OpenAI released data on just how many of ChatGPT's users are facing mental health challenges, and how it's addressing them.

A new large-scale study of European adults suggests that, on average, being religiously educated as a child is associated with slightly poorer self-rated health after the age of 50. The research, published in the journal Social Science & Medicine, also indicates that this association is not uniform, varying significantly across different aspects of health and among different segments of the population.

Past research has produced a complex and sometimes contradictory picture regarding the connections between religiousness and health. Some studies indicate that religious involvement can offer health benefits, such as reduced suicide risk and fewer unhealthy behaviors. Other research points to negative associations, linking religious attendance with increased depression in some populations.

Most of this work has focused on religious practices in adulthood, leaving the long-term health associations of childhood religious experiences less understood. To address this gap, researchers set out to investigate how a religious upbringing might be linked to health outcomes decades later, taking into account the diverse life experiences that can shape a person’s well-being.

The researchers proposed several potential pathways through which a religious upbringing could influence long-term health. These include psychosocial mechanisms, where religion might foster positive emotions and coping strategies but could also lead to internal conflict or distress. Social and economic mechanisms might involve access to supportive communities and resources, while also potentially exposing individuals to group tensions.

Finally, behavioral mechanisms suggest religion may encourage healthier lifestyles, such as avoiding smoking or excessive drinking, which could have lasting positive effects on physical health. Given these varied and sometimes opposing potential influences, the researchers hypothesized that the link between a religious upbringing and late-life health would not be simple or consistent for everyone.

To explore these questions, the study utilized data from the Survey of Health, Aging, and Retirement in Europe, a major cross-national project. The analysis included information from 10,346 adults aged 50 or older from ten European countries. Participants were asked a straightforward question about their childhood: “Were you religiously educated by your parents?” Their current health was assessed through self-ratings on a five-point scale from “poor” to “excellent.” The study also examined more specific health indicators, including physical health (chronic diseases and limitations in daily activities), mental health (symptoms of depression), and cognitive health (numeracy and orientation skills).

The researchers employed an advanced statistical method known as a causal forest approach. This machine learning technique is particularly well-suited for identifying complex and non-linear patterns in large datasets. Unlike traditional methods that often look for straightforward, linear relationships, the causal forest model can uncover how the association between a religious upbringing and health might change based on a wide array of other factors. The analysis accounted for 19 different variables, including early-life circumstances, late-life demographics like age and marital status, and current religious involvement.

The overall results indicated that, on average, having a religious upbringing was associated with poorer self-rated health in later life. The average effect was modest, representing a -0.10 point difference on the five-point health scale. The analysis showed that for a majority of individuals in the sample, the association was negative.

However, the model also identified a smaller portion of individuals for whom the association was positive, suggesting that for some, a religious upbringing was linked to better health outcomes. This variation highlights that an average finding does not tell the whole story.

When the researchers examined different domains of health, a more nuanced picture emerged. A religious upbringing was associated with poorer mental health, specifically a higher level of depressive symptoms. It was also linked to poorer cognitive health, as measured by lower numeracy, or mathematical ability.

In contrast, the same childhood experience was associated with better physical health, indicated by fewer limitations in activities of daily living, which include basic self-care tasks like bathing and dressing. This suggests that a religious childhood may have different, and even opposing, associations with the physical, mental, and cognitive aspects of a person’s well-being in later life.

The study provided further evidence that the link between a religious upbringing and poorer self-rated health was not the same for all people. The negative association appeared to be stronger for certain subgroups. For example, individuals who grew up with adverse family circumstances, such as a parent with mental health problems or a parent who drank heavily, showed a stronger negative link between their religious education and later health.

Late-life demographic factors also seemed to modify the association. The negative link was more pronounced among older individuals (aged 65 and above), females, those who were not married or partnered, and those with lower levels of education. These findings suggest that disadvantages or vulnerabilities experienced later in life may interact with early experiences to shape health outcomes.

The analysis also considered how adult religious practices related to the findings. The negative association between a religious upbringing and later health was stronger for individuals who reported praying in adulthood. It was also stronger for those who reported that they never attended a religious organization as an adult. This combination suggests a complex interplay between past experiences and present behaviors.

The study does have some limitations. The data on religious upbringing and other childhood circumstances were based on participants’ retrospective self-reports, which can be subject to memory biases. The study’s design is cross-sectional, meaning it captures a snapshot in time and cannot establish a direct causal link between a religious upbringing and health outcomes. It is possible that other unmeasured factors, such as parental socioeconomic status, could play a role in this relationship. The measure of religious upbringing was also broad and did not capture the intensity, type, or strictness of the education received.

Future research could build on these findings by using longitudinal data to track individuals over time, providing a clearer view of how early experiences unfold into later life health. More detailed measures of religious education could also help explain why the experience appears beneficial for some health domains but detrimental for others. Researchers also suggest that exploring the mechanisms, such as coping strategies or social support, would provide a more complete understanding.

The study, “Heterogeneous associations between early-life religious upbringing and late-life health: Evidence from a machine learning approach,” was authored by Xu Zong, Xiangjiao Meng, Karri Silventoinen, Matti Nelimarkka, and Pekka Martikainen.

A study of men seeking help for delayed or premature ejaculation in Italy found that those suffering from delayed ejaculation tended to have more severe depressive and anxiety symptoms, and lower sexual desire than men suffering from premature ejaculation. They also tended to be older. The paper was published in IJIR: Your Sexual Medicine Journal.

Premature ejaculation is a sexual condition in which a man reaches orgasm and ejaculates sooner than desired, often within a minute of penetration or with minimal stimulation. It can lead to frustration, anxiety, and reduced sexual satisfaction for both partners. The causes may include psychological factors such as stress, depression, or relationship problems, as well as biological ones like hormonal imbalances or nerve sensitivity.

In contrast, delayed ejaculation is the persistent difficulty or inability to reach orgasm and ejaculate despite adequate sexual stimulation. This condition can also cause emotional distress, relationship strain, and decreased confidence. Delayed ejaculation may result from psychological issues, nerve damage, certain medications, or chronic health conditions such as diabetes. Both conditions are forms of ejaculatory disorders and sexual dysfunction. They can occur occasionally or become chronic depending on underlying causes.

Study author Fausto Negri and his colleagues note that many men experiencing ejaculatory disorders have difficulty expressing their negative feelings and that sexuality and emotional expression are closely connected. With this in mind, they conducted a study aiming to define specific clinical and psychological profiles of individuals suffering from premature and delayed ejaculation and to investigate the association between delayed ejaculation and other domains of sexual functioning.

Study participants were 555 men who were seeking medical help for ejaculation disorders. 76 of them reported for delayed ejaculation, while the rest of them sought help for premature ejaculation. Participants’ average age was approximately 45 years. 53% of participants with delayed ejaculation reported having a stable partner, and this was the case with 64% of participants with premature ejaculation.

Participants completed assessments of erectile function (the International Index of Erectile Function) and depression (the Beck Depression Inventory). Researchers also measured levels of various hormones and collected other medical and demographic information about the participants.

Results showed that participants suffering from delayed ejaculation were older than participants suffering from premature ejaculation (average age of 47 years vs 44 years). They also more often suffered from other disorders. Participants with delayed ejaculation also tended to have more severe symptoms of depression and anxiety. Their sexual desire tended to be lower, as were their orgasmic function scores, compared to participants with premature ejaculation. The two groups did not differ in relationship status, waist circumference, body mass index, or levels of examined hormones.

“Roughly one of ten men presenting for self-reported ejaculatory dysfunction as their main complaint in the real-life setting suffers from DE [delayed ejaculation]. Usually, they are older than men with primary PE [premature ejaculation] and overall less healthy. Likewise, they depict an overall poorer quality of sexual life, with lower SD [sexual desire] and OF [orgasmic function]. Moreover, men with DE have higher chances to report clinically significant depression and anxiety, which significantly impact their overall sexual satisfaction,” the study authors concluded.

The study sheds light on the differences in psychological characteristics between people with different forms of ejaculation disorders. However, it should be noted that the design of the study does not allow any causal inferences to be derived from the results. Additionally, all participants came from the same clinical center. Results on men from other geographical areas might differ.

The paper, “Men with delayed ejaculation report lower sexual satisfaction and more depressive symptoms than those with premature ejaculation: findings from a cross-sectional study,” was authored by Fausto Negri, Christian Corsini, Edoardo Pozzi, Massimiliano Raffo, Alessandro Bertini, Gabriele Birolini, Alessia d’Arma, Luca Boeri, Francesco Montorsi, Michael L. Eisenberg, and Andrea Salonia.

Researchers in Japan have documented the case of a teenager whose psychotic symptoms consistently appeared before her menstrual period and resolved immediately after. A case report published in Psychiatry and Clinical Neurosciences Reports indicates that a medication typically used to treat seizures and bipolar disorder was effective after standard antipsychotic and antidepressant drugs failed to provide relief. This account offers a detailed look at a rare and often misunderstood condition.

The condition is known as menstrual psychosis, which is characterized by the sudden onset of psychotic symptoms in an individual who is otherwise mentally well. These episodes are typically brief and occur in a cyclical pattern that aligns with the menstrual cycle. The presence of symptoms like delusions or hallucinations distinguishes menstrual psychosis from more common conditions such as premenstrual syndrome or premenstrual dysphoric disorder, which primarily involve mood-related changes. Menstrual psychosis is considered exceptionally rare, with fewer than 100 cases identified in the medical literature.

The new report, authored by Atsuo Morisaki and colleagues at the Tokyo Metropolitan Children’s Medical Center, details the experience of a 17-year-old Japanese girl who sought medical help after about two years of recurring psychological distress. Her initial symptoms included intense anxiety, a feeling of being watched, and auditory hallucinations where she heard a classmate’s voice. She also developed the belief that conversations around her were about herself. She had no prior psychiatric history or family history of mental illness.

Initially, she was diagnosed with schizophrenia and prescribed antipsychotic medication, which did not appear to alleviate her symptoms. Upon being transferred to a new medical center, her treatment was changed, but her condition persisted. While hospitalized, her medical team observed a distinct pattern. In the days leading up to her first menstrual period at the hospital, she experienced a depressive mood and restlessness. This escalated to include delusional thoughts and the feeling that “voices and sounds were entering my mind.” These symptoms disappeared completely four days later, once her period ended.

This cycle repeated itself the following month. About twelve days before her second menstruation, she again became restless. Nine days before, she reported the sensation that her thoughts were “leaking out” during phone calls. She also experienced auditory hallucinations and believed her thoughts were being broadcast to others. Her antipsychotic dosage was increased, but the symptoms continued until her menstruation ended, at which point they once again resolved completely.

A similar pattern emerged before her third period during hospitalization. Fourteen days prior, she developed a fearful, delusional mood. She reported that “gazes and voices are entering my head” and her diary entries showed signs of disorganized thinking. An increase in her medication dosage seemed to have no effect. As her period began, the symptoms started to fade, and they were gone by the time it was over. This consistent, cyclical nature of her psychosis, which did not respond to conventional treatments, led her doctors to consider an alternative diagnosis and treatment plan.

Observing this clear link between her symptoms and her menstrual cycle, the medical team initiated treatment with carbamazepine. This medication is an anticonvulsant commonly used to manage seizures and is also prescribed as a mood stabilizer for bipolar disorder. The dosage was started low and gradually increased. Following the administration of carbamazepine, her psychotic symptoms resolved entirely. She was eventually able to discontinue the antipsychotic and antidepressant medications. During follow-up appointments as an outpatient, her symptoms had not returned.

The exact biological mechanisms behind menstrual psychosis are not well understood. Some scientific theories suggest a link to the sharp drop in estrogen that occurs during the late phase of the menstrual cycle. Estrogen influences several brain chemicals, including dopamine, and a significant reduction in estrogen might lead to a state where the brain has too much dopamine activity, which has been associated with psychosis. However, since psychotic episodes can occur at various points in the menstrual cycle, fluctuating estrogen levels alone do not seem to fully explain the condition.

The choice of carbamazepine was partly guided by the patient’s age and the potential long-term side effects of other mood stabilizers. The authors of the report note that carbamazepine may work by modulating the activity of various channels and chemical messengers in the brain, helping to stabilize neuronal excitability. While there are no previous reports of carbamazepine being used specifically for menstrual psychosis, it has shown some effectiveness in other cyclical psychiatric conditions, suggesting it may influence the underlying mechanisms that produce symptoms tied to biological cycles.

It is important to understand the nature of a case report. Findings from a single patient cannot be generalized to a larger population. This report does not establish that carbamazepine is a definitive treatment for all individuals with menstrual psychosis. The positive outcome observed in this one person could be unique to her specific biology and circumstances.

However, case reports like this one serve a significant function in medical science, especially for uncommon conditions. They can highlight patterns that might otherwise be missed and introduce potential new avenues for treatment that warrant further investigation. By documenting this experience, the authors provide information that may help other clinicians recognize this rare disorder and consider a wider range of therapeutic options. This account provides a foundation for future, more systematic research into the causes of menstrual psychosis and the potential effectiveness of medications like carbamazepine.

The report, “Menstrual psychosis with a marked response to carbamazepine,” was authored by Atsuo Morisaki, Ken Ebishima, Akira Uezono, and Takashi Nagasawa.

Trigger warnings are meant to help people emotionally prepare for or avoid potentially upsetting material. But new evidence from a week-long study of young adults suggests they often do neither. Instead, most people who encounter these warnings choose to view the content anyway. The findings also indicate that even individuals with trauma histories or mental health concerns are no more likely to avoid warned content than others. The results provide further support for the growing idea that trigger warnings, while widespread, may not function as intended in everyday digital life.

Trigger warnings are now common in both online and offline environments, appearing ahead of everything from social media posts to college course material. They are typically used to signal content that could be distressing, especially for those with past trauma or mental health challenges. Advocates argue that these warnings give vulnerable people the opportunity to prepare for or avoid harmful content.

But a growing body of lab-based studies has cast doubt on the idea that trigger warnings work in the way people hope. While many assume that warnings prompt avoidance, experiments have shown that most people choose to view the content anyway, and that warnings rarely reduce emotional distress. Until now, however, nearly all of this evidence came from controlled settings. Researchers had not yet studied how people actually respond to trigger warnings in their everyday lives.

The new study, published in the Journal of Behavior Therapy and Experimental Psychiatry, aimed to fill that gap. The researchers set out to track when and how often people encounter trigger warnings on social media, whether they choose to view or avoid the associated content, and whether certain psychological traits—such as symptoms of posttraumatic stress or depression—are linked to different patterns of behavior.

“Over the past (almost decade) my research has been concerned with cutting through online debate about trigger warnings and examining them using an experimental framework. This work has found that in the lab, warnings about upcoming negative content do not reduce people’s emotional reactions to material, nor do they seem effective in deterring the majority of people from viewing negative content when given a neutral/non-distressing alternative,” said study author Victoria Bridgland, a lecturer at Flinders University.

“We were interested in seeing if these findings, particularly about avoidance, extend outside of lab environments. Participating in a lab study is inherently coercive, however participants have no obligation to watch or avoid negative content in daily life. However, aligning with lab findings, we found that the most common response to seeing trigger warnings online in daily life was to view the content, and the most common reason given was because of curiosity—which is also something we hear in lab.”

The study followed 261 young adults between the ages of 17 and 25 over a seven-day period. Participants reported their daily experiences with social media, including whether they saw any trigger warnings and what kind of content those warnings accompanied. They also recorded whether they chose to look at or avoid the content after seeing the warning.

To explore whether psychological traits influenced avoidance behavior, participants completed several standardized assessments at the beginning of the study. These included measures of trauma exposure, symptoms of posttraumatic stress disorder, depression, anxiety, and general well-being. The researchers also asked whether participants had a tendency to deliberately seek out reminders of traumatic experiences, a behavior sometimes referred to as self-triggering.

The researchers wanted to see whether people who had higher levels of psychological distress were more likely to avoid warned content, as trigger warning advocates often suggest. They also looked at how frequently participants encountered these warnings and what motivated their decision to view or avoid the content.

Nearly half of the participants reported seeing at least one trigger warning during the week. Among those who did, the average number of warnings seen was about four. The most common platforms for encountering these warnings were Instagram, TikTok, and Twitter, and the most frequent content types were violent or aggressive material, depictions of physical injury, and sexually explicit content.

When asked how they responded to the warnings, the overwhelming majority said they chose to look at the content. On a scale from “never looked” to “always looked,” most people leaned heavily toward viewing. In fact, only around 11 percent reported consistently avoiding warned material throughout the week, while more than a third said they always approached it. When asked why they looked, more than half cited curiosity—the desire to know what was being hidden—as their main motivation.

The results were not a surprise. “We have known for some time from lab experiments that trigger warnings don’t seem to increase rates of avoidance, and we also know that people are morbidly curious and often self-expose themselves to negative material (even when it serves no real benefit),” Bridgland told PsyPost.

The researchers found no evidence that people with higher psychological vulnerability were more likely to avoid the content. Participants with greater posttraumatic stress symptoms, for example, were just as likely to view the material as those with fewer symptoms. This pattern held across several mental health measures, including depression, anxiety, and a history of trauma exposure.

Interestingly, people who did see trigger warnings tended to score higher on mental health symptom scales and lower on general well-being. The authors suggest that this could be because such individuals spend more time in online spaces where trigger warnings are common, or because the warnings feel more personally relevant and memorable to them. But even within this group, the presence of a warning did not increase the likelihood of avoidance.

The content people chose to avoid, when they did avoid it, varied widely. Some said they were simply uninterested, while others avoided it because it involved specific types of content they preferred not to see, such as animal cruelty or depictions of death. A small number of participants reported avoiding material that felt emotionally overwhelming or clashed with their current mood. Still, these decisions were the exception rather than the rule.

“I’d like for people to be conscious consumers of negative material online and be wary of extremes,” Bridgland said. “For example, if you are someone who finds they often need to avoid or becomes overly distressed or triggered by online content or someone who is deliberately searching for and binge consuming negative content in high volumes which is leading to distress—this is likely a sign that there is some underlying issue that likely warrants therapeutic attention. In either of these cases, be aware that a trigger warning may not be serving a beneficial function.”

As with all research, there are some limitations. First, the study did not measure emotional reactions after viewing the content, so it remains unclear whether the warnings helped people feel more prepared or less distressed. Prior research, however, suggests that trigger warnings tend not to influence emotional responses much, if at all.

Another limitation is that people might behave differently depending on the specific context or type of content. For example, someone might avoid a warning about sexual assault but not one about medical procedures. The study also didn’t capture real-time responses, so there may be subtle moment-to-moment factors—such as mood or fatigue—that influence decisions to view or avoid warned content.

“I’d like to clarify that me and my research team aren’t advocating that we should ban trigger warnings, but we just want people to be aware of the lack of benefits they provide,” Bridgland explained. “This way people can take other precautions to safeguard their mental health online.”

“Since it seems hard to improve antecedent based strategies to help people cope with negative content (as various recent studies have tried to “improve” trigger warnings with no success), I’m exploring ways we can help people after they are exposed. This will also help in the case where shocking/traumatic content exposure happens without warning (which is a common experience online).”

The study, “‘I’m always curious’: Tracking young adults exposure and responses to social media trigger warnings in daily life,” was authored by Victoria M.E. Bridgland, Ella K. Moeck, and Melanie K.T. Takarangi.

A new scientific report reveals that the protein aggregates associated with Parkinson’s disease are not inert clumps of cellular waste, but rather are chemically active structures that can systematically destroy the primary energy molecule used by brain cells. The research, published in the journal Advanced Science, demonstrates that these protein plaques can function like tiny, rogue enzymes, breaking down adenosine triphosphate and potentially starving neurons of the power they need to survive and function.

Scientists have long sought to understand how the accumulation of protein clumps, known as amyloids, leads to the devastating neuronal death seen in neurodegenerative conditions like Parkinson’s disease. These clumps are primarily made of a misfolded protein called alpha-synuclein.

The prevailing view has been that these aggregates cause harm by physically disrupting cellular processes, poking holes in membranes, or sequestering other important proteins. However, a team of researchers led by Pernilla Wittung-Stafshede at Rice University suspected there might be more to the story.

Previous work from the same group had shown that alpha-synuclein amyloids were not chemically inactive. They could facilitate certain chemical reactions on simple model compounds in a test tube. This led the researchers to question if these amyloids could also act on biologically significant molecules inside a cell. They focused on one of the most fundamental molecules in all of life: adenosine triphosphate, the universal energy currency that powers nearly every cellular activity.

Neurons have exceptionally high energy demands and cannot store fuel, making them particularly vulnerable to any disruption in their adenosine triphosphate supply. The team hypothesized that if amyloids could break down this vital molecule, it would represent a completely new way these pathological structures exert their toxicity.

To investigate this possibility, the scientists conducted a series of experiments. First, they needed to confirm that adenosine triphosphate even interacts with the alpha-synuclein amyloids. They used a chemical reaction they had previously studied, where the amyloids break down a substance called para-nitrophenyl orthophosphate.

When they added adenosine triphosphate to this mixture, the original reaction stopped. This competitive effect suggested that adenosine triphosphate was binding to the same active location on the amyloid surface, pushing the other substance out of the way.

Having established that adenosine triphosphate binds to the amyloids, the researchers then tested whether it was being broken down. They mixed prepared alpha-synuclein amyloids with a solution of adenosine triphosphate and used a diagnostic tool called the Malachite Green assay, which changes color in the presence of free phosphate, a byproduct of adenosine triphosphate breakdown.

They observed a steady increase in free phosphate over time, confirming that the amyloids were indeed cleaving the phosphate bonds in adenosine triphosphate. This activity was catalytic, meaning a single amyloid structure could process many molecules of adenosine triphosphate, one after another. The same experiment performed with individual, non-clumped alpha-synuclein proteins showed no such effect, indicating this energy-draining ability is a feature specific to the aggregated, amyloid form.

To understand the mechanism behind this chemical activity, the team used a powerful imaging technique known as cryogenic electron microscopy. This method allowed them to visualize the structure of the alpha-synuclein amyloid at a near-atomic level of detail while it was bound to adenosine triphosphate.

The resulting images revealed a remarkable transformation. The amyloid itself was formed from two intertwined filaments, creating a cavity between them. When adenosine triphosphate entered this cavity, a normally flexible and disordered segment of the alpha-synuclein protein, consisting of amino acids 16 through 22, folded into an ordered beta-strand. This newly formed structure acted like a lid, closing over the cavity and trapping the adenosine triphosphate molecule inside.

This enclosed pocket was lined with several positively charged amino acids called lysines. Since the phosphate tail of adenosine triphosphate is strongly negatively charged, these lysines likely serve to attract and hold the energy molecule in a specific orientation. The structure suggested that this induced-fit mechanism, where the amyloid changes its shape upon binding its target, was a key part of its chemical function.

To prove that these specific lysine residues were responsible for the activity, the researchers genetically engineered several mutant versions of the alpha-synuclein protein. In each version, they replaced one or more of the key lysines in the cavity with a neutral amino acid, alanine. These mutant proteins were still able to form amyloid clumps that looked similar to the original ones.

When they tested the mutant amyloids for their ability to break down adenosine triphosphate, they found the activity was almost completely gone. This result confirmed that the positively charged lysines are essential for the amyloid’s ability to perform the chemical reaction.

In a final step, the scientists solved the high-resolution structure of one of the inactive mutant amyloids (K21A) while it was bound to adenosine triphosphate. The images showed that the energy molecule could still sit in the cavity, but its orientation was different from that seen in the active, non-mutant amyloid.

More importantly, in this inactive complex, the flexible protein segment did not fold over to form the enclosing lid. This finding provided strong evidence that both the proper positioning of adenosine triphosphate by the lysines and the structural rearrangement that closes the cavity are necessary for the breakdown to occur.

The study does have some limitations. The experiments were conducted in a controlled laboratory setting, not in living cells or organisms. The specific structural form of the alpha-synuclein amyloid studied, known as polymorph type 1A, has not yet been identified in the brains of Parkinson’s patients, although similar structures exist.

Also, the rate at which the amyloids broke down adenosine triphosphate was slow compared to natural enzymes. Future research will need to determine if this process occurs within the complex environment of a neuron and if other, more clinically relevant amyloid forms share this toxic capability.

Despite these caveats, the findings introduce a new and potentially significant mechanism of neurodegeneration. The researchers suggest that even a slow reaction could have a profound local effect. An amyloid plaque contains a very high density of these active sites. This could create a zone of severe energy depletion in the immediate vicinity of the plaque, disabling essential cellular machinery.

For instance, cells use chaperone proteins that require adenosine triphosphate to try to break up these very amyloids. If the chaperones approach an amyloid plaque and enter an energy-depleted zone, their rescue function could be disabled, effectively allowing the plaque to protect itself and persist. This work transforms the view of amyloids from passive obstacles into active metabolic drains, opening new avenues for understanding and potentially treating Parkinson’s disease.

The study, “ATP Hydrolysis by α-Synuclein Amyloids is Mediated by Enclosing β-Strand,” was authored by Lukas Frey, Fiamma Ayelen Buratti, Istvan Horvath, Shraddha Parate, Ranjeet Kumar, Roland Riek, and Pernilla Wittung-Stafshede.

A new analysis suggests that when it comes to the mental health benefits of urban green spaces, a moderate amount is best. The research, which synthesized four decades of studies, found that the relationship between the quantity of greenery and mental well-being follows an inverted U-shaped pattern, where benefits decline after a certain point. This finding challenges the simpler idea that more green space is always better and was published in the journal Nature Cities.

Researchers have long established a connection between exposure to nature and improved mental health for city dwellers. However, the exact nature of this relationship has been unclear. Bin Jiang, Jiali Li, and a team of international collaborators recognized a growing problem in the field. Early studies often suggested a straightforward linear connection, implying that any increase in greenness would lead to better mental health outcomes. This made it difficult for city planners to determine how much green space was optimal for public well-being.

More recent studies started to show curved, non-linear patterns, but because they used different methods and were conducted in various contexts, the evidence remained fragmented and inconclusive. Without a clear, general understanding of this dose-response relationship, urban planners and policymakers lack the scientific guidance needed to allocate land and resources to maximize mental health benefits for residents. The team aimed to resolve this by searching for a generalized pattern across the entire body of existing research.

To achieve their goal, the scientists conducted a meta-analysis, a type of study that statistically combines the results of many previous independent studies. Their first step was a systematic search of major scientific databases for all empirical studies published between 1985 and 2025 that examined the link between a measured “dose” of greenness and mental health responses. This exhaustive search initially identified over 128,000 potential articles. The researchers then applied a strict set of criteria to filter this large pool, narrowing it down to 133 studies that directly measured a quantitative relationship between greenness and mental health outcomes like stress, anxiety, depression, or cognitive function.

From this collection of 133 studies, the team focused on a subset of 69 that measured the “intensity” of greenness, as this was the most commonly studied variable and provided enough data for a robust analysis. They further divided these studies into two categories based on how greenness was measured. The first category was “eye-level greenness,” which captures the amount of vegetation a person sees from a ground-level perspective, such as when walking down a street. The second was “top-down greenness,” which is measured from aerial or satellite imagery and typically represents the percentage of an area covered by tree canopy or other vegetation.

A significant challenge in combining so many different studies is that they use various scales and metrics. To address this, the researchers standardized the data. They converted the mental health outcomes from all studies onto a common scale ranging from negative one to one. They also re-analyzed images from the original papers to calculate the percentage of greenness in a consistent way across all studies. After standardizing the data, they extracted representative points from each study’s reported dose-response curve and combined them into two large datasets, one for eye-level greenness and one for top-down greenness.

With all the data points compiled and standardized, the researchers performed a curve-fitting analysis. They tested several mathematical models, including a straight line (linear model), a power-law curve, and a quadratic model, which produces an inverted U-shape. The results showed that for both eye-level and top-down greenness, the quadratic model was the best fit for the collective data. This indicates that as the amount of greenness increases from zero, mental health benefits rise, reach a peak at a moderate level, and then begin to decline as the amount of greenness becomes very high.

The analysis identified specific thresholds for these effects. For eye-level greenness, the peak mental health benefit occurred at 53.1 percent greenness. The range considered “highly beneficial,” representing the top five percent of positive effects, was between 46.2 and 59.5 percent. Any positive effect, which the researchers termed a “non-adverse effect,” was observed in a broader range from 25.3 to 80.2 percent. Outside of this range, at very low or very high levels of eye-level greenness, the effects were associated with negative mental health responses.

The findings for top-down greenness were similar. The optimal dose for the best effect was found to be 51.2 percent. The highly beneficial range was between 43.1 and 59.2 percent, and the non-adverse range spanned from 21.1 to 81.7 percent. These specific figures provide practical guidance for urban design, suggesting target percentages for vegetation cover that could yield the greatest psychological rewards for communities.

The researchers propose several reasons why this inverted U-shaped pattern exists. At very low levels of greenness, an environment can feel barren or desolate, which may increase feelings of stress or anxiety. As greenery is introduced, the environment becomes more restorative.

However, at extremely high levels of greenness, a landscape can become too dense. This might reduce natural light, obstruct views, and create a feeling of being closed-in or unsafe, potentially leading to anxiety or a sense of unease. A dense, complex environment may also require more mental effort to process, leading to cognitive fatigue rather than restoration. A moderate dose appears to strike a balance, offering nature’s restorative qualities without becoming overwhelming or threatening.

The study’s authors acknowledge some limitations. By combining many diverse studies, some nuance is lost, as different populations, cultures, and types of mental health measures are grouped together. The analysis was also limited to the intensity of greenness; there was not enough consistent data available to perform a similar analysis on the frequency or duration of visits to green spaces, which are also important factors.

Additionally, very few of the original studies examined environments with extremely high levels of greenness, so the downward slope of the curve at the highest end is based more on statistical prediction than on a large volume of direct observation.

Future research could build on this foundation by investigating these other dimensions of nature exposure, such as the duration of visits or the biodiversity within green spaces. More studies are also needed that specifically test the effects of very high doses of greenness to confirm the predicted decline in benefits. Expanding this work to differentiate between types of vegetation, like trees versus shrubs or manicured parks versus wilder areas, could provide even more refined guidance for urban planning.

Despite these limitations, this comprehensive analysis provides a new, evidence-based framework for understanding how to design healthier cities, suggesting that the goal should not simply be to maximize greenness, but to optimize it.

The study, “A generalized relationship between dose of greenness and mental health response,” was authored by Bin Jiang, Jiali Li, Peng Gong, Chris Webster, Gunter Schumann, Xueming Liu, and Pongsakorn Suppakittpaisarn.