Attention bias modification for reducing speech anxiety.

The mechanisms mediating the anxiolytic effects of attention bias modification (ABM) remain unclear. Accordingly, we randomly assigned speech-anxious subjects to receive four sessions of one of three training conditions: ABM, inverse ABM, and control. In the ABM condition, subjects viewed pairs of photographs of models displaying facial expressions of disgust and joy on a computer screen. Probes always replaced the positive face, and subjects pushed a button to indicate the identity of the probe (E or F) as rapidly as possible. In the inverse condition, the probes always replaced the negative face, and in the control condition, the probes replaced each face type equally often. After four training sessions, all groups exhibited statistically indistinguishable, but significant, reductions on self-report, behavioral, and physiological measures of speech anxiety. Self-report and behavioral measures of attentional control improved likewise. Contrary to early studies, ABM was not superior to control procedures in producing reductions on measures of social anxiety.