Introduction: Psychological flexibility (PF) has emerged as an important process in understanding adaptive functioning and adjustment. PF refers to the ability to persist or change behavior in the service of long-term life values. Previous research has consistently related PF to diminished psychological distress and enhanced psychological well-being and life satisfaction. Recent research indicates that PF is also associated with lower anger dysregulation and aggressive behavior. PF is composed of six core processes: acceptance, cognitive defusion, flexible present-focused attention, self-as-context, values-based action, and committed action. This study focusses on how PF affects anger by investigating the contribution of some of these specific PF processes to dysfunctional anger. Methods: Participants were 391 undergraduate students (64.4% females; mean age = 21.99; SD = 2.13) recruited at the University of Pisa (Italy). They completed measures of valued living (Valued Living Questionnaire), committed action (Committed Action Questionnaire), cognitive defusion (Cognitive Fusion Questionnaire), acceptance and contact with the present moment (Kentucky Inventory of Mindfulness Skills). The STAXI-II was used to assess four different dimensions of anger: a) aggressive behaviour directed toward other persons or objects (Anger Expression-Out, AE-O); b) inward expression of anger (Anger Expression-In; AE-I); c) attempts to prevent the expression of anger (Anger Control-Out, AC-O); and, d) controlling angry feelings by calming down or cooling off (Anger Control-In, AC-I). Four Hierarchical Regression Analyses were conducted to explore the contribution of the PI processes to each anger dimension. Results: All PF processes were significantly associated with lower anger expression and higher control of anger, with the exception of valued living that did not significantly correlate with two anger domains (i.e., AC-O and AE-O). After controlling for background variables, PF processes explained a statistically significant portion of variance in each dimension of anger (from 5% for AC-O to 27% for AE-I). Four out of five PF processes proved to significantly predict anger dimensions. Acceptance was a significant predictor of lower anger expression, while committed action was a significant predictor of higher anger control. Lower valued living and higher cognitive fusion scores did significantly predict higher AE-I. Discussion: Acceptance seems to play a central role in determining lower levels of anger expression. This finding is in line with hypotheses from the PF model suggesting that aggressive behaviour may function as a way to avoid contact with inner unpleasant experiences. Committed action proved to be important for the control dimensions of anger, suggesting that when people act in accordance with their values they are more inclined to inhibit their aggressive behaviour and angry thoughts, perhaps because value-based actions are in contrast with aggressive behaviour. These results also support hypotheses from the PF model suggesting that rigid cognitive fusion with angry thoughts leads to enhanced feelings of anger. Conclusion: PF processes seem to contribute in different ways to anger expression and control.