Men who routinely reframe upsetting situations in daily life tend to manage mild negative emotions more effectively when they have higher natural resting levels of the stress hormone cortisol, according to a study published in Psychoneuroendocrinology. Resting testosterone levels, examined either alone or alongside cortisol, showed no reliable link to emotional control. The results suggest that everyday psychological habits and baseline stress hormones jointly relate to how men handle mild distress.
Emotion regulation encompasses the mental processes people use to alter the intensity, duration, or expression of their feelings. One widely studied conscious strategy is cognitive reappraisal, which involves deliberately reinterpreting an upsetting scene in a less threatening or more hopeful light to soften its emotional impact. Effective use of cognitive reappraisal is associated with better psychological well-being, whereas trouble regulating emotions is common across many psychiatric disorders.
Executing this mental shift relies on communication between the prefrontal cortex, a brain region behind the forehead involved in planning and self-control, and the amygdala, an almond-shaped structure deeper in the brain that drives rapid reactions to threats. During reappraisal, the prefrontal cortex quiets activity in the amygdala. Both structures contain receptors for cortisol, the primary stress-adaptation hormone produced by the adrenal glands, and testosterone, a reproductive hormone produced mainly by the gonads that also influences social behavior.
Prior work indicated that temporary spikes in cortisol can support emotional reframing in men. For instance, a 2020 experiment with healthy adults found that men exposed to a stressful public-speaking evaluation shortly before viewing unpleasant images were more successful at reframing their emotions than unstressed men, an effect tied to stress-induced cortisol release. Similar benefits appeared when men were given cortisol pills.
Don’t miss the next big psychology finding
Every morning, PsyPost sends a short briefing on the most interesting new peer-reviewed research in psychology and neuroscience.
“These findings are logical given cortisol’s role as a stress hormone – it not only mobilizes our bodies to deal with stress but also helps us to cope with negative feelings,” study lead author Erik Ilkevič, a doctoral candidate in the Department of Neurobiology and Biophysics at Vilnius University in Lithuania, told PsyPost. “However, there is a lack of knowledge about whether cognitive reappraisal is also associated with one’s basal cortisol level – the level that is not manipulated in any way and is simply the participant’s baseline during the experiment.”
Testosterone’s role in reframing emotions has been less certain, though brain imaging suggests it can dampen top-down emotional control in adults. In a 2011 brain-imaging study of healthy men, individuals with higher natural testosterone showed weaker communication between the prefrontal cortex and the amygdala when overriding automatic reactions toward angry and happy faces.
Because the endocrine systems producing cortisol and testosterone continually influence and inhibit one another, scientists often examine them together, notably under the dual-hormone hypothesis, which originally proposed that testosterone promotes status-seeking behavior mainly when cortisol levels are low. Extending joint hormone testing to stress appraisal, a 2025 study of young men by two of the same authors found that higher resting testosterone predicted lower perceived stress after a social challenge only when cortisol was low, while higher cortisol predicted lower stress only when testosterone was low.
Laboratory performance also depends on how intense an emotional trigger is and how practiced someone is at reframing. A 2016 behavioral experiment indicated that reinterpreting unpleasant pictures demanded greater mental effort when the scenes were highly disturbing, whereas people who routinely used reappraisal in daily life experienced less mental strain and stronger reductions in negative feelings.
Building on this work, Ilkevič, senior author Ramunė Grikšienė, a professor of neurobiology at Vilnius University’s Life Sciences Center, and their colleagues in the Hormones and Brain Function Research Group tested whether resting cortisol and testosterone jointly relate to reappraisal performance, and whether everyday reappraisal habits alter those links.
Because viewing emotional photographs involved no threat to social status, the researchers did not base their predictions directly on the dual-hormone hypothesis, expecting instead from prior emotion studies that each hormone’s opposing influence would stand out mainly when the other hormone was low. They focused on healthy young men both to minimize sex-related variations in hormone concentrations and to examine emotional processing in men, who face elevated rates of aggression and suicide linked to emotion regulation difficulties.
The researchers recruited 140 Lithuanian men aged 20 to 36 through university advertisements and social media. Two participants were excluded after reporting diagnoses of depression or schizotypal disorder, leaving a sample of 138 men (median age 27.5) for descriptive and task-level analyses, and one statistical outlier was excluded to leave 137 men in the regression models. The study was not preregistered, though the team posted their data and analysis code publicly and treated follow-up tests at specific hormone levels as exploratory.
Recruiting volunteers for the lengthy laboratory visit turned out to be far easier than Ilkevič anticipated. “Usually, it is difficult to find participants willing to attend a three-hour session, but we easily attracted 140 males who were able to spend that time in the middle of a weekday,” he said. “This shows that the idea of investigating emotion regulation abilities in the context of testosterone and cortisol strongly resonated with the community.”
After arriving at the laboratory between late morning and mid-afternoon, participants completed questionnaires assessing recent anxiety and depression symptoms alongside how frequently they use cognitive reappraisal or emotional suppression in daily life. Roughly an hour after arrival, between 11:23 a.m. and 5:19 p.m., each participant provided a saliva sample via passive drool. Specialists measured baseline cortisol and testosterone concentrations using liquid chromatography coupled with tandem mass spectrometry, a high-precision chemical analysis method.
Following a brief supervised practice round, participants completed a 25-minute computer task featuring 72 photographs of people. Twenty-four showed neutral scenes, such as people walking, which men passively viewed. The other 48 showed unpleasant scenes, split evenly between images of mild distress, such as crying children, and images of severe injuries or open wounds. Men passively viewed half of the unpleasant images at each intensity level while reacting naturally, and actively reframed the other half by reinterpreting the situation in a more positive context or imagining a better outcome.
To help participants apply the strategy consistently, the instructions encouraged realistic reinterpretations of each scene. “For example, when they see a crying child, participants may imagine that nothing bad has happened and that parents are on their way to comfort them,” Ilkevič explained.
Each photograph appeared for six seconds, after which participants rated how negatively it affected them on a scale from 1 to 100. Although the session also included brainwave and pupil recordings and a later social-status manipulation as part of a broader project, the present paper analyzed only the baseline hormones and initial task ratings. The regression models adjusted for body mass index, time of saliva collection, and each man’s emotional ratings when passively viewing unpleasant images of matching intensity.
Across the sample, the task worked as intended: men rated severe images more negatively than mild ones, and active reframing lowered negative ratings relative to passive viewing at both intensity levels. Contrary to the authors’ primary hypotheses, baseline cortisol, baseline testosterone, and the interaction between the two hormones were not statistically significant predictors of reframing performance across the full group, whether images were mild or intense.
A pattern emerged only when accounting for everyday habits during mild unpleasant images. Among men who scored high in habitual reappraisal, higher resting cortisol was associated with lower negative feelings during active reframing. That interaction effect was small, only slightly exceeding the minimum effect size the 137-person regression analysis was statistically powered to detect. Among men with average or low everyday reappraisal habits, cortisol’s link to emotional ratings was not statistically significant.
“Even though the effect was small, it aligns well with the current literature showing positive effects of cortisol on emotion regulation,” Ilkevič noted. “Given that these findings were present only for the slightly negative pictures and only among participants who often use this strategy in daily life, it confirms the idea that hormones should not be investigated in isolation. Hormonal effects are highly context-dependent; therefore, they must be investigated while considering environment, other hormones, participants’ psychological characteristics, and other relevant factors.”
“The effect size we observed was small, suggesting that factors beyond hormonal levels should be considered,” Ilkevič added. “On the one hand, this is revealing because our emotion regulation abilities are not determined by hormonal levels, even though hormones have a significant impact. On the other hand, these findings suggest that baseline deviations in cortisol concentrations might be associated with emotion regulation abilities, which might be important for people undergoing specific treatments or having specific illnesses. However, this is speculation, and given the cross-sectional design of the current study, we cannot comment on causality – it is also possible that those participants who were better at performing cognitive reappraisal had slightly higher basal cortisol levels.”
Looking at the same mild-image interaction from the reverse angle in an exploratory analysis, strong everyday reappraisal habits predicted better laboratory regulation when resting cortisol was high, but unexpectedly predicted worse regulation when resting cortisol was low. In statistical testing, a p-value measures how likely a result at least that extreme would occur by chance alone, with values under 0.05 passing the standard cutoff. Both exploratory slopes cleared that threshold only narrowly, with p-values of 0.031 and 0.041, prompting the authors to call for tentative interpretation.
“Although this effect was marginal (p-value close to the 0.05 threshold) and might be a false positive, it still surprised us, as it is counterintuitive,” Ilkevič told PsyPost. “However, it might just reflect the importance of basal cortisol levels for more efficient cognitive reappraisal, even if you already use that strategy often in daily life.”
During high-intensity graphic scenes, none of the associations involving cortisol, testosterone, or everyday habits were statistically significant, and neither testosterone nor habitual emotional suppression yielded statistically significant effects in any model.
To explain why baseline cortisol showed no broad effect across all men unlike cortisol pills or acute stress in earlier work, the authors suggest that resting cortisol occupies mainly high-sensitivity brain receptors rather than the lower-sensitivity receptors activated by large cortisol spikes. In the researchers’ reading, that mild baseline signal may be insufficient to boost regulation across everyone, yet still enough to fine-tune prefrontal-amygdala circuits in men who have honed those pathways through routine everyday use. For intense graphic scenes, the authors propose that heavy mental demands may overwhelm reappraisal or prompt even habitual reappraisers to switch quietly to simpler tactics like distraction.
As for why testosterone showed no behavioral link, the researchers note that earlier brain-imaging studies used rapid reaction tasks and often detected hormone-related differences in brain activity without changes in outward behavior. With six seconds to view each photograph in the reappraisal task, participants may have had enough time to construct a calming interpretation even if testosterone subtly dampened prefrontal control.
“Surprisingly, we did not observe any robust findings with basal testosterone, even though some other studies suggest that testosterone is associated with impaired emotion regulation,” Ilkevič said. “However, that is actually good news – it means that even those males with higher testosterone levels can successfully engage in cognitive reappraisal and thus effectively regulate their emotions.”
Stepping back from the laboratory data, Ilkevič emphasized that both hormones often suffer from unfair public reputations. “We would like to rehabilitate cortisol a bit,” he explained. “Chronic elevations of cortisol concentrations are, of course, unhealthy, but the optimal basal levels and appropriate cortisol responses to acute stressors are important for the health of both body and mind.”
“Similarly, we also would like to rehabilitate testosterone,” Ilkevič continued. “Stereotypically, testosterone is often assumed to be related to aggression and anger, although the actual literature shows a much more complex picture of testosterone that is heavily context-dependent and cannot be considered in isolation from both environmental and individual factors. Similarly, although we assumed testosterone would be negatively associated with cognitive reappraisal, this was not the case. These findings should further remind us that our assumptions might be wrong and that there is much we still do not understand about the complex hormonal interactions.”
There are a few things to keep in mind when interpreting these findings. Because hormones and task performance were measured in a single session without experimentally changing hormone levels, the design cannot establish cause and effect; as the authors note, routinely reframing emotions over time could shape baseline cortisol levels rather than cortisol aiding reappraisal. Studying only healthy young Lithuanian men also means the findings cannot be generalized to women, older adults, or clinical populations.
In addition, the questionnaire measuring everyday reappraisal habits showed modest internal consistency in this sample, introducing measurement noise that can weaken statistical interactions, and performance was judged solely through self-reported ratings rather than physiological markers. Future research that experimentally administers hormones, includes female participants, and examines physiological data such as brainwaves and pupil dilation could help test how baseline steroid hormones relate to emotional resilience.
That physiological work is already underway using data from the same laboratory sessions to track the autonomic nervous system, which governs involuntary bodily functions including fight-or-flight arousal. “During this study, we also used electroencephalography (EEG) to record brain activity and pupillometry to record changes in pupil dilation, which is used as a proxy for sympathetic nervous system activity,” Ilkevič said. “These measures will allow us to evaluate whether physiological responses in the brain and autonomic nervous system are associated with basal hormone levels, and how they relate to (or differ from) self-reports.”
“Additionally, we are further investigating the potential relationship between cortisol and testosterone by using different emotion regulation strategies to better understand the putative mechanisms,” he added. “Although we did not observe robust effects of testosterone in this study, our previous research showed that testosterone is associated with lower perceived stress when cortisol levels are low. We want to investigate this further by addressing different, more unconscious aspects of the emotion regulation process.”
Beyond emotion regulation, the team is preparing to examine how people’s reluctance to hurt others, known as harm aversion, relates to planned, goal-directed harmful behavior rather than impulsive outbursts. “Finally, we are also planning a new project to investigate mechanisms of harm aversion to predict real-world proactive aggression from laboratory-based measures,” Ilkevič said.
The study, “Investigating associations between cortisol, testosterone and emotion regulation in healthy young men,” was authored by Erik Ilkevič, Ingrida Zelionkaitė, Eglė Mazgelytė, Violeta Cimalanskaitė-Kazlauskienė, Dovilė Karčiauskaitė, Algirdas Utkus, Markus Hausmann, and Ramunė Grikšienė.