The total participants were 190 adolescents (95 boys, 95 girls) from a small high school in the southwestern area of China. Adolescents ranged in age from 15 to 19 years (M = 16.22, SD = 1.60).

Experimental Design

Multi-factorial design was used, with Movie Type and Aggressive Trait as independent variables and Aggression as dependent variable. 2 (Movie Type: violent vs. non-violent) × 3 (Aggressive Trait: HT, MT, LT) repeated two measures analysis of variance (ANOVA) was conducted with Movie Type and Aggressive Trait as between-group factor, and Goal Word as within-group factor. According to score distribution, participants who got score at the top of 27% on Buss Perry Aggression Questionnaire (BPAQ) were defined as HT, and the last 27% were seen as LT, and the rest were considered as MT.

The Main Effect of Movie Type, Movie Type × Aggressive Trait Interaction

In the study, we would like to verify the aggressively priming effect of violent movie and aggressive word on implicit aggression, and postulated each participant had a mean aggressively priming score (APS), which meant the calculated score for the mean RT value of aggressive words minus nonaggressive words and divide by 2 for the violent movie group (THE BIG FAT KILL), and found whether significant difference of APS in group type (See Table 1).

Note: *p < 0.05.

Multivariateanalysis of covariance (MANCOVA) was used to test whether significant difference was found in APS among independent variables (See Table 1). Table 1 showed no significant difference in main effect of movie type on aggression (F (1,184) = 1.79, p > 0.05).

As can be seen in Table 1, there was a significant Movie Type × Aggressive Trait interaction (F (2, 184) = 4.09, p < 0.05). Further simple effect analysis showed that aggression of HT adolescents was significantly elicited by violent movie clips (F (2, 184) = 5.46, p < 0.01). But no significant aggression difference was elicited by nonviolent movies (F (2, 184) = 2.79, p > 0.05; See Table 2).

Note: **p < 0.01.