SUPPLEMENTARY INFORMATION

Abstract

METHODS Participants. All participants were mentally and physically healthy. The average (± SD) body mass index was 21.4 ± 2.7 kg/m 2 in women and 22.8 ± 2.4 kg/m 2 in men (p=0.18). Screening procedures consisted of medical and psychological examinations, and routine blood and urine analyses. Exclusion criteria included any sleep disorders, history of shift work or jet lag within the past 3 months, any medical conditions, drug abuse, and cigarette smoking. The participating women had regular menstrual cycles between 26-31 days, were nulliparous, not currently breast feeding, and free of hormonal contraceptives or gynecological pathology. A total of 6 women had never been on oral contraceptives and 3 of them had taken them ≥3 years before entering the study. The prior use of contraceptives was not documented for 2 women but at least 1 month elapsed between the time of recruitment and the laboratory experiment. Moreover, these two women had regular ovulatory cycles prior to enrolment and during the experiments as confirmed by plasma and urinary progesterone levels, respectively. All female participants were studied selectively during the follicular phase of the menstrual cycle. They entered the laboratory during days 5-9 after menses. Eight of the 11 women were also studied during the mid-luteal phase of their menstrual cycle (subjects no. 1-8; SI Appendix, Table S2; lab entry during days 19-23 after menses). For all 11 women, plasma progesterone was measured around day 21 of their menstrual cycle to confirm ovulation prior to lab entry. At that time, plasma progesterone was on average (± SEM) 40.36 ± 5.06 nmol/L. As reported previously in the 8 women studied at both phases of the menstrual cycle (1), urinary progesterone levels increased significantly during the ML phase compared to the MF 1