BACKGROUND Heart rate variability measures are reported to be altered in patients with various diseases, implying related abnormality of cardiac autonomic function. Could associated anxiety-induced mental stress, invariably present in all patients, contribute to these alterations? To address this question, we compared heart rate variability parameters computed from students during university examinations, a known cause of mental stress, with those computed during holidays when students are mentally relaxed. We felt that the changes in heart rate variability, if any, could be attributed to mental stress alone. METHODS We compared short-term heart rate variability parameters computed from 5-minute electrocardiogram recordings during supine rest in 18 healthy students, along with recordings of respiratory rate and blood pressure, during university examinations and holidays. RESULTS The mean of all the RR intervals was significantly lower, while the mean arterial pressure was significantly higher at the time of examinations. Measures of total variability, namely standard deviation of the normal-to-normal RR intervals and total spectral power, were significantly reduced during examinations. Low frequency normalized units (LF nu), high frequency (HF) nu and the LF/HF ratio were not significantly different. CONCLUSIONS The decreased mean RR interval indicates an overall increase in heart rate or cardiac sympatho-excitation during examinations. This is associated with a decrease in total heart rate variability at the time of mental stress. None of the parameters indicative of the ratio of the extent of modulation of cardiac sympathetic to parasympathetic activity were different. These findings may be useful when interpreting measures of heart rate variability in patients.