Martijn van Teffelen

Imagery-enhanced cognitive restructuring: Efficacy 91 4 on hostile interpretation bias) observed changes in hostile beliefs, but not in hostility traits (Hawkins & Cougle, 2013). Given that we investigated the efficacy of one session of CR, we urge future studies to implement mental imagery techniques more broadly in longer or multi-session treatment protocols. The present study possesses several notable limitations. First, our sample includes participants that scored above average on a hostility measure. Caution should be paid to generalizing these findings as further studies are needed that include participants with higher, more clinical levels of hostility (e.g., persons with hostility scores higher than +1 or +2SD), as many factors (e.g., increased psychosocial stressors and comorbidity) were not controlled for. Second, the majority of the present sample was female (72%). Given that women may express aggressive behavior more indirectly (e.g., through social exclusion) than men (e.g., physical aggression) (Björkqvist et al., 1992) the present findings may differ from those studies using samples including relatively more men. Third, the present design included one single session of (I-)CR. Results can therefore not be generalized to more extensive CBT programs for hostility. Fourth, our small sample size may explain some non-significant findings. Specifically, the observed effects in this study ( b = -0.55) are lower than we expected ( b = 0.98) when we determined the a-priori sample size. The finding that one session of I-CR holds the potential of more efficaciously changing hostile beliefs compared to traditional CR has several clinical or social implications. First, it is likely that the efficacy of I-CR can be further enhanced when it is more broadly integrated in traditional CBT protocols. For example, clinicians can consider adding an imagery component to thought records, or combining imagery exercises as a preparation to behavioral experiments (e.g., McEvoy et al., 2015). Moreover, we selected only one type of CR intervention, namely evidence gathering. One could also think of integrating imagery with other CR techniques such as the ‘pie-chart’ technique, or multidimensional evaluation (Beck, 2011). Second, parts of the present I-CR intervention may be used to test future hypotheses on the working mechanism of CBT and other clinical interventions. For example, it would be interesting to test whether the efficacy of I-CR is attributable to extinction learning (i.e., imaginary exposure) or stimulus revaluation (i.e., changing the meaning of a stimulus). In sum, we observed that both I-CR and CR efficaciously reduced hostile beliefs, aggressive inclinations and anger compared to an active control condition. I-CR was more efficacious and sustainable than CR and AC in reducing hostile beliefs and aggressive inclinations. Findings suggest that implementing imagery techniques in existing treatment protocols can enhance its’ efficacy.