Personality and Individual Differences 45 (2008) 618–623

Contents lists available at ScienceDirect

Personality and Individual Differences

journal homepage: www.elsevier .com/locate /paid

How impulsiveness, trait anger, and extracurricular activities might affectaggression in school children

Marc Wittmann a,*, Estibaliz Arce a, Carmen Santisteban b

a Department of Psychiatry, University of California San Diego, 9500 Gilman Drive #9116A, La Jolla, CA 92093-9116, USAb Instituto de Estudios Biofuncionales, Universidad Complutense de Madrid, Spain

a r t i c l e i n f o

Article history:Received 13 February 2008Received in revised form 24 June 2008Accepted 1 July 2008Available online 19 August 2008

Keywords:ImpulsivenessAggressionAngerSex differencesDevelopmental differences

0191-8869/$ - see front matter � 2008 Elsevier Ltd. Adoi:10.1016/j.paid.2008.07.001

* Corresponding author. Tel.: +1 (858) 552 8585x53E-mail address: wittmann@ucsd.edu (M. Wittman

a b s t r a c t

Individual differences in personality traits such as impulsivity and trait anger as well as environmentalvariables have an impact on aggressiveness. We tested a model incorporating the related variables ofimpulsiveness, trait anger, and aggression and incorporated the possible mediating influences of lei-sure-time activities. Regression analyses of data from 1129 pre-adolescents and 1093 adolescents(55.4% females) from a study evaluating the Spanish version of the Buss and Perry aggression question-naire (AQ; Santisteban, Alvarado, & Recio, 2007) showed a relation between impulsiveness (BarrattImpulsiveness Scale) and aggression (AQ). Trait anger (Spielberger’s State-Trait Anger Expression Inven-tory), the time spent watching TV and playing video games, and the time spent doing homework (allrelated to impulsiveness) also are related to physical, as well as verbal aggression (with low to moderatecoefficients). Multiple mediation analyses confirm that media violence exposure and homework, respec-tively, can have aggravating and attenuating effects on self-reported aggression. These results providekey variables for longitudinal studies which could reveal the causal nature of the results found withour cross-sectional design.

� 2008 Elsevier Ltd. All rights reserved.

1. Introduction

Despite the considerable amount of evidence accumulating onthe effects of media violence on aggressive behavior, controversyon this topic still exists (Anderson & Bushman, 2002a; Kutner & Ol-son, 2008). The influence of the time spent watching television onaggressive behavior has been widely reported in children, adoles-cents, and adults (Anderson & Bushman, 2002b; Bartholow et al.,2005; Bushman & Huesmann, 2006; Johnson, Cohen, Smailes,Kasen, & Brook, 2002; Joy, Kimball, & Zabrack, 1986; Santisteban,Alvarado, & Recio, 2007). Another form of entertainment that hasacquired wide acceptance among youngsters is playing videogames, which has also been related to aggressive behavior (Funk,Baldacci, Pasold, & Baumgardner, 2004; Gentile, Lynch, Linder, &Walsh, 2004; Santisteban et al., 2007). Nevertheless, the scientificcommunity is still debating the influence that TV and video gamesexert on people’s behavior (see, for example, Kids, TV viewing, andaggressive behavior (Letters). Science 297, July 5, 2002). Thosewho oppose the influential role of TV and video games on aggres-sive behavior claim that causality cannot easily be establishedsince violent individuals are more likely to watch aggressive actson TV and perform them in video games (Olson, 2004). An attemptto reconcile both views is the notion of a bidirectional relationship

ll rights reserved.

88; fax: +1 (858) 642 1429.n).

between media violence and aggressive behavior, where both cau-sal directions contribute to the associations found.

Aggression can be defined as behavior intended to harm anindividual when it is clear that he or she wants to avoid beingharmed (Joireman, Anderson, & Strathman, 2003). Apparentaggressive behavior can only be explained by integrating a multi-tude of factors. For example, the general aggression model(GAM), encompasses three main factors that are active in any so-cial interaction: (1) the individual and situational inputs, (2) theinternal states of cognition, emotion, and arousal, and (3) the out-comes of decision processes that lead to either thoughtful orimpulsive action (Anderson & Bushman, 2002b). Focusing on per-sonality traits, a recent variant of this model postulates thataggressive behavior is founded on individual differences, such asimpulsiveness, sensation seeking, and considering future conse-quences (Joireman et al., 2003).

The multidimensionality of impulsiveness has resulted in agreat variety of behavioral and self-report instruments that, how-ever, often lack significant inter-correlations (Arce & Santisteban,2006; Wittmann & Paulus, 2008). One definition we shall use hereis that impulsiveness is a behavior defined as responding to a stim-ulus without appropriately evaluating its consequences (Gerbing,Ahadi, & Patton, 1987). In the context of aggression, the lack ofimpulse control is regarded as one determinant of aggressivebehavior (Vigil-Colet & Codorniu-Raga, 2004). As opposed to non-impulsive aggressive acts, which are planned, premeditated, and

M. Wittmann et al. / Personality and Individual Differences 45 (2008) 618–623 619

instrumental for social gain and dominance, impulsive aggressionis spontaneous, unprovoked, and out of proportion Barratt, Stand-ford, Dowdy, Liebman, and Kent (1999).

A main characteristic of the GAM model by Joireman and col-leagues (2003) is that individual differences in hostility and angermediate the relationship between personal traits, such as impul-siveness or sensation seeking, and aggression. Thus, only individu-als who are anger prone will translate impulsive urges into verbalor physical aggression. In extending this three-part structure ofaggressive behavior (impulsiveness ? trait anger ? aggression)and applying it to pre-adolescents and adolescents, we attemptedto include variables that can promote or reduce aggression and arerelated to activities after school. Environmental factors, such asexposure to violent media, can change the knowledge structureof an individual (perceptual and person schemata, behavioralscripts), produce a desensitization effect in the long run, and leadto more aggressive behavior (Anderson & Bushman, 2002b).

Besides the aggression-promoting influence of virtual violence,effects of aggression reduction through certain mediators are possi-ble. Studies reveal that antisocial behavior and aggression are neg-atively correlated with academic achievement (Barriga et al., 2002;Williams & McGee, 1994). Aggression has also been related to lowreading abilities (Barrera et al., 2002; Davis, Byrd, Arnold, Auinger,& Bocchini, 1999). Impulsive children have poorer selective atten-tion, deficient reading skills, and show more often aggressive socialbehavior (Thompson, Teare, & Elliott, 1983). Furthermore, it hasspecifically been shown that impulsiveness scores are negativelyrelated to success at school, e.g., doing well in reading and mathe-matics tests (Merrell & Tymms, 2001; Vigil Colet & Morales Vives,2005). Thus, the current literature suggests that the time spentdoing homework and leisure reading is associated with lowerimpulsiveness and aggression. Keeping in mind the bi-directional-ity of this association, it is possible that pursuing these activitiesat home could attenuate the effects of impulsiveness on aggression.

In our model depicted in Fig. 1, we assume that, aside from an ef-fect of impulsiveness on aggression, anger-trait will have an influ-ence on aggression. In addition, exposure to media (TV, videogames) may aggravate aggression, whereas doing homework andreading books presents an attenuating influence. Impulsivity, onthe other hand, might be related to anger as a trait as well as affectingactivities done at home. More impulsive individuals will be morelikely to watch TV and play video games and be less inclined to dotheir homework and read books. We were specifically interestedin developmental aspects of the discussed relationships (differences

a4, a5

a1, a2, a3

M6: hours/week homewM7: hours/week readin

M1: hours TV workdayM2: hours TV weekendM3: hours/wk video ga

M4: STAXI anger traitM5: STAXI anger cont

X: BISSum score

Impulsiveness

c’

a6, a7

Fig. 1. Positive and negative effects in the relationships between impulsivity and aggresvideo games, doing homework, and reading books. The coefficients {ai} and {bi} are the inddirect path coefficient between X and Y.

between adolescents and pre-adolescents), as well as in replicatingand expanding knowledge of sex differences in the assessed vari-ables, especially in aggression, i.e., that men are more readily in-volved in physical aggressive behavior than women (Bjorkqvist,1994; Ramirez, Santisteban, Fujihara, & van Goozen, 2002).

2. Method

2.1. Participants

Participants consisted of 2222 pre-adolescents and adolescentsbetween the ages of 9 and 17 (55.4% females) recruited and testedin 27 schools in the Madrid area. Permission to test the studentswas sought from the participating schools. The parents were in-formed by the schools prior to the study and asked whether theirchildren were allowed to participate. Using a stratified samplingmethod and then randomly selecting school sites, strata were iden-tified according to the criteria of (1) private or state school, (2)location (metropolitan area, dormitory towns, rural environment),(3) size of population at location and school-age population size.For data analyses, children aged between 9 and 11 were consideredpre-adolescents (n = 1129, 46.1% male, 53.9% female), and thosebetween the ages of 14 and 17 were assigned to the group of ado-lescents (n = 1093, 43% male, 57% female). All participants com-pleted the self-report questionnaires in their classrooms.

2.2. Instruments

After filling out a socio-demographic questionnaire, the partici-pants were asked to indicate the approximate number of hours perday they spent watching television on a weekday and on a week-end day, separately, and how many hours per day they played vi-deo games. They also reported the number of hours per dayspent doing homework and the number of hours devoted to extra-curricular reading per week.

We used the Spanish version (Recio, Santisteban, & Alvarado,2004) of the Barratt Impulsiveness Scale (BIS-11) that consists of30 items that can be grouped into a total score and 3 subscales:Non-planning impulsiveness (‘‘I plan tasks carefully”, ‘‘I changejobs”), motor impulsiveness (‘‘I do things without thinking”, ‘‘Ibuy things on impulse”), and attention/cognitive impulsiveness(‘‘I concentrate easily”, ‘‘I get easily bored when solving thoughtproblems”) (Patton, Stanford, & Barratt, 1995).

b6, b7ork g

ssmes

rol

Y1: Agression - physical Y2: Aggression - verbal

b1, b2, b3

b4, b5

sion. Possible mediating effects are anger-trait, spending time watching TV, playingirect path coefficients for the effects of X on M and M on Y, respectively. The c0 is the

Table 1Age-group differences (pre-adolescents vs. adolescents) and effect sizes

Variable Pre-adolescents

Adolescents t df = 2220(max)

p Effect sizeCohen’s d

Mean score Mean score

BIS motor 18.26 20.35 �10.04 0.0001*** �0.457BIS attention/

cognition17.01 18.64 �9.71 0.0001*** �0.437

BIS non-planning

24.63 27.31 �15.52 0.0001*** �0.716

STAXI anger-trait

18.10 22.01 �16.57 0.0001*** �0.705

STAXI anger-control

53.47 53.92 �1.17 0.242

AQ – physical 18.81 20.05 �4.17 0.0001*** �0.177AQ – verbal 10.12 12.03 �11.91 0.0001*** �0.506Hours TV

weekday1.95 2.53 �12.85 0.0001*** �0.560

Hours TVweekend

2.98 3.33 �5.93 0.0001*** �0.260

Hours/day videogames

3.10 3.18 �1.42 0.157

Hours/dayhomework

2.85 3.15 �6.49 0.0001*** �0.279

Hours/weekreadingbooks

2.72 2.45 5.92 0.0001*** 0.254

Note: Significant coefficients: ***p < 0.001; significant p values adjusted to 12 sta-tistical tests would be: p < 0.00416.

Table 2Sex differences and effect sizes

Variable Male Female t df = 2220(max)

p Effect sizeCohen’s dMean

scoreMeanscore

BIS motor 19.49 19.24 1.19 0.236BIS attention/

cognition18.09 17.66 2.46 0.014* 0.111

BIS non-planning 26.57 25.65 5.01 0.0001*** 0.233STAXI anger-trait 19.98 20.15 �.662 0.508STAXI anger-

control53.33 54.19 �2.23 0.026* �0.096

AQ – physical 21.66 17.64 13.70 0.0001*** 0.586AQ – verbal 11.41 10.80 3.67 0.0001*** 0.157Hours TV

weekday2.28 2.21 1.46 0.143

Hours TV 3.16 3.15 .247 0.805

620 M. Wittmann et al. / Personality and Individual Differences 45 (2008) 618–623

Aggression was measured with the Spanish translation of theBuss and Perry aggression questionnaire (AQ) adapted for pre-ado-lescents and adolescents (Buss & Perry, 1992; Santisteban et al.,2007). The validated Spanish version for adolescents and pre-ado-lescents of the questionnaire maintains the structure of the originalAQ (Santisteban, Recio, Alvarado, & Leenen, 2004). The 29 items inthis questionnaire provide a global measure of aggression in foursubscales: physical aggression (PA) with 9 items, verbal aggression(VA) with 5 items, anger (A) with 7 items, and hostility (H) with 8items. To probe our model we used the subscales of physical andverbal aggression, which represent outward aggressive behaviors.

A Spanish version of Spielberger’s State-Trait Anger ExpressionInventory (STAXI) was used to measure anger (del Barrio, Aluja, &Spielberger, 2004). On three four-point rating scales participantsindicated the extent to which each item describes their experienceof anger at this moment (anger state), in general (anger-trait), andwhen they get upset (control and expression of anger). We usedthe ‘trait anger’ and the ‘anger-control and expression’ subscalesto test our hypotheses.

2.3. Statistics

We first calculated t tests for the age group (pre-adolescents vs.adolescents) and sex differences on the variables of interest. Wethen conducted a series of regression analyses – separately forthe subgroups – to test for the effects between the variables inour model. We finally performed a mediation analysis over all sub-jects to test the extent to which mediator variables transmit the as-sumed effect of impulsiveness on aggression. Since we wereconstrained by our cross-sectional analysis, we did this only fordescriptive purposes.

A variable can be considered a mediator M if it carries the influ-ence of a given independent variable X to a given dependent vari-able Y. In our model of multiple mediation effects (Fig. 1), a = {ai},b = {bi} and c0 represent the path coefficients for the effects ofX ? M, M ? Y, and X ? Y, respectively. The path coefficient c0 rep-resents the direct path, whereas (ai,bi) represent the indirect paths.The total effect c, i.e., the initial effect of X on Y when mediators arenot introduced in the model, is the sum of the indirect (ai,bi) anddirect effects (c0). We used the multiple mediator approach pro-posed by Preacher and Hayes (2004), including a bootstrappingprocedure (n = 5000, confidence intervals to be set at 95% and99%) for testing the indirect effects.

weekendHours/day video

games3.42 2.72 12.66 0.0001*** 0.671

Hours/dayhomework

2.79 3.19 �8.51 0.0001*** �0.370

Hours/weekreading books

2.53 2.64 �2.36 0.018* �0.102

Note: Significant coefficients: *p < 0.05, ***p < 0.001; significant p values adjusted to12 statistical tests: p < 0.00416.

3. Results

3.1. t Tests for group comparisons

As reflected in Table 1, adolescents score higher than pre-ado-lescents in most of the variables. Adolescents are more impulsive,report more anger, and are more physically and verbally aggres-sive. They are also to a greater extent involved in the assessed vari-ables of how they spend time at home (watching TV, doinghomework, reading books). Regarding sex differences (Table 2),male subjects show more non-planning impulsivity and scorehigher on the physical and verbal aggression subscales. Male sub-jects spend more time playing video games, and female subjectsspend more time doing homework.

3.2. Regression analyses

In pre-adolescents, BIS motor and BIS attention/cognition (high-er impulsivity predicts more aggression), as well as STAXI anger-trait (more anger-trait results in more aggression) and STAXIanger-control (more anger-control contributes to less verbal

aggression) significantly affect the verbal aggression score (for de-tails, see Table 3). Physical aggression in pre-adolescents is pre-dicted by BIS motor and BIS non-planning and STAXI anger-traitas well as STAXI anger-control. In addition, the amount of videogames played predicts physical aggression.

In adolescents, motor impulsivity and STAXI anger-trait signifi-cantly predict verbal aggression (see Table 3). More hours spentdoing homework is related to less verbal aggression and morehours spent watching video games predicts more verbal aggres-sion. Physical aggression is predicted by a greater amount ofvariables in adolescents. BIS motor, BIS non-planning, STAXI an-ger-trait, STAXI anger-control, as well as the hours spent doing

Table 3Significant regression coefficients (b) for pre-adolescents and adolescents includingall the independent variables of impulsivity, anger, extracurricular activities on thedependent variables of verbal and physical aggression

Variables Pre-adolescents Adolescents

Verbalaggression

Physicalaggression

Verbalaggression

Physicalaggression

BIS motor 0.232*** 0.186*** 0.167** 0.129**

BIS attention/cognition

0.139**

BIS non-planning 0.159*** 0.156***

STAXI anger-trait 0.326*** 0.285*** 0.465*** 0.354***

STAXI anger-control

�0.194** �0.179***

Hours TV weekdayHours TV weekendHours/day video

games0.166*** 0.091* 0.107**

Hours/dayhomework

�0.102* �0.158***

Hours/weekreading books

R2 0.382*** 0.425*** 0.328*** 0.390***

Note: Significant coefficients: *p < 0.05, **p < 0.01, ***p < 0.001.

Table 4Significant regression coefficients (b) for male and female participants including allthe independent variables of impulsivity, anger, extracurricular activities on thedependent variables of verbal and physical aggression

Variables Male Female

Verbalaggression

Physicalaggression

Verbalaggression

Physicalaggression

BIS motor 0.135** 0.170*** 0.288*** 0.178***

BIS attention/cognition

BIS non-planning 0.155*** 0.126*

STAXI anger-trait 0.411*** 0.371*** 0.434*** 0.328***

STAXI anger-control

�0.152*** �0.142***

Hours TV weekday 0.103*

Hours TV weekendHours/day video

games0.086* 0.109**

Hours/dayhomework

�0.162***

Hours/weekreading books

R2 0.310*** 0.386*** 0.471*** 0.427***

Note: Significant coefficients: *p < 0.05, **p < 0.01, ***p < 0.001.

Table 5Summary of mediation analyses incorporating the model variables across all subjects

Independentvariable X

Mediating variableM1–Mj

Dependentvariable Y

Effect of X onM (a)

Effect of M oY (b)

Barratt TV weekdays (h) AQ - physical 0.0272*** 0.3764Impulsiveness TV weekends (h) 0.0201*** �0.3565*

Scale Video games (h) 0.0247*** 0.9730***

Homework (h) �0.0156 �0.6426***

Sum score Reading books (h) �0.0286*** �0.3047STAXI trait 0.3061*** 0.4167***

STAXI control �0.0481 �0.1238***

TV weekdays (h) AQ – verbal 0.0272*** �0.1122TV weekends (h) 0.0201*** 0.2093*

Video games (h) 0.0247*** 0.2761**

Homework (h) �0.0156*** �0.1223Reading books (h) �0.0286*** 0.1068STAXI trait 0.3061*** 0.3028***

STAXI control �0.0481 �0.0193

Note: Significant coefficients: *p < 0.05, **p < 0.01, ***p < 0.001; the coefficients are not sta

M. Wittmann et al. / Personality and Individual Differences 45 (2008) 618–623 621

homework and playing video games contribute to physicalaggression.

In male participants (Table 4), BIS motor and STAXI anger-traitaffect significantly verbal aggression. More variables affect physi-cal aggression in the young males. BIS motor, BIS non-planning,STAXI anger-trait, STAXI anger-control as well as the hours watch-ing TV and playing video games influence physical aggression. Fe-male participants’ verbal aggression is significantly related tomotor impulsivity, STAXI anger-trait, and the hours spent playingvideo games. Physical aggression in females is influenced by BISmotor and BIS non-planning, STAXI anger-trait, STAXI anger-con-trol, and the amount of homework done per day.

3.3. Mediation analyses

Mediation analyses (Table 5) reveal significant indirect effectsbetween impulsiveness and both physical and verbal aggression(ai and bi paths) for the mediator trait anger and the number ofhours playing video games (aggravating effect). In addition, doinghomework functions as a mediator (attenuating effect) betweenimpulsiveness and physical aggression and watching TV on theweekends is a mediator (aggravating effect) on verbal aggression.Impulsiveness also has a direct influence (c0) on aggression. Thus,the total effect c is the sum of a direct effect of impulsiveness onaggression and the revealed indirect effects (partial mediation).

4. Discussion

The aim of this study was to test a model containing severalvariables related to aggression in a representative sample of pre-adolescents (9–11 years) and adolescents (14–17 years). We foundthat trait anger had the strongest relationship with aggression.Impulsivity also showed to influence verbal and physical aggres-sion. In addition, the way leisure-time is spent relates to aggres-sion. The more time is spent playing video games and watchingTV, the higher the aggression scores. The fact that playing violentvideo games has a stronger influence (more consistent over differ-ent subgroups) on aggression than watching TV can be explainedby the fact that in video games the player has an active role in dis-playing forms of aggression, such as harming or killing virtualopponents, as opposed to the more passive role of watching TV(Dill & Dill, 1998). On the other hand, the time spent doing home-work might diminish aggression. Doing homework requires a con-siderable amount of self control as the urge to choose immediategratification (i.e., playing, socializing) has to be subdued inthe interest of more delayed rewards (scholarly success) (Mischel,

n Indirecteffect (ab)

Total indirecteffect (ab)

Directeffect (c0)

Totaleffect (c)

Degree ofmediation

0.0102 0.1794** 0.1937*** 0.3731*** Partial�0.0072

0.0240**

0.0101**

0.00870.1276**

0.0060

�0.0030 0.1005** 0.0817*** 0.1821*** Partial0.0042*

0.0068**

0.0019�0.0031

0.0927**

0.0009

ndardized values.

622 M. Wittmann et al. / Personality and Individual Differences 45 (2008) 618–623

Shoda, & Rodriguez, 1989). The reason why homework, but notreading, is related to aggression can probably be understood bythe fact that doing homework usually requires more effort and selfcontrol. As opposed to the way books are often read, i.e., they arechosen for entertainment and pleasure, feedback on quality of per-formance is continuously provided on homework.

Complementing these findings with mediation analyses appliedover all subjects, we found that next to the direct effect betweenimpulsiveness and aggression also indirect effects occur. Trait an-ger, the strongest mediator, is considered a decisive factor deter-mining the expression of aggressive behavior (Joireman et al.,2003). In addition, the way leisure-time is spent has a mediatingeffect on aggression. The more time spent playing video gamesand watching TV, the greater the impact of impulsiveness onaggression. The time spent doing homework diminishes the effectthat impulsiveness has on aggression.

Our results coincide with previous reports linking aggressionand impulsiveness, aggression and media exposure, and betweenimpulsiveness and academic achievements (for references, see Sec-tion 1). Moreover, this study provides new evidence in combiningall these factors and shows that impulsivity and trait anger as wellas the time spent doing particular activities after school can influ-ence aggression.

Next to these general trends across subgroups, we also foundsome age-related discrepancies, i.e., the way free time is spenthas a stronger impact in adolescents as compared to younger chil-dren. While for the pre-adolescents, only the time spent playing vi-deo games related to physical aggression (but not to verbalaggression), for adolescents the hours spent playing video gamesand doing homework influenced physical as well as verbal aggres-sion. This may reflect the fact that they have been indulging inthese activities over a longer period of time, leading to long-termbehavioral changes. Alternatively, a developmental interpretationwould state that in puberty adolescents, who show more aggres-sion than pre-adolescents, might also be more susceptible tobehavioral modulation.

In addition to a developmental effect, a gender effect was det-ected. In the male group, activities after school (i.e., watching TVon weekdays and playing video games) affect physical aggression,but not verbal aggression. In the female group, verbal aggression isaffected by the time spent watching videos and physical aggressionis influenced by the amount of homework. Studies have repeatedlyshown that males score higher on physical and (with a smaller dif-ference) verbal aggression than females (Archer, 2004; Buss &Perry, 1992; von Collani & Werner, 2005). Our study also foundthat values for physical aggression and verbal aggression were sig-nificantly higher in males than in females. Moreover, the strongerinfluence of extracurricular activities on physical as compared toverbal aggression points to the fact that physical aggression playsa more prominent role in males and that this factor may also bemore easily influenced.

Naturally, there are limitations to our study. (1) With ourcross-sectional design we are unable to reach strong conclusionsregarding the causal effects proposed. Nevertheless, the contextof available empirical evidence gathered in longitudinal studies,which show long-term effects of media exposure on aggression,supports the interpretation of the effects found in our study.Playing video games might increase aggression in individuals.Doing homework might have an attenuating effect on aggression.(2) Our chosen inventory for aggression is based on self-reports.However, the AQ by Buss and Perry (1992) has proven to havehigh ecological validity by predicting real-life aggressive behav-ior (Bushman & Wells, 1998). (3) We only registered the timespent on the leisure activities and did not qualitatively assesswhich type of TV programs or video games children and adoles-cents were exposed to. The assessment of additional information

on video-game content might have led to more specific results,differentiating the impact of levels of violence in entertainmentmedia.

However, the claim that video games cause aggression in chil-dren has to be regarded with caution. First of all, the regressioncoefficients in our study indicative of this relationship are smallto (at most) moderate. A majority of children in industrializednations play video games (to a great proportion of violent nature)but most of them do not become more aggressive. Children arecapable of making the distinction between simulation and reality(Kutner & Olson, 2008; Olson et al., 2007). Individuals in their com-plex social context have to be identified who are at risk of beingnegatively affected by violent video games.

Regarding the prevention of aggression as related to impulsive-ness, our data point to two variables, which are not trait related(such as impulsivity and anger) and influence aggression: theamount of time spent playing video games and doing homework.The present findings on the possibly enhancing and diminishing ef-fects of certain variables on aggression provide useful informationfor future intervention studies that control for the impact certainacademic and leisure-time activities have on problematic behaviorin individual children and adolescents.

Acknowledgments

This work was supported by the Spanish Ministerios de AsuntosSociales AS (RS/MS2001-16/01) and Educación y Ciencia MEC(SEJ2004-05290).

References

Anderson, C. A., & Bushman, B. J. (2002a). The effects of media violence on society.Science, 295, 2377–2379.

Anderson, C. A., & Bushman, B. J. (2002b). Human aggression. Annual Review ofPsychology, 53, 27–51.

Arce, E., & Santisteban, C. (2006). Impulsivity: A review. Psicothema, 18, 213–220.Archer, J. (2004). Sex differences in aggression in real-world settings: A meta-

analytic review. Review of General Psychology, 8, 291–322.Barratt, E. S., Standford, M. S., Dowdy, L., Liebman, M. J., & Kent, T. A. (1999).

Impulsive and premeditated aggression: A factor analysis of self-reported acts.Psychiatry Research, 86, 163–173.

Barrera, M., Biglan, A., Taylor, T. K., Gunn, B., Smolkowsky, K., Black, C., et al. (2002).Early elementary school intervention to reduce conduct problems: Arandomized trial with Hispanic and non-Hispanic children. Prevention Science,3, 83–94.

Barriga, A. Q., Doran, J., Newell, S. R., Morrison, E. M., Barbetti, V., & Robbins, B. D.(2002). Relationships between problem behaviors and academic achievementin adolescents: The unique role of attention problems. Journal of Emotional andBehavioral Disorders, 10, 233–240.

Bartholow, B. D., Sestir, M. A., & Davis, E. B. (2005). Correlates and consequences ofexposure to video game violence: Hostile personality, empathy, and aggressivebehavior. Personality and Social Psychology Bulletin, 31, 1573–1586.

Bjorkqvist, K. (1994). Sex differences in physical, verbal, and indirect aggression:A review of recent research. Sex Roles, 30, 177–188.

Bushman, B. J., & Huesmann, C. A. (2006). Short-term and long-term effects ofviolent media on aggression in children and adults. Archive of Pediatric andAdolescence Medicine, 160, 348–352.

Bushman, B. J., & Wells, G. L. (1998). Trait aggressiveness and hockey penalties:Predicting hot tempers on the ice. Journal of Personality and Social Psychology, 83,969–974.

Buss, A. H., & Perry, M. (1992). The aggression questionnaire. Journal of Personalityand Social Psychology, 63, 452–459.

Davis, T. C., Byrd, R. S., Arnold, C. L., Auinger, P., & Bocchini, J. A. (1999). Low literacyand violence among adolescents in a summer sports program. Journal ofAdolescent Health, 24, 403–411.

del Barrio, V., Aluja, A., & Spielberger, Ch. (2004). Anger assessment with the STAXI-CA: Psychometric properties of a new instrument for children and adolescents.Personality and Individual Differences, 37, 227–244.

Dill, K. E., & Dill, J. C. (1998). Video game violence: A review of the empiricalliterature. Aggression and Violent Behavior, 3, 407–428.

Funk, J. B., Baldacci, H. B., Pasold, T., & Baumgardner, J. (2004). Violence exposure inreal-life, video games, television, movies, and the internet: Is theredesensitisation? Journal of Adolescence, 27, 23–39.

Gentile, D. A., Lynch, P. J., Linder, J. R., & Walsh, D. A. (2004). The effect of violentgame habits on adolescent hostility, aggressive behaviors, and schoolperformance. Journal of Adolescence, 27, 5–22.

M. Wittmann et al. / Personality and Individual Differences 45 (2008) 618–623 623

Gerbing, D. W., Ahadi, S. A., & Patton, J. H. (1987). Towards a conceptualization ofimpulsivity: Components across the behavioural and self-report domains.Multivariate Behavioural Research, 22, 357–379.

Johnson, J. G., Cohen, P., Smailes, E. M., Kasen, S., & Brook, J. S. (2002). Televisionviewing and aggressive behavior during adolescence and adulthood. Science,295, 2468–2471.

Joireman, J., Anderson, J., & Strathman, A. (2003). The aggression paradox:Understanding links among aggression, sensation seeking, and theconsideration of future consequences. Journal of Personality and SocialPsychology, 84, 1287–1302.

Joy, L. A., Kimball, M. M., & Zabrack, M. L. (1986). Television and children’saggressive behavior. In T. M. Williams (Ed.), The impact of television: A naturalexperiment in three communities (pp. 303–360). New York: Academic Press.

Kutner, L. K., & Olson, C. K. (2008). Grand theft childhood: The surprising truth aboutviolent video games and what parents can do. New York: Simon & Schuster.

Merrell, C., & Tymms, P. B. (2001). Inattention, hyperactivity and impulsiveness:Their impact on academic achievement and progress. The British Journal ofEducational Psychology, 71, 43–56.

Mischel, W., Shoda, Y., & Rodriguez, M. (1989). Delay of gratification in children.Science, 244, 933–938.

Olson, C. K. (2004). Media violence research and youth violence data: Why do theyconflict. Academic Psychiatry, 28, 144–150.

Olson, C. K., Kutner, L. K., Warner, D., Almerigi, J. B., Baer, L., Nicholi, A. M., et al.(2007). Factors correlated with violent video game use by adolescent boys andgirls. Journal of Adolescent Health, 41, 77–83.

Patton, J. H., Stanford, M. S., & Barratt, E. S. (1995). Factor structure of the Barrattimpulsiveness scale. Journal of Clinical Psychology, 51, 768–774.

Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirecteffects in simple mediation models. Behavior Research Methods, Instruments, andComputers, 36, 717–731.

Ramirez, J. M., Santisteban, C., Fujihara, T., & van Goozen, S. (2002). Differencesbetween experience of anger and readiness to angry action. Aggressive Behavior,28, 429–438.

Recio, P., Santisteban, C., & Alvarado, J. M. (2004). Estructura factorial de unaadaptación española del test de impulsividad de Barratt. Revista de Metodologíade las Ciencias del Comportamiento, Suplemento 2004, 517–521.

Santisteban, C., Alvarado, J. M., & Recio, P. (2007). Evaluation of a Spanish version ofthe Buss and Perry aggression questionnaire: Some personal and situationalfactors related to the aggression scores of young subjects. Personality andIndividual Differences, 42, 1453–1465.

Santisteban, C., Recio, P., Alvarado, J. M., & Leenen, I. (2004). Comparación de laestructura factorial con una clasificación jerárquica en una escala deagresividad. Revista de Metodología de las Ciencias del Comportamiento,Suplemento 2004, 549–555.

Thompson, R. W., Teare, J. F., & Elliott, S. N. (1983). Impulsivity: From theoreticalconstructs to applied interventions. The Journal of Special Education, 17,157–169.

Vigil-Colet, A., & Codorniu-Raga, M. J. (2004). Aggression and inhibition deficits, therole of functional and dysfunctional impulsivity. Personality and IndividualDifferences, 37, 1431–1440.

Vigil Colet, A., & Morales Vives, F. (2005). How impulsivity is related to intelligenceand academic achievement. The Spanish Journal of Psychology, 8, 199–204.

von Collani, G., & Werner, R. (2005). Self-related and motivational constructs asdeterminants of aggression. An analysis and validation of a German version ofthe Buss–Perry aggression questionnaire. Personality and Individual Differences,38, 1631–1643.

Williams, S., & McGee, R. (1994). Reading attainment und juvenile delinquency.Journal of Child Psychology and Psychiatry, 35, 442–459.

Wittmann, M., & Paulus, M. P. (2008). Decision making, impulsivity and timeperception. Trends in Cognitive Sciences, 12, 7–12.