A brief breathing challenge reveals a cardiac correlate of generalized anxiety beyond resting physiology and interoceptive measures

Physiological markers measured at rest, such as heart rate variability, have proved relatively weak predictors of mental health symptoms. One potential reason is that resting measures are influenced by uncontrolled state factors, such as spontaneous breathing, whereas a standardized challenge holds these constant and may reveal regulatory differences not evident at rest. Here we asked whether respiratory sinus arrhythmia (RSA) during slow paced breathing, a challenge that strongly increases RSA, carries information about symptoms beyond resting RSA and interoceptive measures. In 207 young adults (mean age 23.6 years, SD = 3.1), electrocardiogram (ECG) was recorded at rest and during slow (6 breaths/min) and fast (20 breaths/min) paced breathing. Lower RSA during slow breathing, relative to that predicted by resting RSA, was associated with more severe generalized anxiety symptoms ({beta} = -0.21, BF = 13), also when depressive symptoms were controlled, whereas resting RSA was not. The association was concentrated among participants with at least moderate anxiety symptoms, whose resting RSA was typical but whose rest-adjusted slow-breathing RSA was 0.62 SD lower, and who in exploratory analyses also showed weaker phase coupling between heart period and respiration. Interoceptive accuracy, measured with heartbeat counting or heart rate discrimination, was unrelated to symptoms, self-report and slow-breathing RSA, and interoceptive questionnaires tracked distress shared across anxiety and depressive symptoms rather than generalized anxiety symptoms. At the group level, slow-breathing RSA thus carries information about generalized anxiety symptoms beyond resting RSA and self-reported interoception. Brief paced-breathing challenges may therefore increase the sensitivity of autonomic markers in anxiety research and merit testing in clinical samples.