Barking up the wrong tree in attentional bias modification? Comparing the sensitivity of four tasks to attentional biases

Background and objectives Attentional bias modification (ABM) is a potentially exciting new development in the treatment of anxiety disorders. However, reported therapeutic benefits have not always been replicated. To gauge the sensitivity of tasks used in ABM treatment and assessment, we used a counterbalanced within-subject design to measure their discriminant sensitivity to neutral and threatening facial expressions, comparing them with other well-known tasks that measure visual attention. Methods We compared two tasks often used in the assessment and treatment of attention bias (the dot-probe and the spatial cueing paradigms) with two well-known visual attention tasks (the irrelevant singleton and attentional blink paradigms), measuring their sensitivity to processing differences between threatening and neutral expressions for non-clinical observers. Results The dot-probe, spatial cueing and irrelevant singleton paradigms showed little or no sensitivity to processing differences between facial expressions while the attentional blink task proved very sensitive to such differences. Furthermore, the attentional blink task provided an intriguing picture of the temporal dynamics of attentional biases that the other paradigms cannot do. Limitations These results need to be replicated with larger samples, including a comparison of a group of individuals diagnosed with social anxiety disorder and normal controls. Conclusions Our results indicate that the sensitivity of putative attentional bias measures should be assessed experimentally for more powerful assessment and treatment of such biases. If the attentional blink task is indeed particularly sensitive to attentional biases, as our findings indicate, it is not unreasonable to expect that interventions based on this task may be more effective than those based on the tasks that are currently used.