The overall aim of the project is to develop three internet-delivered cognitive behavior therapies for inappropriate anger expression to test the isolated and combined effects of two types of regulation strategies, emotional awareness and cognitive reappraisal. Treatments will be evaluated in a stratified single blind parallel randomized block design, along with a pooled interrupted time series design. Self-recruited adults with elevated levels of anger will be stratified based on baseline level of anger. The particular aims are to examine:

1. the overall and separate effects of emotional awareness training, cognitive reappraisal training, and emotional awareness and cognitive reappraisal training on anger and aggression;
2. whether treatment effects are differentially moderated by baseline levels of emotional awareness, cognitive reappraisal, anger expression-out, anger expression-out, anger control-in, and trait anger;
3. whether the treatment effects are mediated by changes in emotional awareness, cognitive reappraisal, and anger control-in.

1. Following the intention-to-treat principle, all primary data models will include all individuals as randomized. 3-month follow-up will be considered as the primary end-point. We will analyze both (1) overall treatment effect of all treatments (active treatment phase and 3-month follow-up compared to baseline) and (2) the relative efficacy for the emotional awareness training, cognitive reappraisal training, and emotional awareness and cognitive reappraisal training, respectively. We expect that, compared to a prolonged baseline phase, all treatments will result in greater improvements in on both primary and secondary outcomes. Further, we predict that the combined intervention will produce, on average, the strongest effects on primary and secondary outcomes relative to the separate interventions. We expect that cognitive reappraisal training will result in greater improvements relative to emotional awareness as a standalone treatment.
2. Growth models that include the interaction terms between experimental conditions (contrast coded) and baseline levels of moderator variables (as continuous measures) will be utilized to examine moderated treatment effects. Significant interactions will be examined following recommendations for probing interactions in growth models.
3. Weekly measurements of mediators will be analyzed with state-of-the-art methods for longitudinal mediation analysis, and we aim to to examine associations among process and outcome variables both within and between individuals as a function of condition and treatment phase.

Inclusion Criteria:

• experience of inappropriate anger expression that leads to significant distress or interferes with daily life (based on the STAXI-II)

Exclusion Criteria:

• risk for violent behavior that may harm others
• prior convictions of violence/abuse
• ongoing psychological treatment that may conflict with the offered treatment
• change in psychotropic medication within 2 months prior to enrollment
• severe psychiatric disorder, such as severe depression or risk of suicide
• alcohol or drug problems
• life circumstances that interfere with treatment (e.g., homeless)
• pronounced language skill deficits and learning difficulties (as the treatment format assumes normal reading and writing skills)