Detecting Social Desirability Bias with Human-Computer Interaction: A Mouse-Tracking Study

Social desirability bias undermines self-report accuracy, necessitating novel approaches to detect and mitigate its impact. This study aimed to investigate the influence of social desirability on questionnaire responses by analyzing mouse cursor movements and answering behaviors. Respondents (n=238) completed a health and wellness questionnaire while their mouse cursor data was recorded. The results revealed that individuals under a higher social desirability treatment exhibited significantly longer response times and slower mouse cursor speeds, supporting the hypothesis that they may engage in more cautious and deliberate responding. However, no significant differences were found in terms of mouse cursor deviations or answer switches between the two groups. These findings suggest that analyzing mouse cursor movements can provide valuable insights into the influence of social desirability bias on questionnaire responses, offering a potentially scalable method for detection and future intervention.