Neuroimaging and behavioral evidence that violent video games exert no negative effect on human empathy for pain and emotional reactivity to violence
- Social, Cognitive and Affective Neuroscience Unit, Department of Cognition, Emotion, and Methods in Psychology, Faculty of Psychology, University of Vienna, Austria
- Department of Clinical Neuroscience, Division of Psychology, Karolinska Institute, Sweden
Figures
Figure 1
Schematic depiction of the experimental tasks.
(A) Empathy-for-pain task. In trials of the Self condition, participants passively received electrical stimuli. In the Other condition, participants experienced how another person (a confederate) received electrical stimuli. The stimuli were either painful or not painful. In the cue phase, an arrow indicated the recipient (downwards: Self; right: Other) and the intensity (blue: not painful; red: painful) of the next stimulus. In the stimulation phase, the stimulus was delivered. After half of the trials, participants were asked to rate the last stimulus. The confederate depicted has given informed consent that his photograph can be published. (B) Emotional reactivity task. Participants were presented pictures with different content (violent or neutral) and different context (real or game context). After observing a block of pictures, participants rated their current unpleasantness on a visual analog scale from 0 to 100.
Figure 2
Behavioral results.
Depicted are participants’ ratings during the empathy-for-pain task (A and B) and the emotional reactivity task (C). Ratings were given on a visual analog scale (range: 0–100). (A) Empathy for pain, painfulness rating. Question text: ‘How painful for the other?’. (B) Empathy for pain, unpleasantness rating. Question text: ‘How unpleasant for yourself?’. Note that an apparent trend toward a three-way interaction Group*Session*Intensity is not supported by the respective Bayesian hypothesis test (BF = 0.130, Table 1). (C) Emotional reactivity, unpleasantness rating. Question text: ‘How unpleasant?’. Boxes: the middle line marks the group mean of participant ratings in the respective condition; the box represents the 95% credible interval of the posterior predictive distribution of mean ratings. Dots depict the individual mean ratings of participants, lines depict the 95% credible interval of the posterior predictive distribution of mean ratings of single participants. Control game group: N = 44. Violent game group: N = 45.
Figure 3
Results of the whole-brain analyses for region-of-interest definition.
(A) Empathy-for-pain task. Clusters represent areas where brain activity was increased when the confederate received a painful electrical stimulus, compared to a non-painful stimulus. (B) Emotional reactivity task. Clusters represent areas where brain activity was increased during the observation of violent images, compared to neutral images. All results p<0.05 FWE-corrected. This figure was made with the software MRICron (https://www.nitrc.org/projects/mricron).
Tables
Table 1
Posterior parameter means of models for ratings in the empathy-for-pain task.
Dependent variable: empathy ratings (visual analog scale, range: 0–100). Factor codings: Group: control game group = –1, violent game group = 1; Session: first session = –1, second session = 1; Intensity: non-painful = –1, painful = 1. Bayes factors were derived from comparing a model where the respective parameter was unrestricted to a model where it was restricted to zero. †These Bayes factors were derived from comparing a model where the parameter was restricted to be negative to a model where it was restricted to zero (one-sided hypothesis test).
| Fixed effect | β | 95% Credible interval | Bayes factor | |
|---|---|---|---|---|
| A) Painfulness ratings | ||||
| Group | 0.66 | –1.14 | 2.46 | 0.127 |
| Session | 0.42 | –0.31 | 1.16 | 0.072 |
| Intensity | 27.86 | 25.32 | 30.29 | >100 |
| Group*Session | –0.48 | –1.24 | 0.25 | 0.102 |
| Group*Intensity | –1.12 | –3.48 | 1.32 | 0.207 |
| Session*Intensity | –0.26 | –1.38 | 0.83 | 0.069 |
| Group*Session*Intensity | –0.78 | –1.87 | 0.32 | 0.324† |
| B) Unpleasantness ratings | ||||
| Group | –1.13 | –4.83 | 2.56 | 0.251 |
| Session | –0.63 | –1.94 | 0.71 | 0.141 |
| Intensity | 17.48 | 14.93 | 20.11 | >100 |
| Group*Session | –0.93 | –2.33 | 0.40 | 0.254 |
| Group*Intensity | –0.95 | –3.63 | 1.68 | 0.197 |
| Session*Intensity | –1.20 | –2.12 | –0.30 | 0.996 |
| Group*Session*Intensity | –0.45 | –1.36 | 0.49 | 0.130† |
Table 2
Posterior parameter estimates of models for ratings in the emotional reactivity task.
Dependent variable: unpleasantness ratings (visual analog scale, range: 0–100). Factor codings: Group: control game group = –1, violent game group = 1; Content: neutral = –1, violent = 1; Context: real = –1, game = 1. Bayes factors were derived from comparing a model where the respective parameter was unrestricted to a model where it was restricted to zero. †These Bayes factors were derived from comparing a model where the parameter was restricted to be negative to a model where it was restricted to zero (one-sided hypothesis test).
| Fixed effect | β | 95% Credible interval | Bayes factor | |
|---|---|---|---|---|
| Group | 1.26 | –2.78 | 5.44 | 0.349 |
| Content | 37.08 | 32.98 | 41.48 | >100 |
| Context | –7.24 | –9.15 | –5.39 | >100 |
| Group*Content | 2.28 | –1.92 | 6.52 | 0.151† |
| Group*Context | –1.23 | –3.01 | 0.47 | 0.306 |
| Content*Context | –5.36 | –7.39 | –3.34 | >100 |
| Group*Content*Context | 0.33 | –1.58 | 2.10 | 0.094† |
Table 3
Comparison of trait empathy levels between experimental group and control group.
Bayes factors were derived from comparing a model where the mean difference violent video game (VVG) group – Control group was positive to a model where it was restricted to zero (one-sided Bayesian t-test).
| VVG group (N=83) | Control group (N=132) | |||||
|---|---|---|---|---|---|---|
| QCAE subdimension | Mean | SD | Mean | SD | t | Bayes factor |
| Perspective Taking | 1.93 | 0.43 | 2.01 | 0.51 | –1.189 | 0.074 |
| Online Simulation | 1.96 | 0.41 | 1.92 | 0.47 | 0.588 | 0.259 |
| Emotional Contagion | 1.58 | 0.48 | 1.63 | 0.55 | –0.659 | 0.098 |
| Peripheral Responsivity | 1.58 | 0.58 | 1.58 | 0.62 | 0.021 | 0.155 |
| Proximal Responsivity | 1.67 | 0.58 | 1.79 | 0.53 | –1.538 | 0.063 |
Table 4
Results of the Bayes factor design analysis.
Depicted are the estimated probabilities of inferential decisions for each dependent variable and assumed true effect size d. Inc.: Inconclusive evidence, no decision. H0: evidence for the null hypothesis. H1: evidence for the alternative hypothesis. ρ=correlation between repeated measurements, i.e., test-retest reliability. The estimated probabilities of correct decisions (evidence for H0 when d=0.0, evidence for H1 when d>0.0) are marked in bold.
| Empathy | Emotional reactivity | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Painfulness ratings | Unpleasantness ratings | Neural response | Unpleasantness ratings | |||||||||
| (ρ=0.75) | (ρ=0.90) | (ρ=0) | (only second session) | |||||||||
| Effect size | Inc. | H0 | H1 | Inc. | H0 | H1 | Inc. | H0 | H1 | Inc. | H0 | H1 |
| d=0.0 | 0.29 | 0.69 | 0.02 | 0.29 | 0.69 | 0.02 | 0.30 | 0.69 | 0.02 | 0.30 | 0.68 | 0.02 |
| d=0.2 | 0.58 | 0.20 | 0.23 | 0.44 | 0.05 | 0.50 | 0.49 | 0.43 | 0.08 | 0.55 | 0.33 | 0.13 |
| d=0.3 | 0.48 | 0.06 | 0.46 | 0.14 | 0.00 | 0.85 | 0.56 | 0.30 | 0.14 | 0.57 | 0.18 | 0.25 |
| d=0.4 | 0.28 | 0.02 | 0.71 | 0.02 | 0.00 | 0.98 | 0.57 | 0.21 | 0.22 | 0.50 | 0.08 | 0.42 |
Table 5
Cross-task correlations.
Above diagonal: posterior means of correlations. Below diagonal: 95% credible intervals of correlations. Unpl.=unpleasantness. Emo.Reac.=emotional reactivity. aMCC = anterior midcingulate cortex. AI = anterior insula. Amy = amygdala. l.=left. r.=right.
| Emp: Pain | Emp: Unpl. | ER: Unpl. | Emp: aMCC | Emp: lAI | Emp: rAI | ER: lAmy | ER: rAmy | ||
|---|---|---|---|---|---|---|---|---|---|
| Behavior | Empathy: Pain | 0.630 | 0.280 | 0.015 | 0.043 | 0.133 | –0.037 | –0.037 | |
| Empathy: Unpl. | (0.544,0.708) | 0.227 | 0.039 | 0.084 | 0.211 | 0.058 | 0.096 | ||
| Emo. Reac.: Unpl. | (0.191,0.366) | (0.130,0.323) | –0.010 | 0.028 | –0.028 | 0.060 | 0.043 | ||
| Neural | Empathy: aMCC | (–0.178,0.215) | (–0.170,0.245) | (–0.212,0.190) | 0.055 | 0.069 | 0.040 | 0.050 | |
| Empathy: l. AI | (–0.151,0.235) | (–0.118,0.276) | (–0.167,0.214) | (–0.201,0.285) | 0.104 | 0.019 | 0.021 | ||
| Empathy: r. AI | (–0.056,0.310) | (–0.009,0.393) | (–0.205,0.153) | (–0.156,0.289) | (–0.146,0.323) | 0.080 | 0.085 | ||
| Emo. Reac.: l. Amy | (–0.178,0.096) | (–0.079,0.195) | (–0.080,0.202) | (–0.164,0.241) | (–0.172,0.213) | (–0.124,0.266) | 0.507 | ||
| Emo. Reac.: r. Amy | (–0.173,0.098) | (–0.042,0.230) | (–0.095,0.180) | (–0.150,0.251) | (–0.171,0.207) | (–0.117,0.271) | (0.370,0.628) |
Appendix 1—table 1
Results of whole-brain analyses for region of interest (ROI) definition, empathy-for-pain task.
Tested contrast: Other Pain – Other No Pain, data only taken from the first session. We report the first local maximum within each cluster. Effects were tested for significance with a significance threshold of p<0.05, FWE-corrected. We only report clusters larger than 10 voxels.
| MNI coordinates | |||||
|---|---|---|---|---|---|
| Brain region | x | y | z | z-Value | Cluster size |
| R Insula | 30 | 22 | –14 | 6.15 | 27 |
| R Inferior frontal gyrus, orbital part | 50 | 26 | –6 | 6.29 | 125 |
| R Inferior frontal gyrus, opercular part | 50 | 18 | 8 | 5.50 | 23 |
| L Insula | –34 | 20 | 0 | 6.35 | 343 |
| L Superior frontal gyrus, medial part | –2 | 34 | 34 | 6.32 | 86 |
| L Supramarginal gyrus | –58 | –56 | 30 | 6.26 | 110 |
| L Inferior parietal lobule | 58 | –56 | 40 | 5.69 | 12 |
Appendix 1—table 2
Results of whole-brain analyses for region of interest (ROI) definition, emotional reactivity task.
Tested contrast: Violent – Neutral. We report the first local maximum within each cluster. Effects were tested for significance with a significance threshold of p<0.05, FWE-corrected. We only report clusters larger than 10 voxels.
| MNI coordinates | |||||
|---|---|---|---|---|---|
| Brain region | x | y | z | z-Value | Cluster size |
| L/R Thalamus | -6 | –28 | -6 | >10 | 1405 |
| L/R Inferior occipital lobe | –40 | –68 | –10 | >10 | 30,408 |
| R Amygdala | 20 | 0 | –14 | 5.64 | 26 |
| L Amygdala | –20 | –2 | –12 | 6.09 | 13 |
| L/R Hypothalamus | 4 | –4 | –10 | 6.41 | 27 |
| L Precentral gyrus | –44 | 4 | 30 | >10 | 521 |
| R Precentral gyrus | 46 | 8 | 32 | 8.21 | 1229 |
| R Caudate | 14 | 12 | 10 | 5.80 | 33 |
| L Inferior frontal gyrus, triangular part | –42 | 34 | 16 | 7.05 | 159 |
| R Supramarginal gyrus | 66 | –22 | 28 | 6.11 | 87 |
| L Midcingulum | 0 | 4 | 34 | 6.70 | 34 |
| R Supplementary motor area | 6 | 14 | 62 | 6.54 | 116 |
| L Cerebellum | –22 | –38 | –42 | 7.47 | 42 |
| R Cerebellum | 24 | –34 | –42 | 6.97 | 45 |
Appendix 2—table 1
Posterior parameter estimates and contrasts of models for functional magnetic resonance imaging (fMRI) signal in the empathy-for-pain task.
Dependent variable: fMRI signal extracted from the respective region of interest (ROI) (standardized to unit variance): Fixed effect: terms in standard font describe the mean regression parameter of the respective event averaged across all conditions. Terms in italic font describe the fixed effect of the respective condition on the regression parameter. Factor codings: Group: control game group = –1, violent game group = 1; Session: first session = –1, second session = 1; Intensity: non-painful = –1, painful = 1. β/σe: Mean model parameter divided by the mean error standard deviation. Bayes factors were derived from comparing a model where the respective parameter was unrestricted to a model where it was restricted to zero. †These Bayes factors were derived from comparing a model where the parameter was restricted to be negative to a model where it was restricted to zero (one-sided hypothesis test).
| aMCC | Left AI | Right AI | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Fixed effect | β/σe | 95% CI | BF | β/σe | 95% CI | BF | β/σe | 95% CI | BF |
| Self cue | 0.25 | (0.09, 0.41) | 34.770 | 0.16 | (0.00, 0.32) | 1.925 | 0.33 | (0.20, 0.45) | >100 |
| Group | 0.03 | (–0.13, 0.20) | 0.264 | –0.01 | (–0.17, 0.16) | 0.260 | 0.03 | (–0.10, 0.16) | 0.257 |
| Intensity | 0.14 | (0.03, 0.26) | 3.494 | 0.33 | (0.20, 0.45) | >100 | 0.41 | (0.31, 0.51) | >100 |
| Session | 0.08 | (–0.03, 0.19) | 0.582 | 0.10 | (–0.01, 0.21) | 1.054 | 0.05 | (–0.06, 0.15) | 0.301 |
| Group*Intensity | –0.01 | (–0.13, 0.10) | 0.201 | –0.02 | (–0.14, 0.11) | 0.209 | –0.01 | (–0.11, 0.09) | 0.191 |
| Group*Session | 0.05 | (–0.05, 0.16) | 0.276 | 0.00 | (–0.11, 0.10) | 0.179 | –0.03 | (–0.13, 0.07) | 0.236 |
| Intensity*Session | –0.02 | (–0.11, 0.08) | 0.182 | –0.08 | (–0.18, 0.02) | 0.577 | –0.01 | (–0.09, 0.07) | 0.150 |
| Group*Intensity*Session | –0.03 | (–0.12, 0.07) | 0.193 | –0.04 | (–0.14, 0.06) | 0.253 | –0.03 | (–0.11, 0.05) | 0.241 |
| Other cue | 0.41 | (0.26, 0.56) | >100 | 0.06 | (–0.11, 0.22) | 0.299 | 0.13 | (0.00, 0.26) | 1.459 |
| Group | 0.11 | (–0.04, 0.26) | 0.627 | 0.03 | (–0.14, 0.21) | 0.292 | 0.09 | (–0.04, 0.22) | 0.601 |
| Intensity | 0.14 | (0.05, 0.24) | 8.814 | 0.19 | (0.10, 0.28) | >100 | 0.20 | (0.12, 0.28) | >100 |
| Session | 0.00 | (–0.11, 0.10) | 0.175 | 0.08 | (–0.02, 0.20) | 0.576 | 0.07 | (–0.02, 0.16) | 0.596 |
| Group*Intensity | 0.00 | (–0.10, 0.09) | 0.161 | 0.02 | (–0.07, 0.11) | 0.165 | –0.01 | (–0.09, 0.07) | 0.169 |
| Group*Session | –0.05 | (–0.16, 0.05) | 0.307 | 0.02 | (–0.08, 0.13) | 0.207 | 0.01 | (–0.08, 0.10) | 0.183 |
| Intensity*Session | –0.03 | (–0.13, 0.06) | 0.198 | –0.07 | (–0.17, 0.02) | 0.516 | –0.09 | (−0.17,–0.02) | 2.283 |
| Group*Intensity*Session | –0.05 | (–0.14, 0.05) | 0.402† | –0.06 | (–0.15, 0.03) | 0.547† | –0.01 | (–0.09, 0.07) | 0.190† |
| Self stimulation | 1.02 | (0.86, 1.17) | >100 | 1.38 | (1.22, 1.55) | >100 | 0.82 | (0.69, 0.95) | >100 |
| Group | –0.16 | (–0.32, 0.00) | 1.690 | 0.01 | (–0.16, 0.18) | 0.286 | –0.02 | (–0.16, 0.12) | 0.245 |
| Intensity | 0.53 | (0.39, 0.68) | >100 | 0.63 | (0.49, 0.78) | >100 | 0.64 | (0.51, 0.77) | >100 |
| Session | –0.04 | (–0.16, 0.08) | 0.280 | –0.10 | (–0.22, 0.02) | 0.617 | –0.02 | (–0.12, 0.08) | 0.218 |
| Group*Intensity | 0.01 | (–0.14, 0.16) | 0.253 | 0.00 | (–0.15, 0.14) | 0.289 | 0.00 | (–0.13, 0.13) | 0.264 |
| Group*Session | 0.10 | (–0.02, 0.23) | 0.737 | 0.09 | (–0.03, 0.21) | 0.585 | 0.10 | (0.00, 0.19) | 1.162 |
| Intensity*Session | –0.04 | (–0.14, 0.05) | 0.220 | –0.14 | (−0.24,–0.04) | 5.631 | –0.06 | (–0.14, 0.02) | 0.385 |
| Group*Intensity*Session | –0.03 | (–0.12, 0.07) | 0.191 | –0.01 | (–0.11, 0.09) | 0.184 | –0.04 | (–0.11, 0.04) | 0.234 |
| Other stimulation | 0.33 | (0.19, 0.48) | >100 | 0.28 | (0.10, 0.46) | 20.821 | 0.23 | (0.07, 0.38) | 12.416 |
| Group | 0.13 | (–0.03, 0.28) | 0.852 | 0.09 | (–0.10, 0.27) | 0.490 | 0.07 | (–0.08, 0.23) | 0.378 |
| Intensity | 0.32 | (0.23, 0.41) | >100 | 0.39 | (0.30, 0.48) | >100 | 0.36 | (0.27, 0.45) | >100 |
| Session | 0.11 | (0.00, 0.22) | 1.236 | 0.07 | (–0.04, 0.18) | 0.455 | 0.05 | (–0.05, 0.14) | 0.286 |
| Group*Intensity | 0.05 | (–0.04, 0.14) | 0.290 | –0.03 | (–0.12, 0.05) | 0.191 | 0.02 | (–0.07, 0.12) | 0.231 |
| Group*Session | –0.01 | (–0.12, 0.10) | 0.174 | –0.09 | (–0.20, 0.02) | 0.663 | –0.03 | (–0.13, 0.07) | 0.235 |
| Intensity*Session | –0.14 | (−0.22,–0.04) | 16.742 | –0.11 | (−0.19,–0.02) | 2.595 | –0.14 | (−0.22,–0.06) | 19.384 |
| Group*Intensity*Session | –0.01 | (–0.10, 0.09) | 0.176† | 0.01 | (–0.08, 0.10) | 0.143† | –0.04 | (–0.12, 0.04) | 0.434† |
| Rating | 4.16 | (3.82, 4.48) | >100 | 5.01 | (4.73, 5.28) | >100 | 3.24 | (2.96, 3.52) | >100 |
| Group | 0.02 | (–0.17, 0.20) | 0.289 | –0.06 | (–0.25, 0.12) | 0.335 | 0.05 | (–0.09, 0.18) | 0.333 |
| Session | –0.16 | (–0.50, 0.18) | 0.700 | 0.10 | (–0.19, 0.37) | 0.475 | –0.06 | (–0.34, 0.22) | 0.464 |
| Group*Session | 0.04 | (–0.15, 0.23) | 0.294 | –0.02 | (–0.20, 0.16) | 0.303 | –0.05 | (–0.17, 0.09) | 0.309 |
Appendix 2—table 2
Posterior parameter estimates and contrasts of models for functional magnetic resonance imaging (fMRI) signal in the emotional reactivity task.
Dependent variable: fMRI signal extracted from the respective region of interest (ROI) (standardized to unit variance): Fixed effect: terms in standard font describe the mean regression parameter of the respective event averaged across all conditions. Terms in italic font describe the fixed effect of the respective condition on the regression parameter. Factor codings: Group: control game group = –1, violent game group = 1; Content: neutral = –1, violent = 1; Context: real = –1, game = 1. β/σe: Mean model parameter divided by the mean error standard deviation. Bayes factors were derived from comparing a model where the respective parameter was unrestricted to a model where it was restricted to zero. †These Bayes factors were derived from comparing a model where the parameter was restricted to be negative to a model where it was restricted to zero (one-sided hypothesis test).
| Left amygdala | Right amygdala | |||||
|---|---|---|---|---|---|---|
| Fixed effect | β/σe | 95% CI | BF | β/σe | 95% CI | BF |
| Pictures | 3.59 | (3.21, 3.96) | >100 | 4.25 | (3.91, 4.57) | >100 |
| Group | 0.22 | (–0.15, 0.57) | 0.639 | –0.02 | (–0.33, 0.30) | 0.298 |
| Content | 1.13 | (0.87, 1.38) | >100 | 1.06 | (0.78, 1.31) | >100 |
| Context | –0.01 | (–0.23, 0.22) | 0.252 | –0.06 | (–0.29, 0.17) | 0.281 |
| Group*Content | –0.04 | (–0.29, 0.22) | 0.324† | –0.02 | (–0.27, 0.25) | 0.338† |
| Group*Context | –0.16 | (–0.38, 0.07) | 0.554 | –0.25 | (−0.48,–0.02) | 2.204 |
| Content*Context | –0.06 | (–0.23, 0.12) | 0.227 | 0.00 | (–0.18, 0.18) | 0.205 |
| Group*Content*Context | –0.01 | (–0.18, 0.17) | 0.205† | 0.02 | (–0.17, 0.20) | 0.163† |
| Rating | 1.02 | (0.73, 1.30) | >100 | 0.97 | (0.72, 1.24) | >100 |
| Group | –0.10 | (–0.36, 0.19) | 0.315 | –0.05 | (–0.30, 0.21) | 0.255 |
Appendix 3—table 1
Pictures used in the emotional reactivity paradigm.
Picture ID: International Affective Pictures System (IAPS) ID numbers for pictures of the Neutral Real and Violent Real conditions; arbitrary internal ID numbers for pictures of the Neutral Game and Violent Game conditions.
| Condition | Picture ID | Content | Valence | Arousal |
|---|---|---|---|---|
| Neutral real | 2038 | Woman reading alone | 5.60 | 2.17 |
| 2200 | Man | 5.52 | 1.71 | |
| 2210 | Man | 5.10 | 2.00 | |
| 2215 | Man | 5.30 | 1.63 | |
| 2383 | Woman on the phone | 5.33 | 1.30 | |
| 2393 | Two workers | 5.37 | 1.37 | |
| 2440 | Woman | 5.48 | 1.42 | |
| 2495 | Man | 5.27 | 1.87 | |
| 2570 | Man | 5.50 | 1.57 | |
| 2749 | Man smoking | 5.33 | 1.90 | |
| Violent real | 3010 | Dead body | 1.45 | 6.81 |
| 3015 | Dead body | 1.32 | 7.16 | |
| 3016 | Dead body | 1.87 | 6.20 | |
| 3060 | Mutilation | 1.42 | 7.00 | |
| 3120 | Mutilation | 1.63 | 6.27 | |
| 6530 | Man hits woman | 2.81 | 4.55 | |
| 6550 | Man threatens woman with knife | 2.06 | 5.7 | |
| 6560 | Man threatens woman with gun | 1.87 | 6.13 | |
| 6561 | Man hits woman | 3.29 | 4.23 | |
| 6571 | Man threatens man with gun | 3.23 | 4.06 | |
| Neutral game | ng1 | Woman | 5.40 | 1.67 |
| ng2 | Three women smoking | 5.32 | 1.84 | |
| ng3 | Man | 5.13 | 1.68 | |
| ng4 | Woman | 5.40 | 1.80 | |
| ng5 | Woman | 5.39 | 1.35 | |
| ng6 | Man | 5.23 | 1.42 | |
| ng7 | Woman | 5.61 | 1.81 | |
| ng8 | Man | 5.45 | 1.48 | |
| ng9 | Man | 5.65 | 1.48 | |
| ng10 | Two workers | 5.52 | 1.52 | |
| Violent game | vg1 | Man attacks man with chainsaw | 2.13 | 5.90 |
| vg2 | Mutilation | 2.23 | 5.65 | |
| vg3 | Dead body | 2.43 | 5.33 | |
| vg4 | Dead body | 2.52 | 5.16 | |
| vg5 | Dead body | 2.52 | 4.71 | |
| vg6 | Man shoots man in the head | 2.63 | 5.00 | |
| vg7 | Man shoots man in the head | 2.58 | 5.29 | |
| vg8 | Man chokes man | 3.13 | 4.37 | |
| vg9 | Man shoots man in the head | 3.06 | 4.52 | |
| vg10 | Man threatens man with gun | 3.61 | 3.74 |
Appendix 5—table 1
Linear mixed effects models for ratings in the empathy-for-pain task.
Dependent variable: empathy ratings (visual analog scale, range: 0–100). Factor codings: Group: control game group = –1, violent game group = 1; Session: first session = –1, second session = 1; Intensity: non-painful = –1, painful = 1.
| Fixed effect | β | SE | df | t | p |
|---|---|---|---|---|---|
| Painfulness ratings | |||||
| Group | 0.03 | 0.74 | 87.7 | 0.04 | 0.967 |
| Session | 0.16 | 0.43 | 85.9 | 0.38 | 0.706 |
| Intensity | 25.49 | 0.98 | 85.7 | 25.93 | <0.001 |
| Group*Session | –0.68 | 0.43 | 85.9 | –1.56 | 0.122 |
| Group*Intensity | –0.92 | 0.98 | 85.7 | –0.96 | 0.340 |
| Session*Intensity | –0.23 | 0.47 | 82.0 | –0.50 | 0.621 |
| Group*Session*Intensity | –0.67 | 0.47 | 82.0 | –1.42 | 0.160 |
| Unpleasantness ratings | |||||
| Group | –1.06 | 1.46 | 86.36 | –0.73 | 0.469 |
| Session | –0.81 | 0.55 | 82.36 | –1.47 | 0.145 |
| Intensity | 14.97 | 1.11 | 86.68 | 13.51 | <0.001 |
| Group*Session | –1.04 | 0.55 | 82.36 | –1.89 | 0.063 |
| Group*Intensity | –0.43 | 1.11 | 86.68 | –0.39 | 0.696 |
| Session*Intensity | –0.99 | 0.43 | 84.78 | –2.31 | 0.023 |
| Group*Session*Intensity | –0.13 | 0.43 | 84.78 | –0.30 | 0.762 |
Appendix 5—table 2
Linear mixed effects models for unpleasantness ratings in the emotional reactivity task.
Dependent variable: unpleasantness ratings (visual analog scale, range: 0–100). Factor codings: Group: control game group = –1, violent game group = 1; Content: neutral = –1, violent = 1; Context: real = –1, game = 1.
| Fixed effect | β | SE | df | t | p |
|---|---|---|---|---|---|
| Group | 0.96 | 1.48 | 87 | 0.65 | 0.519 |
| Content | 29.17 | 1.50 | 87 | 19.40 | <0.001 |
| Context | –5.03 | 0.71 | 87 | –7.10 | <0.001 |
| Group*Content | 1.49 | 1.50 | 87 | 0.99 | 0.325 |
| Group*Context | –0.58 | 0.71 | 87 | –0.82 | 0.415 |
| Content*Context | –4.80 | 0.67 | 87 | –7.18 | <0.001 |
| Group*Content*Context | –0.03 | 0.67 | 87 | –0.05 | 0.962 |
Appendix 6—table 1
Posterior parameter means of models for ratings in the empathy-for-pain task, including the trait covariate Neuroticism.
Dependent variable: empathy ratings (visual analog scale, range: 0–100). Factor codings: Group: control game group = –1, violent game group = 1; Session: first session = –1, second session = 1; Intensity: non-painful = –1, painful = 1. The variable Neuroticism was scaled to mean 0 and standard deviation 1. Bayes factors were derived from comparing a model where the respective parameter was unrestricted to a model where it was restricted to zero. †These Bayes factors were derived from comparing a model where the parameter was restricted to be negative to a model where it was restricted to zero (one-sided hypothesis test).
| Fixed effect | β | 95% Credible interval | Bayes factor | |
|---|---|---|---|---|
| Painfulness ratings | ||||
| Group | 0.622 | –1.147 | 2.385 | 0.138 |
| Neuroticism | –0.624 | –2.401 | 1.135 | 0.140 |
| Session | 0.518 | –0.240 | 1.270 | 0.105 |
| Intensity | 27.208 | 24.593 | 29.772 | >100 |
| Group*Neuroticism | –1.767 | –3.597 | 0.024 | 0.541 |
| Group*Session | –0.506 | –1.278 | 0.239 | 0.100 |
| Neuroticism*Session | –0.691 | –1.411 | 0.009 | 0.225 |
| Group*Intensity | –1.253 | –3.788 | 1.276 | 0.255 |
| Neuroticism*Intensity | 0.537 | –2.022 | 3.069 | 0.174 |
| Session*Intensity | –0.374 | –1.521 | 0.791 | 0.082 |
| Group*Neuroticism*Session | –0.155 | –0.898 | 0.578 | 0.046 |
| Group*Neuroticism*Intensity | –0.383 | –2.999 | 2.205 | 0.158 |
| Group*Intensity*Session | –0.599 | –1.743 | 0.506 | 0.112 |
| Neuroticism*Intensity*Session | 0.628 | –0.536 | 1.826 | 0.113 |
| Group*Neuroticism*Intensity*Session | –0.717 | –1.865 | 0.473 | 0.265† |
| Unpleasantness ratings | ||||
| Group | –0.562 | –4.120 | 3.058 | 0.202 |
| Neuroticism | 3.430 | –0.141 | 7.296 | 1.007 |
| Session | –0.258 | –1.534 | 1.131 | 0.081 |
| Intensity | 17.004 | 14.246 | 19.772 | >100 |
| Group*Neuroticism | –2.029 | –5.759 | 1.932 | 0.390 |
| Group*Session | –0.734 | –2.060 | 0.589 | 0.127 |
| Neuroticism*Session | –1.682 | –3.021 | –0.377 | 1.697 |
| Group*Intensity | –0.875 | –3.536 | 1.770 | 0.181 |
| Neuroticism*Intensity | 1.977 | –0.917 | 4.697 | 0.390 |
| Session*Intensity | –1.192 | –2.124 | –0.269 | 1.236 |
| Group*Neuroticism*Session | –0.904 | –2.301 | 0.443 | 0.168 |
| Group*Neuroticism*Intensity | 0.496 | –2.258 | 3.253 | 0.167 |
| Group*Intensity*Session | –0.246 | –1.161 | 0.696 | 0.060 |
| Neuroticism*Intensity*Session | –0.464 | –1.402 | 0.486 | 0.081 |
| Group*Neuroticism*Intensity*Session | –0.850 | –1.796 | 0.092 | 0.466† |
Appendix 6—table 2
Posterior parameter means of models for ratings in the empathy-for-pain task, including the trait covariate SSRT (stop-signal reaction time).
Dependent variable: empathy ratings (visual analog scale, range: 0–100). Factor codings: Group: control game group = –1, violent game group = 1; Session: first session = –1, second session = 1; Intensity: non-painful = –1, painful = 1. The variable SSRT was scaled to mean 0 and standard deviation 1. Bayes factors were derived from comparing a model where the respective parameter was unrestricted to a model where it was restricted to zero. †These Bayes factors were derived from comparing a model where the parameter was restricted to be negative to a model where it was restricted to zero (one-sided hypothesis test).
| Fixed effect | β | 95% Credible interval | Bayes factor | |
|---|---|---|---|---|
| Painfulness ratings | ||||
| Group | 0.990 | –1.016 | 3.014 | 0.173 |
| SSRT | 0.060 | –2.183 | 2.346 | 0.128 |
| Session | 0.529 | –0.374 | 1.431 | 0.101 |
| Intensity | 27.712 | 25.107 | 30.359 | >100 |
| Group*SSRT | –0.413 | –2.724 | 1.953 | 0.139 |
| Group*Session | –0.539 | –1.517 | 0.363 | 0.097 |
| SSRT*Session | –0.290 | –1.389 | 0.794 | 0.068 |
| Group*Intensity | –1.590 | –4.051 | 0.933 | 0.294 |
| SSRT*Intensity | 0.291 | –2.671 | 3.222 | 0.165 |
| Session*Intensity | 0.051 | –1.035 | 1.145 | 0.063 |
| Group*SSRT*Session | 0.587 | –0.440 | 1.611 | 0.105 |
| Group*SSRT*Intensity | 2.030 | –0.898 | 4.986 | 0.402 |
| Group*Intensity*Session | –0.830 | –1.944 | 0.230 | 0.173 |
| SSRT*Intensity*Session | 1.138 | –0.170 | 2.400 | 0.333 |
| Group*SSRT*Intensity*Session | –0.409 | –1.710 | 0.861 | 0.128† |
| Unpleasantness ratings | ||||
| Group | –0.080 | –4.225 | 3.970 | 0.214 |
| SSRT | 0.661 | –3.864 | 5.040 | 0.262 |
| Session | –0.476 | –1.878 | 1.018 | 0.103 |
| Intensity | 17.149 | 14.265 | 19.969 | >100 |
| Group*SSRT | –3.128 | –7.644 | 1.276 | 0.613 |
| Group*Session | –0.817 | –2.328 | 0.692 | 0.151 |
| SSRT*Session | 0.688 | –0.962 | 2.386 | 0.127 |
| Group*Intensity | –1.531 | –4.313 | 1.325 | 0.259 |
| SSRT*Intensity | 1.193 | –1.907 | 4.284 | 0.216 |
| Session*Intensity | –1.089 | –2.014 | –0.211 | 0.822 |
| Group*SSRT*Session | 1.253 | –0.465 | 2.938 | 0.269 |
| Group*SSRT*Intensity | 1.527 | –1.569 | 4.545 | 0.277 |
| Group*Intensity*Session | –0.092 | –1.015 | 0.834 | 0.051 |
| SSRT*Intensity*Session | –0.137 | –1.136 | 0.900 | 0.055 |
| Group*SSRT*Intensity*Session | 0.999 | –0.040 | 2.031 | 0.021† |
Additional files
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Neuroimaging and behavioral evidence that violent video games exert no negative effect on human empathy for pain and emotional reactivity to violence
eLife 12:e84951.
https://doi.org/10.7554/eLife.84951
