service. Washington, DC: U.S. Government Printing Office.
Warrington, E. K., & Weiskrantz, L. (1970). Amnesiac syndrome: Consolidation or retrieval? Nature, 228, 628-630.
Wharton, R., & Mandell, F. (1985). Violence on television and imitative behavior: Impact on parenting practices. Pediatrics, 75, 1120-1123.© B.L. Yates 2000