An exploratory study of 197 healthy university students in Israel found that associations between sex hormones and some forms of self-reported impulsivity differed according to cortisol and anxiety levels.
The findings came from 66 men, 68 women using hormonal birth control and 63 naturally cycling women. The researchers measured testosterone, estradiol and cortisol in a single saliva sample, then assessed impulsivity with the 45-item UPPS Impulsive Behavior Scale and anxiety with the state version of the State-Trait Anxiety Inventory.
Estradiol patterns varied with cortisol and anxiety
Higher estradiol was associated with more “lack of premeditation”—acting without thinking ahead—among participants with lower cortisol. Among those with higher cortisol, higher estradiol was associated with less lack of premeditation. Both associations were statistically significant.
Estradiol showed a different pattern for perseverance, described in the study as follow-through. Higher estradiol was associated with poorer follow-through among people with lower anxiety, but with better follow-through among people with higher anxiety. Both associations were significant.
Higher testosterone was associated with poorer follow-through only among participants reporting higher anxiety. Cortisol did not significantly alter testosterone’s associations with the impulsivity measures.
Neither cortisol nor anxiety significantly changed the hormone associations with urgency or sensation seeking. Anxiety did alter estradiol’s association with planning, but the result among participants with higher anxiety fell just short of statistical significance.
What the study measured
Participants provided one saliva sample between noon and 6 p.m. and completed questionnaires about current anxiety and impulsivity. The researchers statistically controlled for progesterone. Naturally cycling women were tested on day 21 of a confirmed regular menstrual cycle, described as the mid-luteal phase.
Across the three groups, men had several times more testosterone than either group of women, while naturally cycling women had the highest estradiol levels. Cortisol did not differ between the groups. The reported interaction patterns did not differ significantly among men, women using hormonal birth control and naturally cycling women, and remained after the researchers adjusted for group.
Why the findings need cautious interpretation
The study was cross-sectional, so it cannot show whether hormone levels influence impulsivity, impulsive habits influence hormone levels, or another factor affects both. The participants were healthy, non-smoking undergraduate volunteers in their 20s recruited from college notice boards, which limits how broadly the findings can be applied.
The analysis also included eight regression models and 16 hormone-by-context tests without correcting for multiple comparisons. Five tests reached significance. In a corrected simpler correlation table, only one of 28 associations remained significant.
The study measured impulsivity through self-report rather than behavioral tasks. It also relied on one afternoon saliva sample, even though hormone levels can vary across the day and across the menstrual cycle. The naturally cycling women were assessed only during the mid-luteal phase, not around the estradiol peak near ovulation.
Overall, the results are best understood as leads for further research rather than evidence that changing hormone levels changes impulsivity. The researchers noted that repeated hormone sampling, behavioral measures and larger samples are needed to assess whether the patterns hold up.
What this means for readers
The study adds context to research on impulsivity by examining hormones alongside stress-related cortisol and anxiety rather than treating hormone levels as acting in isolation. However, its findings came from a specific group of healthy students and do not establish effects of contraception or hormone therapy, or support choosing either treatment.
AI tools were used to assist with the preparation of this article under our editorial standards.
Leave a Reply