One promising avenue toward a better understanding of the pathophysiology of positive emotional disturbances is to examine high-frequency heart rate variability (HRV-HF), which has been implicated as a potential physiological index of disturbances in positive emotional functioning. To date, only a few psychopathology relevant studies have systematically quantified HRV-HF profiles using more ecologically valid methods in everyday life. Using an experience-sampling approach, the present study examined both mean levels and intra-individual variability of HRV-HF - as well as comparison measures of cardiovascular arousal, sympathetic activity, and gross somatic movement - in everyday life, using ambulatory psychophysiological measurement across a six-day consecutive period among a spectrum of community adult participants with varying degrees of positive valence system disturbance, including adults with bipolar I disorder (BD; n=21), major depressive disorder (MDD; n=17), and healthy non-psychiatric controls (CTL; n=28). Groups did not differ in mean HRV-HF, but greater HRV-HF instability (i.e., intra-individual variation in HRV-HF) was found in the BD compared to both MDD and CTL groups. Subsequent analyses suggested that group differences in HRV-HF variability were largely accounted for by variations in clinician-rated manic symptoms. However, no association was found between HRV-HF variability and dimensional measures of positive affectivity. This work provides evidence consistent with a quadratic relationship between HRV-HF and positive emotional disturbance and represents a valuable step toward developing a more ecologically valid model of positive valence system disturbances and their underlying psychophysiological mechanisms within an RDoC framework.