Summary:The relation between anxiety and cortical ac tivity was compared in two samples of normal volun teers. One group was studied with the noninvasive xenon-133 inhalation technique for measuring cerebral blood flow (CBF) and the other with positron emission tomography (PET) using lsFlurodeoxyglucose (1sFDG) for measuring cerebral metabolic rates (CMR) for glu cose. The inhalation technique produced less anxiety than the PET procedure, and for low anxiety subjects, there was a linear increase in CBF with anxiety. For higher anxiety subjects, however, there was a linear deWe observed that normal volunteers participating in studies of cerebral blood flow (CBP) and metab olism (CMR) manifest anxiety. The effects of anx iety on these measures are not obvious . Greater anxiety is associated with autonomic arousal (see Beck, 1978), but its relation to cortical arousal re mains to be determined. There is evidence for an inverted-U relationship between anxiety and per formance (Yerkes and Dodson, 1908;Courts, 1939; Hebb, 1955;Malmo, 1959;Eysenck, 1985) with better performance associated with intermediate anxiety. In a preliminary study examining the rela tionship between anxiety and CMR for glucose, we found a curvilinear relation between state anxiety and CMR in a sample of 18 normal volunteers (Gur et aI., 1981; Reivich et aI., 1983). In a subsequent study measuring CBP with the xenon-133 inhala- tion technique (Gur et ai., 1987), we found an in verted-U relationship between anxiety and perfor mance, as well as between anxiety and CBP, mea sured during cognitive activity.The purpose of the present study was to examine the effects of anxiety on cortical CBP and CMR. The xenon-133 inhalation method is noninvasive, whereas the PET 18PDG technique for measuring CMR (as applied in our PET center) requires both arterial and venous catheterization. Thus, CBP measures are taken during a situation probably in voking less anxiety. This permits comparison of the effects of anxiety on cortical activity in higher and lower anxiety-provoking conditions. The hy pothesis of an underlying curvilinear, inverted-U relationship between anxiety and cortical activity stipulates a positive linear relationship between anxiety and cortical activity for low anxiety states and a negative linear relationship for high anxiety states. This would lead to the expectation that the CBP measurement condition will more likely reveal both components of this relationship, whereas the higher anxiety associated with CMR measurement conditions may diminish the positive linear compo nent.