Table 2.

Information about the selected studies on video games for emotional training.

Study	Sample	Age	Study design	Conditions	Time spent playing	Measures	Main outcomes	Effect sizes
Bailey and West, 2013	31	18–45 years old	Randomized controlled trial	Three-conditions: – Action video game group – Non-action video game group – No contact control group	Ten sessions (1 h each) across 10 consecutive days	Pre and post training: – The Emotion Search Task while ERPs (Event Related Potentials) are recorded	Playing an action game resulted in a change in the affective information processing, with an the an increase in the amplitude of the ERPs over the right frontal and posterior regions that was similar for angry, happy, and neutral faces	– For the action group, the latency of the P3 ERP component was shorter for the post-test than for the pre (η2 = 0.44); for the no-contact and non-action group, the main effect of occasion was not significant (η2 = 0.17)
Bouchard et al., 2012	41	24.9 ± 3.9 years old	Randomized Controlled Trial	Two conditions: – Video game group – Training-as-usual control group	Three sessions of 30-min (one per day for 3 days)	Pre and post training: – Concentration of salivary cortisol and heart rate – Perceived sense to control stress	Decrease of stress level after the training with the video game	– When comparing efficacy of the ImPACT program on the main measure of stress, there was a significant difference in cortisol response (η2 = 0.17) documenting that the program was effective in better controlling stress than training as usual – The ImPACT program had a significant positive impact on stress levels measured thought heart rate (η2 = 0.11) during the apprehension phases – Significant interaction Time × condition on self-reported perceived stress (η2 = 0.31), with ImPACT program contributing for the greater decrease
Dennis and O'Toole, 2014	78	17–50 years old	Randomized Controlled Trial	Four conditions: – Video game group short training – Active control group short training – Video game group long training – Active control group long training	Single session (45 min for the long training condition and 25 for the short training condition)	Pre and post training: – Tries social stress test (TSST) (stress reactivity) – Baseline mood questionnaire – Threat bias (dot-probe task)	Evidence that an alternative delivery strategy for ABMT— a gamified mobile app — shows transfer of benefits to independent, untrained lab-based measures of anxiety and stress reactivity after a single session of training. The long but not the short active training condition reduced the core cognitive process implicated in ABMT (threat bias)	– Threat bias: significant main effect of Training Condition on attentional bias (η2 = 0.13), and disengagement (η2 = 0.19); the effect for vigilance did not reach significance (η2 = 0.01) – Significant main effect of training condition on self-reported anxiety (η2 = 0.10)
Dennis-Tiwary et al., 2016	42	18-38 years old	Randomized controlled trial	Two conditions: – Video game group – Active control group	One session (25min of total gameplay with two 10-min breaks)	Pre and post training: – State-trait anxiety inventory (STAI) – Threat bias (Dot-probe task) while recording EEG activity –Trier social stress test (TSST)	Improved behavioral performance during the anxiety-related stress task among female participants only	The main effect of Training Group did not reach significance for observed behavioral performance (η2 = 0.02); there was a significant interaction between training group and gender for observed behavioral performance (η2 = 0.13): behavioral performance was better for the ABMT vs. placebo training (PT) condition, but only for females (p = 0.02); for the PT condition only, males had better performance compared to females (p = 0.02) – The interaction between Training Group and Gender was significant for N2 amplitudes to threat vs. non-threat (η2 = 0.11): post-training N2 amplitudes to threat vs. non-threat were greater following ABMT vs. PT, but only for males (p = 0.02); for the PT condition only, post-training N2 amplitudes to threat vs. non-threat were greater for females vs. males (p = 0.02) – No other effects reached significance
Naugle et al., 2014	22	20.5 ± 1.1 years old	Quasi-experimental	Two conditions: – Video game group – Traditional physical activity group	Three sessions, 2 exercise activities per session for 20 min each	Pre and post training: – Heart rate – Enjoyment – Positive and negative affect scale (PANAS)	Wii boxing and Wii tennis elicited the highest levels of enjoyment and produced an increase in positive emotions after game play	– Significant effect of Activity on average heart rate (d = 3.01); however, the main effect was superseded by a significant Activity × Group interaction (d = 1.07) – Significant effect of Activity for level of enjoyment (d = 1.08); follow-up tests indicated that the participants enjoyed playing Wii tennis significantly more than they did all the other exercise activities except for Wii boxing – Significant Activity × Time interaction (d = 0.74): the post hoc tests showed that positive affect significantly (a) decreased after walking and running on the treadmill, and (b) increased after playing Wii tennis and Wii boxing