A new computerized intervention for Social Anxiety Disorder – A randomized controlled trial.


In a recent study, Amir and Taylor (2012)  investigated a new, computerized intervention for individuals with generalized social anxiety disorder. This intervention aimed to modify individuals’ interpretations of ambiguous social and non-social situations in order to reduce their symptoms of social anxiety. Results indicated that the intervention was successful in modifying individuals’ interpretations and that those individuals who received the treatment experienced significant reductions in anxiety. These results are important as they suggest that brief interventions focused on subtly modifying how individuals interpret situations can lead to significant changes in anxiety.


What were the goals of this study?

There were two main goals for this study:

  • The first goal was to examine the effect of this computerized intervention on anxiety symptoms. The authors hypothesized that that this intervention would change individuals’ interpretations bias, which would in turn reduce anxiety and impairment due to anxiety.
  • The secondary goal of the study was to examine whether this brief intervention had any impact on depression and/or trait anxiety.


What is an interpretation bias and why did the authors select this bias as a target for treatment?

An interpretation bias is a bias in which ambiguous situations are interpreted in a threatening manner. Cognitive theories of anxiety posit that this bias is a key reason why individuals develop anxiety. For those with social anxiety disorder, these theories argue that individuals have a tendency towards interpreting social situations in a threatening manner, which then leads to greater avoidance of social situations in an effort to reduce/avoid distress. Many interventions for anxiety are based on these theories and target the avoidance aspect via exposure techniques or the interpretation bias through cognitive strategies. This intervention focused on the interpretation bias.

How did the authors examine the study goals?

The authors recruited individuals for the study through several different resources (e.g., newspapers, referrals, community centers) and then verified that the individual had generalized social anxiety disorder. Those that had generalized social anxiety disorder and met other study criteria were asked to participate in the study and were assessed with several self-report and clinician-administered measures of anxiety and depression. Individuals were then randomized into either the intervention or the placebo group and were re-assessed after completing 8-12 sessions of the intervention. Individuals were also asked to complete follow-up assessments three months after the intervention was completed.

Describe the intervention.

The intervention was a computerized task that the authors called IMP, which stands for Interpretation Modification Program. The task presented individuals with a word, which was either neutral or threatening and then a sentence that described an ambiguous social or non-social situation. After the sentence was presented, individuals were asked to indicate whether the word and sentence were related by pressing a key on the computer. Individuals were reinforced with statement that indicated that they were correct when they made a neutral interpretation and were punished with a statement that indicated that they were incorrect when they made a threatening interpretation.

Describe the placebo.

The placebo was similar to the intervention, except that it had a different reinforcement and punishment schedule. More specifically, in the placebo, individuals were reinforced when they made a neutral interpretation 50% of time and were also reinforced for making a threatening interpretation 50% of the time. The same was true for punishment.

What were the main findings from this study?

There were four main findings:

1. Results of the study demonstrated that the intervention did produce changes in interpretation bias on both social and non-social situations. After completing the sessions, those that received the treatment made less threatening and more neutral interpretations. The authors did find that the placebo intervention also produced some changes on interpretations, but not as much as the active intervention, IMP.

2. After completing the intervention, individuals receiving IMP experienced less impairment due to anxiety than the group receiving the placebo.

3. After completing the intervention, individuals receiving IMP also experienced less depression and trait anxiety than the group receiving the placebo.

4.  After a three-month follow-up, the researchers found that those receiving IMP not only maintained their gains in anxiety and impairment, but they actually continued to improve. Unfortunately, the same was not true for depression.

What are the clinical implications of this study?

The results of this study provide more support for cognitive theories of social anxiety disorder and also strongly suggest that working on modifying individuals biased interpretation of social and non-social situations can be beneficial. Moreover, these results demonstrate that change can be made via a computerized intervention that could be delivered as a standalone treatment, or as part of a more comprehensive therapy.

How can I learn more about this study?

If you would like to learn more about this study, please click the following link:


How can I learn more about this computerized intervention?

If you would like to learn more about this computerized intervention, you may consider reading more articles by Nader Amir, Ph.D. You may find a list of his articles via his university website:


You may also learn more by consulting his business webpage which features computerized interventions:



Amir, N., & Taylor, C. T. (2012). Interpretation training in individuals with generalized social anxiety disorder: a randomized controlled trial. J Consult Clin Psychol, 80(3), 497-511. doi: 10.1037/a0026928



Leave a Reply

Your email address will not be published. Required fields are marked *