Specificity of computationally modeled interpretation biases in health anxiety and the contributions of repetitive negative thinking in a community sample of individuals with elevated health anxiety

Cognitive-behavioral models of health anxiety suggest interpretation biases for ambiguous information play a role in its etiology. This study sought to clarify how specific these biases are in health anxiety and how other potential causal processes, like repetitive negative thinking (RNT), are associated with interpretation biases and health anxiety. Participants from the community (N = 124) with at least mildly elevated levels of health anxiety (Short Health Anxiety Inventory scores ≥ 18) completed a self-report questionnaire assessing repetitive negative thinking (RNT), as well as an interpretation bias task based on the Word Sentence Association paradigm. Task outcomes were computationally modelled using drift diffusion modeling and interpretation bias was operationalized as drift rate, i.e., the rate of evidence accumulation to decide between two choices (in this case, the emotional interpretation or the benign interpretation). Results showed that individuals with health anxiety exhibit interpretation biases specific to health-threatening stimuli but not general anxiety, dysphoric, or positive stimuli. In addition, RNT partially explained the association between interpretation biases for health anxiety-related stimuli and health anxiety symptom severity. Our work suggests health anxiety-specific interpretation biases and repetitive negative thinking may be important targets for intervention, whether through psychotherapy or computerized interventions (e.g., cognitive bias modification).