Induced Aggressive Mood And

Induced Aggressive Mood and Explicit Memory

MICHAEL J. LANG The purpose of these 2 studies is to examine whether an aggressive mood (induced by playing a violent video game) biases a person’s recall of aggressive and John Carroll University nonaggressive words. Study 1 examined the effects that violent video games have on the player and observer. Study 2 examined the effects of an induced aggressive mood on a person’s recall of a list of aggressive and nonaggressive words. In both studies, participants were randomly assigned to play a violent video game, which consisted of fighting a computer opponent using martial arts, or to play a nonviolent video game, which involved racing a high-performance car on a racetrack. Overall, in Study 1, the participants who were exposed to the violent video game had a greater feeling of aggression than the participants who were exposed to the nonviolent video game. Overall, in Study 2, the participants recalled significantly more aggressive words than nonaggressive words, but participants who played the violent video game did not recall significantly more aggressive words than those who played the nonviolent video game.

OOD CONGRUENT MEMORY (MCM) REFERS TO unpleasant words, the unpleasant words just pop out the tendency of a person in a particular at them. Mmood to recall information that is congru- Research with depressed participants supports ent with that mood (Christianson, 1992; Mayer, Bower’s (1981) selective attention theory. Based on McCormick, & Strong, 1995; Watkins, Vache, Verney, the results of two studies, McDowall (1984) concluded Muller, & Mathews, 1996). For example, a person that depressed people do not have a problem recall- whose mood is depressed is more likely to recall in- ing pleasant words when presented with only pleas- formation with an unpleasant meaning than infor- ant words; however, when they are presented with mation that has a pleasant meaning. both pleasant and unpleasant words, they selectively Although extensive research has examined MCM, attend to mood-congruent material and recall more the bulk of it has involved depressed participants. For unpleasant words than pleasant words. example, studies show that depressed participants are Berkowitz (1973) looked for MCM with a differ- more likely to recall words that are congruent to their ent kind of mood, aggression. Elementary-age chil- mood (Denny & Hunt, 1992; Ruiz-Caballero & dren selected a word to ﬁnish an incomplete sentence Gonzalez, 1994; Watkins, Mathews, Williamson, & before and after they read a comic book. Children Fuller, 1992). either read a war comic (aggressive condition) or read Additional support for MCM has been shown by a neutral comic (nonaggressive condition). Berkowitz participants who were induced to be either sad or found that the children who read the war comic re- happy (Gilligan & Bower, 1983). These researchers trieved more aggressive words from memory to ﬁn- found that participants whose moods were induced ish their incomplete sentences than the children who to be happy recalled more happy phrases than par- read the neutral comic. Bower (1981) also showed ticipants whose moods were induced to be sad. that angry participants recalled more aggressive words Bower (1981) proposed selective attention as an than participants with other moods. explanation for MCM and suggested that participants Researchers have reported MCM using explicit actively attend to information or material that is con- memory with depressed participants (Denny & Hunt, gruent to their moods. For example, when a person 1992; Ruiz-Caballero & Gonzalez, 1994; Watkins et with a depressed mood looks at a list of pleasant and al., 1992) and implicit memory with aggressive par-

PSI CHI JOURNAL OF UNDERGRADUATE RESEARCH Winter 1999 149 Copyright 1999 by Psi Chi, The National Honor Society in Psychology (Vol. 4, No. 4, 149–154 / ISSN 1089-4136). Faculty Supervisor: Janet D. Larson INDUCED AGGRESSIVE MOOD Lang ticipants (Berkowitz, 1973). MCM involving explicit high-performance race car on a racetrack. For this memory with participants in an aggressive mood has experiment the player only attempted to drive the not been examined. car at the highest speed possible in order to avoid Feelings of aggression may be induced by having any violent content of the video game. people play violent video games (Irwin & Gross, 1995; Measures. Participants completed two question- Schutte, Malouff, Post-Gorden, & Rodasta, 1988). Spe- naires. The first questionnaire assessed the participant’s cifically, Irwin and Gross (1995) found that children age, how long he had been playing video games, and who played a violent video game displayed signifi- how many hours a week he spent playing video games. cantly more physical aggression toward objects and a The State-Trait Anger Expression Inventory (STAXI; confederate than did the children who played the Spielberger, 1988) was used to measure participants’ nonviolent video game. Schutte et al. (1988) found feeling of aggression. The state anger scale that as- that children who played a violent video game were sesses how a person is feeling at the time they are more aggressive afterwards than children who played filling out the questionnaire was used for this study. a nonviolent video game. Despite minimal research Coefficient alpha reliability for this sample of 40 men on the effects of observing a video game, researchers was .76. A comparison sample of 44 college age male are interested in the effects of observing television participants was used to gather normative data on the on aggression. Boyatzis, Matillo, and Nesbitt (1995) STAXI for students at this institution. and Harris (1992) found that observing violent tele- Procedure. Half of the participants were ran- vision can increase aggression in the viewer. Because domly assigned to play or observe a violent video observing video games is similar to observing televi- game, whereas the remainder of the participants were sion, researchers can extrapolate potential hypoth- randomly assigned to play or observe a nonviolent eses with regard to observing video games from stud- video game. Participants were run in pairs, so the ies of the effects of watching television. Because most order in which they arrived determined which par- video games are for one player, there is often some- ticipant played or observed the video game. Upon one observing the video game being played. arrival the participants filled out the first questionnaire. Participants in the player condition then played Experiment 1 the video game for 8 min while participants in the In order to test the hypothesis that an aggressive observer condition observed the video game being mood can effect memory, a study was conducted to played. If the participant in the player condition was examine video games’ potential for inducing an ag- unfamiliar with the video game, he was given a 2 min gressive mood in male college students who played practice session. During this practice session the ob- them or observed them being played. It was expected server also watched the video game being played. that the participants who were exposed to a violent Prior to playing the video game for 8 min, both the video game would have a greater feeling of aggres- player and the observer were told they could talk sion than the participants who were exposed to a freely but that the observer could not touch the con- nonviolent video game. It was further predicted that trol pad. At the end of the 8 min of play the partici- participants who played the violent video game would pants completed the STAXI. have a greater feeling of aggression than participants who observed the violent video game. Results Three one-way analyses of variance (ANOVAs) Method were performed to examine differences between the Participants. Forty male students between the four experimental conditions with respect to age, ages of 18 and 22 (M = 19.5) from a midwestern uni- hours spent each week playing video games, and the versity participated. Participants were members of an number of years playing video games (see Table 1 for introductory psychology class and received credit for means). Age, F(3, 36) = 1.46, p = .24, hours spent play- their participation. ing video games, F(3, 36) = 2.71, p = .06, or years Apparatus. A Sony Playstation connected to a 19- playing video games, F(3, 36) = .57, p = .64, were not inch (48-cm) color television monitor was used to play significantly different between conditions. the video games. Two different high-action video A 2 × 2 (Type of Video Game × Involvement) games were used for this experiment. The violent ANOVA was performed to test the hypothesis there video game was Namco’s Tekken 2 (1996), in which would be an effect for type of video game on the feel- the player fights a computer opponent using martial ing of aggression. Table 2 shows the cell means and arts. The nonviolent video game was Electronic Arts’ standard deviations for each condition. There was a Need for Speed II (1997), in which the player races a main effect for type of video game, F(1, 36) = 12.80,

TABLE 1 Descriptive Statistics of the Participants: Experiment 1

Mean Hours spent each week Average years playing age playing video games video games Play violent 19.60 7.10 10.90 Video game (1.17) (5.00) (3.57) Observe violent 19.70 4.90 8.90 Video game (0.95) (5.95) (3.87) Play nonviolent 18.90 2.15 9.90 Video game (0.88) (2.77) (3.63) Observe nonviolent 19.80 2.80 8.90 Video game (1.23) (2.40) (4.84)

Note. Standard deviations are in parentheses. N = 10 per condition. p = .001, with the participants exposed to the violent Discussion video game having greater feelings of aggression than The participants who were exposed to the vio- those participants exposed to the nonviolent video lent video game had a greater feeling of aggression game. The main effect for involvement, F(1, 36) = .33, than the participants who were exposed to the non- p = .57, and the significant interaction, F(1, 36) = 2.05, violent video game. Furthermore, the participants p = .16, failed to achieve significance. A planned com- who played the violent video game rated their feel- parison using a t test on the STAXI scores for the ings of aggression significantly higher than the stan- participants who were exposed to the violent video dardization sample. This finding is consistent with game failed to show a significant difference between past research (Anderson & Ford, 1986; Irwin & Gross, the player and the observer on feeling of aggression, 1995; Nelson & Carlson, 1985; Schutte et al., 1988). t(18) = 1.05, p = .31. The participants who played the nonviolent video A one-sample t test was performed to compare game rated their feeling of aggression significantly each of the player conditions to the standardization lower than the participants in the standardization sample (M = 11.57, SD = 2.19) on the STAXI. The sample. There was not, however, a significant differ- participants who played the violent video game scored ence between the players of the violent video game significantly higher than the standardization sample, and the observers of the violent video game on their t(9) = 2.33, p = .027, and the participants who played feeling of aggression. the nonviolent video game scored significantly lower, It appears that the nonviolent video game had a t(9) = –1.67, p = .001, on feelings of aggression than calming effect on the player whereas the violent video the standardization sample. game induced aggression in the player. In research

TABLE 2 State-Trait Anger Scale Mean Scores: Experiment 2

Violent video game Nonviolent video game Marginal means MSDMSDMSD Player of video game 13.90a 2.81 10.40a .70 12.15 1.76 Observer of video game 12.50 3.14 11.00 1.15 11.75 2.15

Marginal means 13.20a 2.98 10.70a .93

Note. N = 10 per condition. aMean scores signiﬁcantly different at the .05 level.

TABLE 3 Means and Standard Deviations of the Participants’ Age and Video Game Experience: Experiment 2

Mean Hours spent each week Average years playing age playing video games video games Played violent 19.41 2.08 10.08 Video game (0.99) (1.83) (4.52)

Played nonviolent 19.20 4.73 9.33 Video game (1.08) (5.19) (2.50)

Note. Standard deviations are in parentheses. N = 15 per condition.

conducted by Nelson and Carlson (1985), partici- ceived credit toward a class requirement for their pants overwhelmingly preferred playing nonviolent participation. video games to playing violent video games. The calm- Materials. A pilot study was conducted in order ing effects of nonviolent video games could be one to establish what words undergraduates think are reason why participants would rather play nonviolent aggressive. A list of 40 words (20 aggressive and 20 video games than violent video games. nonaggressive) was compiled based on the participants’ judgment of each word’s meaning and the Experiment 2 word frequency of each word (see Appendix for word The second study examined whether an aggres- list). Aggressive and nonaggressive words were not sive mood biases a person’s recall of aggressive and signiﬁcantly different with regard to word frequency, nonaggressive words. Based on past research (Berko- t(38) = 1.32, p = .21. witz 1973; Bower 1981), it was expected that partici- Apparatus. The same apparati were used as in pants whose induced mood was aggressive would Experiment 1. A 13-inch (33-cm) computer monitor recall more aggressive words than nonaggressive was used to display the list of 40 words selected dur- words from a list studied after mood manipulation. ing the pilot study. A second hypothesis was that the participants whose Procedure. Participants were randomly assigned induced mood was aggressive would recall more to play one of the video games. Before the partici- aggressive words than the participants whose mood pants played a video game they were asked to ﬁll out was induced to be nonaggressive. a questionnaire that assessed their video game–playing history. If the participants were unfamiliar with Method the video game, they were give a 2-min practice ses- Participants. Thirty male students between the sion to familiarize themselves with the controls. Af- ages of 18 and 22 (M = 19.31) participated in the ter completion of the 8-min video game–playing ses- study. The majority of the participants were mem- sion, participants studied a list of 40 words (20 ag- bers of an introductory psychology class, and they re- gressive and 20 nonaggressive) in preparation for a

TABLE 4 Average Aggressive and Nonaggressive Words Recalled: Experiment 2

Induced aggressive mood Induced nonaggressive mood Marginal means Word type MSDMSDMSD Aggressive 5.00 1.41 6.00 2.36 5.50a 1.89 Neutral 3.40 1.30 4.20 1.94 3.80a 1.62

Marginal means 4.20 1.36 5.10 2.15

Note. N = 15 per condition. aMean scores signiﬁcantly different at the .05 level.

152 PSI CHI JOURNAL OF UNDERGRADUATE RESEARCH Winter 1999 Copyright 1999 by Psi Chi, The National Honor Society in Psychology (Vol. 4, No. 4, 149–154 / ISSN 1089-4136). INDUCED AGGRESSIVE MOOD Lang recall test. Each word was presented in random or- sive words is that the participants naturally attended der on a computer screen for 3 s. Each time a new to the aggressive words due to a history of being ex- word appeared on the screen, participants were told posed to violent and aggressive material via contem- to say it aloud. After the participants studied the words porary media. Violence and aggression are recurrent they counted backwards from 72 by 3. After the themes in newspapers and in the movies and televi- distractor task, participants were given 2 min to write sion programs we watch. Perhaps due to the constant down as many words as they could from the words exposure to violence and aggression, the participants’ presented on the computer screen. moods were already induced to be aggressive before the start of the experiment. If the participants had Results aggressive moods prior to the beginning of the ex- Three t tests were performed to examine differ- periment, then it would have biased their recall of ences between the video-game conditions with respect information, thus providing one possible explanation to age, hours spent each week playing video games, as to why, regardless of video game played, they re- and the number of years playing video games (see called more aggressive words than nonaggressive Table 3 for means). Age was not significantly differ- words. ent between the conditions, t(28) = .60, p = .54. Hours The word list is a major limitation to this study. spent each week playing video games was not signifi- Although the aggressive and nonaggressive words cantly different between conditions, t(28) = 1.53, p = were matched on frequency, they were not, however, .11. Years playing video games was not significantly matched on meaning. Perhaps the reason why par- different between conditions, t(28) = .59, p = .55. ticipants, regardless of video game played, recalled A 2 × 2 (Type of Video Game × Type of Word more aggressive words than nonaggressive words is Recalled) ANOVA with word type as a within-subjects because the aggressive words were more interesting variable was performed on the number of aggressive and thus easier to imagine than the nonaggressive and nonaggressive words recalled. Table 4 shows the words. Any meaning attributed to the results of this cell means and the standard deviations for each con- study must be made cautiously due to the limitations dition. There was a main effect for type of word re- presented by the word list. called, F(1, 28) = 17.59, p < .01. The main effect for Berkowitz (1973) found evidence of MCM by test- type of video game, F(1, 28) = 3.01, p = .094, and the ing children’s implicit memory; children finished interaction between type of mood induction and type more incomplete sentences with aggressive words af- of word recalled, F(1, 28) < 1, were not significant. ter reading a war comic than children who read a Two planned comparisons based on the hypoth- neutral comic. Further research involving participants eses were performed. The participants who played whose moods are induced to be aggressive might look the violent video game recalled significantly more for evidence of MCM by testing participants’ implicit aggressive words than nonaggressive words, t(28) = memory. Mayer et al. (1995) used a category-retrieval 3.23, p < .05; however, the participants who played task to test implicit memory in depressed patients. the violent video game and the participants who This task involved having the participants respond to played the nonviolent video game differed signifi- a certain word (i.e., marriage, divorce, or love) by cantly on the number of aggressive words recalled, listing a word that began with a certain letter. Future t(28) = 1.41, p = .17. research in the area of MCM involving participants whose moods are induced to be aggressive could test General Discussion implicit memory using a category-retrieval task or Because participants who played the violent video some other test of implicit memory. game recalled significantly more aggressive words than nonaggressive words, the results appear to show sup- References port for the first hypothesis. However, the difference Anderson, C. A., & Ford, C. M. (1986). Affect of the game player: in recall was due to the fact that, regardless of video Short-term effects of highly and mildly aggressive video games. Personality and Social Psychology Bulletin, 12, 390–402. game played, participants recalled significantly more Berkowitz, L. (1973). Words and symbols as stimuli to aggressive aggressive words than nonaggressive words. Also, the responses. In J. F. Knutson (Ed.), The control of aggression: results did not support the second hypothesis because Implications from basic research (pp. 113–143). Chicago: Aldine. Bower, G. H. (1981). Mood and memory. American Psychologist, participants who played the violent video game did 36, 129–148. not recall significantly more aggressive words than Boyatzis, C. J., Matillo, G. M., & Nesbitt, K. M. (1995). Effects of those who played the nonviolent video game. the “Mighty Morphin Power Rangers” on children’s aggression with peers. Child Study Journal, 25, 45–55. A possible explanation for why participants, over- Christianson, S. A. (Ed.). (1992). The handbook of emotion and all, recalled more aggressive words than nonaggres- memory: Research and theory. Hillsdale, NJ: Lawrence Erlbaum.

Denny, E. B., & Hunt, R. R. (1992). Affective valence and memory choice of arcade games and their consequences upon young in depression: Dissociation of recall and fragment completion. male players. Journal of Applied Social Psychology, 15, 124–139. Journal of Abnormal Psychology, 101, 575–580. Ruiz-Caballero, J. A., & Gonzalez, P. (1994). Implicit and explicit Gilligan, S. G., & Bower, G. H. (1983). Reminding and mood- memory bias in depressed and nondepressed subjects. Cogni- congruent memory. Bulletin of the Psychonomic Society, 21, 431– tion and Emotion, 8, 555–569. 434. Schutte, N. S., Malouff, J. M., Post-Gorden, J. C., & Rodasta, A. L. Harris, M. B. (1992). Television viewing, aggression, and ethnicity. (1988). Effects of playing videogames on children’s aggres- Psychological Reports, 70, 137–138. sive and other behaviors. Journal of Applied Social Psychology, 18, Irwin, A. R., & Gross, A. M. (1995). Cognitive tempo, violent video 454–460. games, and aggressive behavior in young boys. Journal of Family Spielberger, C. D. (1988). State-Trait Anger Expression Inventory Violence, 10, 337–350. (STAXI). Odessa, FL: PAR/Psychological Assessment Resources. Mayer, J. D., McCormick, L. J., & Strong, S. E. (1995). Mood- Watkins, P. C., Mathews, A., Williamson, D. A., & Fuller, R. D. congruent memory and natural mood: New evidence. Person- (1992). Mood-congruent memory in depression: Emotional ality and Social Psychology Bulletin, 21, 736–746. priming or elaboration? Journal of Abnormal Psychology, 101, 581– McDowall, J. (1984). Recall of pleasant and unpleasant words 586. in depressed subjects. Journal of Abnormal Psychology, 93, 401– Watkins, P. C., Vache, K., Verney, S. P., Muller, S., & Mathews, A. 407. (1996). Unconscious mood-congruent memory bias in depres- Nelson, T. M., & Carlson, D. R. (1985). Determining factors in sion. Journal of Abnormal Psychology, 105, 34–41.

APPENDIX

Aggressive and Nonaggressive Word List

Aggressive Words Nonagressive Words

Aggression Abstract Assault Colorless Annihilate Indifferent Ransacked Relaxed Decapitate Unexcessive Pulverize Unextreme Eradicate Disinterest Exterminate Customary Destruction Inactive Obliterate Tranquil Massacre Dullness Violence Slumber Ferocious Impartial Slaughter Melancholy Decimation Impersonal Torture Tolerance Devastation Peaceful Execute Unrufﬂed Warfare Pleasant Conquer Sluggish