Potentially interactive effects of hypertension and age on the performance of neuropsychological and information processing tests were examined in 123 untreated hypertensive and 50 normotensive men. After covarying education, average alcohol consumption, trait anxiety, and depression scores, results indicated an interaction of age and hypertension. Young hypertensive men (23-40 years) scored significantly worse than young normotensive men on tests of attention/executive function and working memory; middle-aged hypertensive (41-56 years) and normotensive participants were not distinguished by any measures. Hypertensive men performed significantly more poorly than normotensive men on tests of manual dexterity. Results suggest that neuropsychological sequelae of hypertension are more pronounced in young than in middle-aged hypertensive individuals and are independent of various demographic, psychosocial, and alcohol-related factors.

To determine the stability of individual differences in cellular immune reactions to acute mental stress, we correlated enumerative and functional lymphocyte responses to an evaluative speech task across two experimental sessions scheduled 2 weeks apart in 30 young men. Relative to pretask baseline measurements, the speech stressor elicited a diminished proliferative response to phytohemagglutinin and concanavalin A, a decrease in circulating CD19 lymphocytes, and an increase in both CD8 and CD56 lymphocytes across the two occasions of testing. Test-retest correlations were significant for the magnitude of change in proliferative response to PHA (r = .50, p < .005) and in numbers of circulating CD8 and CD56 cells (r = .53, and .42, respectively; p's < .02). Concomitant cardiovascular responses also correlated significantly over the two experimental sessions (heart rate: r = .78, p < .0001; systolic and diastolic blood pressure: r = .79 and .48, p < .0001 and .007). These data provide initial evidence that interindividual variability of cellular immune responses to acute psychological stress is moderately reproducible on retesting and may therefore denote a stable dimension of individual differences.