(PS8-52) A Preliminary Study of Computerized Interpretation Bias Modification Training for Social Anxiety Disorder: A Randomized Controlled Trial

Abstract Text: Social anxiety disorder (SAD) is characterized by a tendency to negatively interpret ambiguous social situations or catastrophize mild negative events, a cognitive pattern known as interpretation bias. Cognitive bias modification for interpretation bias (CBMi) aims to retrain individuals with SAD to interpret ambiguous social scenarios more benignly. Despite growing evidence supporting the efficacy of CBMi, its therapeutic impact may be limited by low engagement due to monotony and boredom. To enhance effectiveness, this study developed a mobile-based CBMi program incorporating psychoeducation, personalization, and priming video components. This randomized controlled trial compared the effect of the enhanced CBMi program to a stress management psychoeducation (SMP) program in reducing SAD symptoms.

As of March 1, 2025, 51 participants underwent diagnostic interviews, with 19 meeting the criteria for SAD and being randomly assigned to either the CBMi (n=10) or SMP (n=9) groups. Both groups completed six training sessions over two weeks via mobile devices. Training involved sentence-completion tasks, where participants filled in missing letters (e.g., “They are discussing something that impress_d them during your presentation”), followed by a consolidation question to reinforce the intended interpretation. The CBMi condition aimed to promote benign interpretation of social scenarios, whereas the SMP condition provided training on general stress management skills. Participants completed self-report measures, including the Liebowitz Social Anxiety Scale (LSAS), Mini Social Phobia Inventory (MSPIN), and Brief Fear of Negative Evaluation (BFNE) at pre-, post-, and follow-up assessments. Symptom changes across time and groups were assessed using repeated measures ANOVAs.

Among the 19 enrolled participants, 18 have completed post-training assessments (CBMi: 9; SMP: 9; mean age=25.06 years; 77.8% female; 83.3% White; training completion=98%). A repeated measures ANOVA revealed a significant main effect of time on the LSAS total score (F(1,16)=35.17, p<.001, η²=0.69), indicating a significant reduction in social anxiety symptoms in both groups (% of reduction: CBMi=27.5 vs. SMP=21.4). A similar pattern was observed for the MSPIN, with a significant main effect of time (F(1,16)=11.47, p=.004, η²=0.42)(% of reduction: CBMi= 21.9 vs. SMP=19.4). Additionally, a significant main effect of time was found for the BFNE scores (F(1,16)=17.49, p=.001, η²=0.52), reflecting a reduction in fear of negative evaluation in both groups (% of reduction: CBMi= 13.1 vs. SMP= 8.0). Overall, CBMi showed a numerically greater reduction in social anxiety symptoms compared to SMP.

Preliminary findings suggest that mobile-based CBMi training effectively reduces social anxiety symptoms and fear of negative evaluation in individuals with SAD, with greater symptoms reduction than SMP. The incorporation of psychoeducation, personalization, and ecologically valid components may enhance engagement and therapeutic outcomes. While both interventions yielded improvements, CBMi showed superior outcomes, supporting its potential as a scalable, accessible intervention for SAD. Further research with a larger sample is warranted.