Altered autonomic arousal in psychosis: an analysis of vulnerability and specificity.

Vulnerability-stress models implicate that alterations of the autonomous nervous system contribute to the development of psychosis. Previous research has found autonomic arousal alterations in psychotic disorders and at-risk individuals that are not explained by medication alone. To test whether these alterations are associated with the extent of an individual's vulnerability and whether they are specific to psychosis, we compared participants with psychosis (n=23), first-degree relatives of individuals with psychosis (n=21), and healthy participants with attenuated positive symptoms (n=23) to participants with depression (n=24) and healthy controls (n=24). At rest, skin conductance level was assessed and photoplethysmography was applied to measure time- and frequency-domain heart rate variability (HRV). Univariate and multivariate analyses of covariance with perceived stress and psychophysiological values as dependent variables showed significant between-group differences for perceived stress (p=.010), heart rate (p=.022), time-domain HRV indices (all ps≤.027), and vagal activity (p=.017). Group differences in sympathetic activity were nonsignificant (p=.069). In an additional analysis with medication as a second between-group factor, the physiological between-group differences remained significant or trend significant (all ps≤.060). With the exception of sympathetic activity, participants with psychosis exhibited more extreme arousal than the control groups. First-degree relatives and participants with attenuated symptoms showed comparable autonomic activity to healthy controls. Thus, the hypothesized association of an alteration of arousal and vulnerability to psychosis was not confirmed. However, particularly low time-domain HRV was found for psychosis, with significant differences to healthy controls (all ps≤.007) and to depression (all ps≤.004), with the latter indicating a specificity to psychosis.