Martijn van Teffelen

General discussion 163 7 Third, the finding that I-CR outperformed traditional CR in a randomized trial suggests that integrating mental imagery in standard CR procedures not only benefits patients with social anxiety disorder (McEvoy et al., 2015) or patients with post-traumatic stress and personality disorder (Arntz & Weertman, 1999; Ehlers et al., 2005; Smucker et al., 1995). It also provides clinical hostility research with an exciting new therapeutic avenue. However, research is still in its early stages and important issues need to be addressed before continuing its’ implementation. Chapter 5 provided a detailed description of the I-CR procedure, including the discussion of problems a therapist may run into during I-CR and potential solutions. Examples of such problems are participants finding it difficult to generate or hold mental images, or participants actively avoiding re-experiencing angry affect. Nonetheless, a number of questions remain to be answered, for example: What is the impact of patients` level of creativity or general imagery capacity on the efficacy of I-CR? What is the impact of multiple I-CR sessions? Do effects sustain over time? And do results generalize do everyday situations? Moreover, in the present protocol we selected ‘evidence-gathering’ as intervention procedure. It may be worthwhile investigating if other CR-techniques, such as the so-called ‘pie-chart’ technique or multidimensional evaluation aligns equally well with the proposed imagery techniques. Hopefully, the present work stimulates further research to facilitate further implementation. Fourth, it was shown in chapter 6 that CBM-I efficaciously reduced hostile interpretation bias and aggressive behavior in a participant sample with clinical levels of hostility. This suggests that people with clinical levels of hostility show small but significant improvements in aggressive behavior after the repeated stimulation of benign interpretations in random ambiguous scenarios across eight weeks. Consequently, this demonstrates that computerized CBM-I holds the potential of an intervention that can be implemented at relatively low cost. Perhaps, CBM-I can be offered as an add-on when patients are on waitlist for therapist- provided therapy. Also, for some people, the threshold for seeking professional help for hostility problems may be too high. For these people and for non-treatment seeking people, CBM-I as a stand-alone intervention may be valuable. However, CBM-I research in hostility is still in its early stages and a number of questions remain to be unanswered prior to further implementation, such as: What is the optimal dose-response effect? Does efficacy sustain over time? Is the intervention effective in everyday clinical practice? And can CBM-I serve as an add-on to standard treatment, for example when people are on waitlist? Fifth, for a significant number of patients, hostility problems are a primary reason to seek help. However, hostility is not described as a separate ‘disorder’ in traditional classification instruments such as the DSM-5. Instead, aspects of hostility are included in other disorders. The transdiagnostic approach in the present work may therefore also have diagnostic implications. Chapter 2 revealed up to five separate hostility dimensions. When these dimensions are mapped on traditional DSM-5 classifications, it becomes clear that hostility criteria for Intermittent