Keywords: Adult Anxiety | Cognitive Biases / Distortions
Presentation Type: Symposium
Research highlights the benefits of computerized interventions designed to shift the way individuals interpret information. In Cognitive Bias Modification for Interpretations (CBM-I), participants read and imagine themselves in ambiguous scenarios that resolve benignly. CBM-I can successfully reduce threat interpretations and subsequent anxiety. However, some studies, particularly those conducted online, have mixed results. The current study tested multiple single-session variants of training to maximize the strength of CBM-I effects. The goal was to determine which variants are optimal, so future trials can follow-up on the strongest variations. This study included 13 variations of CBM-I, a non-CBM-I cognitive training condition, a neutral training condition, and a waitlist control condition. This study was conducted over the Internet via Amazon’s Mechanical Turk. Individuals with high trait anxiety (N=975) completed baseline measures of anxiety and interpretation bias, and were randomized to 1 of 16 conditions. After training, participants completed measures of interpretation bias and reactions to an anticipated social stressor. Results suggest that all 13 CBM-I variations (but not the neutral training or the non-CBM-I cognitive training) led to significantly more positive (0.43ds<0.81) and less negative (-1.14ds<-0.74, except an audio variation) interpretations compared to the waitlist control. However, when testing the effects of CBM-I on a different measure of interpretation bias, and a measure assessing responses to anticipated social situations, only four variations differed from the waitlist control: standard CBM-I, an audio variation, a variation that alternated between positive and negative scenarios in blocks, and a variation that targeted participants’ ability to reframe negative scenarios. We also assessed if baseline interpretation bias moderated condition effects. Results suggest that only the standard CBM-I condition was moderated by baseline positive (p<.0005) and negative (p<.0002) interpretations. Results will be discussed in light of theoretical and clinical implications, and will provide recommendations for future CBM-I research.
Assistant Professor, Clinical Psychology
West Virginia University
Sunday, November 19
10:15 AM – 11:45 AM
The asset you are trying to access is locked. Please enter your access key to unlock.