The combined oral contraceptives (COCs) are a mixture of synthetic oestrogen and progestogen. In contrary to beneficial action of COCs in protecting uterine endometrium, a growing number of studies indicate an increased risk of cardiovascular, psychophysiological, and cognitive side effects through a reduction in vagal tone. These side effects are mediated by the catalysation of progesterone into allopregnanolone, which is known to attenuate vagal neurotransmission within brain stem (1). It is widely accepted that physical exercise stimulates the vagus nerve by a non-pharmacological intervention (2). Our objective in this study was to investigate the underlying mechanism in the therapeutic action of exercise as a vagal stimulator in COC-induced side effects that include anxiety, memory dysfunction, and cardiovascular damage. Adult female Sprague-Dawley rats were randomly divided into 4 groups as control, exercise, non-exercised COC and exercised COC groups (n=8 each). The treadmill exercise protocol required 25 m/min, 1 h/day, 3 days/week, for a total of 1 0 weeks, which corresponds to 60% of the maximum aerobic power (3). The groups received vehicle (distilled water, p.o.) or COC containing levonorgestrel and ethinylestradiol (1.0 µg/kg/day and 5.0 µg/kg/day, respectively) for 10 weeks. At the end of the 10th week, passive avoidance test (PAT) was performed to evaluate 72 h memory retention. Spatial memory was measured using the Y-maze. The hole-board test was performed to evaluate anxiety level. Vagal tone was assessed by heart rate variability (HRV) under an irreversible urethane anaesthesia (1.2 g/kg), after which the contraction / relaxation responses of the aortic rings were obtained with carbachol following precontraction with phenylephrine. Serum and brain tissues were obtained to measure allopregnanolone level by ELISA. The data were analysed using one-way ANOVA followed by post hoc Tukey test. The non-parametric Kruskal-Wallis test was applied when appropriate. p < 0.05 was considered statistically significant. Serum allopregnanolone level was increased in the exercised COC group (p<0.05); however, brain tissue allopregnanolone levels were not different among experimental groups. The non-exercised COC group demonstrated a significant memory impairment in PAT as compared to the control group (p<0.05), which was accompanied by memory deficits in the Y-maze score (p<0.05). Despite that PAT score was increased in the exercised COC group, it was not statistically significant. Anxiety was increased in the COC group as compared to non-treated exercise group (p<0.01). The lowest HRV scores were measured in the non-exercised-COC group, without reaching statistical significance. The aorta strips of all treatment groups showed decreased contractile responses to phenylephrine when compared to control aortae (p<0.001), while relaxation responses were similar in all groups. Although COC treatment did not diminish cardiovascular functions, it increased anxiety and thereby inhibited exploratory behaviour. When added to COC treatment, exercise tended to increase allopregnanolone, which could have a role in increasing the vagal tone. Taken together, more research is needed to elucidate how exercise influences the metabolism of neurosteroids and brain stem autonomic circuits.