Effect of psychological treatment on cognitive bias in generalized anxiety disorder.

Measures of attention and implicit memory for threatening words were obtained from anxious patients before and after psychological treatment, and compared with data from non-anxious control Ss collected over the same period. Findings confirmed the expectation that the presence of threatening distractors would be associated with greater interference with the performance of anxious patients than with that of controls, in both color-naming and attentional search tasks, but failed to confirm the previous finding of related differences in priming on a word completion task. Treatment significantly reduced selective interference effects in anxious patients, and abolished evidence of differences between the treated patients and controls. It is suggested that cognitive bias effects in anxiety may either depend on state factors alone, or may represent a more enduring individual difference that becomes apparent only when vulnerable individuals are primed by mood state or stressful events.