Cognitive bias modification for interpretation (CBM-I) is a promising treatment modality for social anxiety disorder, but effect sizes are relatively small across investigations (Hallion & Ruscio, 2011). Additionally, the extent to which CBM-I impacts other cognitive, emotional, and behavioral outcomes is unclear. This study investigated whether two empirically supported treatment components for anxiety disorders, imaginal exposure (IE) and relaxation, augment the effects of CBM-I and increase the extent to which the effects of CBM-I generalize to behavioral and affective outcomes. We randomly assigned 111 undergraduate students with social anxiety to undergo IE, relaxation, or neutral thinking prior to CBM-I, then measured their interpretation biases, as well as their speaking time, anxiety, and subjective evaluations of performance during a speech task. Results indicated that individuals who received IE prior to CBM-I evidenced more adaptive interpretation biases and less behavioral avoidance during the speech task than did individuals who engaged in relaxation or neutral thinking. However, they did not report differential anxiety, habituation, or evaluations of their performance on the speech task. These findings suggest that the combination of CBM-I with adjunct components that target behavioral, rather than cognitive, mechanisms can facilitate transfer of the effects of CBM-I to reducing behavioral avoidance.