Trust and well-being, sharing at work, nighttime baby care, and more research

People with higher levels of trust are more likely to report greater well-being than those who are less trusting, according to research in Psychological Bulletin. Researchers conducted a meta-analysis of 241 studies on the link between trust and well-being. The studies involved more than 2.5 million participants, ages 6 to 84, from around the world. The researchers found that people who put more trust in others, institutions, or society in general felt happier and more satisfied with life than people who trust less. Some of the studies examining long-term trends suggested a virtuous cycle between trust and well-being: Not only did trust strengthen well-being, but feeling good engendered more trust in people and institutions over time.
DOI: 10.1037/bul0000480

What people share at work—and how it’s received—matters more than most people think, suggests research in Emotion. Across two studies, 358 participants in the United States reported on 26,324 workplace conversations, including what they shared, why they shared it, how others responded, and how they felt afterward. Overall, sharing positive personal information and sharing to connect were associated with increased emotional well-being, greater energy to work, stronger feelings of connection, and lower levels of dwelling on negative thoughts. By comparison, sharing negative information or venting predicted worse outcomes, including increased rumination and lower well-being. When sharers perceived colleagues’ responses as matching their motivation (e.g., offering emotional support in response to an emotional disclosure), they experienced boosts in mood and connection. But when sharers perceived a mismatched response, as occurred about half the time, they reported decreased well-being and increased rumination.
DOI: 10.1037/emo0001539

There’s little difference between how well men and women hear crying babies, despite mothers performing 3 times more nighttime infant care than fathers, according to a study in Emotion. In a study with 142 adults in Denmark without children, women were 14% more likely to wake up to a whisper-level sound, whether it was a baby crying or a regular alarm sound. However, once the volume increased, there was no significant difference between men and women. In a second study, 117 first-time parents in Denmark tracked their nighttime caregiving over a week. The results showed that mothers were 3 times more likely to attend to their babies at night than fathers were. The study authors suggest a small part of the discrepancy may be explained by two factors: Mothers generally take maternity leave before fathers take paternity leave, therefore gaining earlier experience soothing their babies, and when mothers breastfeed, they are usually the parent who must wake up at night anyway.
DOI: 10.1037/emo0001478

Couples with a shared worldview experience less uncertainty and find greater meaning in their lives, suggests research in the Journal of Personality and Social Psychology. Researchers conducted five studies of 1,288 adults in the United States and Canada, pooling data from lab-based tasks, online surveys, and experiments to investigate the connection between shared reality, uncertainty, and meaning in life. The first study found that couples who perceived a higher shared reality—feelings, attitudes, and beliefs in general—with their partner also reported greater meaning in life. Results from the second and third studies, which examined shared realities around racism and the COVID-19 pandemic, indicated that a shared worldview decreased uncertainty, which in turn was associated with a greater sense of meaning in life. The fourth and fifth experimental studies emphasized how reduced uncertainty helped connect shared experiences with a strong sense of meaning in life.
DOI: 10.1037/pspi0000472

Individuals with borderline personality disorder (BPD) exhibit difficulties in both interpersonal emotion regulation and working memory, particularly when faced with social distractions, according to a study in Personality Disorders: Theory, Research, and Treatment. Researchers asked 70 women with BPD symptoms to rate how often they used interpersonal emotion regulation strategies. The women also performed a working memory task with neutral, happy, fearful, and angry faces as distractors. Women with more severe BPD symptoms reported less frequent use of the strategies, especially those intended to enhance positive affect through interactions with others, such as by smiling. They made more mistakes in the task when distracted by faces, particularly happy faces. Better performance during distractor-free trials was associated with more frequent use of enhancing positive affect. The findings suggest that people with BPD are more easily disturbed by others’ happy expressions, perhaps because they perceive positive social cues as more distracting. This could partially explain why they experience less positive affect when interacting with others.
DOI: 10.1037/per0000722

When we forgive others for a past offense, we experience less intensity in our negative feelings about the event, suggests research in the Journal of Experimental Psychology: General. In two online studies, 452 participants recalled real-life personal offenses they had either forgiven or not. They also completed questionnaires about the recall of the event, to which the researchers assigned a vividness score. Participants who had forgiven the person who wronged them reported less negative affect when remembering the event. In the third study, 595 participants were randomly assigned the role of victim or perpetrator and asked to evaluate incidents of wrongdoing. The results were largely similar to those of the previous studies. The emotional nature of their recall was more neutral and less intense if they had forgiven. In the fourth study, forgiveness among 432 participants was associated with less desire for revenge and avoidance and more feelings of goodwill.
DOI: 10.1037/xge0001787

By focusing on individual-level decisions meant to reduce greenhouse gas emissions, people could become less committed to collective carbon-mitigation actions, suggests research in PNAS Nexus. Researchers randomly assigned 3,895 participants in the United States to one of three experimental conditions: an active learning condition in which they were asked to rank the relative effectiveness of 21 climate-mitigation behaviors and then received feedback; an information condition in which they were passively exposed to information about the relative mitigation potential of the 21 behaviors; and a no-information control condition. After the interventions, people in both the active learning and passive groups expressed less commitment to behaviors whose effects were overestimated (e.g., recycling) and more commitment to behaviors whose effects were underestimated (e.g., less meat consumption and airplane travel) in their carbon reduction potential. However, participants in both interventions focused on individual-level behaviors also reported being less likely to commit to climate-related collective actions like voting or joining public demonstrations.
DOI: 10.1093/pnasnexus/pgaf191opens in new window

Most people believe they are better than others at overcoming social pressure, even if they know that social scientific research indicates most people cave to such pressure, according to a study in Current Psychology. Researchers presented 414 online participants with a first-hand account from the classic 1963 Milgram shock study before asking them to predict the level of shock they and the “average person” would deliver in such a study. Before making their predictions, only half were told that 65% of those in the original Milgram research were completely obedient. This group predicted that they would deliver less voltage (and the average person would deliver more) than the other half of participants. These results were almost identical to obedience levels reported by previous studies and fit the researchers’ hypothesis that most people would underpredict their likely obedience in a Milgram-like study.
DOI: 10.1007/s12144-025-07962-1opens in new window

Epiphanies aren’t just satisfying—they shape how our brains represent information and help sear this knowledge into memory, according to a study in Nature Communications. Researchers asked 31 participants in Germany to solve puzzles—finding “hidden pictures” in visually ambiguous images while undergoing functional magnetic resonance imaging (fMRI) brain scans—and indicate their perceived suddenness, emotion, and certainty (signs of insight) about the solution. fMRI results during puzzle solving indicated stronger activity changes in participants’ visual cortex areas, accompanied by activation in the amygdala and hippocampus when their solutions were insights. Activity changes in participants’ ventral occipito-temporal cortex reflected how their brains reorganized perception after identifying previously ambiguous images. Greater insight was also associated with enhanced functional connectivity between the hippocampus and higher-level cortical areas. The connection between insight generation and hippocampal activity, which plays a role in memory, was further supported by the fact that participants better remembered insight-based solutions than analytically derived answers in a memory test 5 days later.
DOI: 10.1038/s41467-025-59355-4opens in new window

Scientists have created a computational model called Centaur that can predict and simulate human behavior across a wide variety of psychological experiments with unprecedented accuracy, according to a study in Nature. Researchers fed Centaur data from more than 60,000 people who made more than 10 million decisions in 160 psychology experiments to train it to capture underlying patterns in human thought, learning, and decision-making. Centaur not only captured the behavior of participants who were not part of the training data better than existing cognitive models; it also generalized across various cognitive tasks, from reasoning to decision-making, accurately simulating human behavior even in variations in tasks not previously encountered. Furthermore, with some fine-tuning, the model’s internal representations came to resemble human brain activity even though it was never explicitly trained to match neural data. Finally, Centaur’s analysis of human behavior patterns led to the identification of a decision-making strategy that combines two heuristic strategies in a way that researchers had not considered before. This combined strategy explained human behavior more accurately and outperformed other heuristic strategies usually considered.
DOI: 10.1038/s41586-025-09215-4opens in new window

People consistently rate empathy as more genuine, supportive, and emotionally satisfying when they believe it comes from a human, even when the content is identical to an AI-generated response, suggests research in Nature Human Behaviour. Researchers conducted nine studies in which AI-generated empathic responses to 6,282 online participants’ emotional situations were labeled as being generated by either another person or AI. Participants rated responses attributed to humans as more empathic and supportive, and as eliciting more positive and fewer negative emotions, than AI-attributed ones. Moreover, when participants believed that AI had aided the human responses, perceived empathy and support plummeted. These effects were found across various AI models.
DOI: 10.1038/s41562-025-02247-wopens in new window

People who work in the arts, design, entertainment, sports, and media in the United States have higher rates of lifetime diagnosed depression and more frequent mental distress than employees in other sectors, according to a study in JAMA Network Open. Researchers analyzed survey data from 469,129 workers in 37 states collected between 2015 and 2019 that captured industry and occupation, demographic information, and mental health outcomes. Overall, 14.2% of participants reported lifetime depression, and 9.6% reported frequent mental distress. Frequent mental distress prevalence was 1.32 times higher among workers in the arts, design, entertainment, sports, and media compared with workers in other fields. Other occupations with high mental distress were food service, health care workers, sales, and building and grounds maintenance.
DOI: 10.1001/jamanetworkopen.2025.14212opens in new window

Esketamine (Spravato), used alone, appears to improve depressive symptoms of treatment-resistant depression (TRD), according to a study in JAMA Psychiatry. Researchers conducted a double-blind, randomized clinical trial of 378 adults with TRD at 51 outpatient centers in the United States. Participants, who had been antidepressant free for 2 or more weeks before the study onset, were administered either 56-mg or 84-mg doses of esketamine twice weekly for 4 weeks. Both groups experienced statistically significant and clinically meaningful improvement of depressive symptoms compared with a placebo group. The positive results were initially observed 24 hours after the first dose and were maintained through Day 28 of the double-blind treatment phase. The most common side effects were nausea, dissociation, dizziness, and headache.
DOI: 10.1001/jamapsychiatry.2025.1317opens in new window

Students who rely on generative AI to help them learn may be missing out on basic skills, according to a study in PNAS. Researchers randomly assigned nearly 1,000 high school math students in Turkey to one of two AI chatbots: GPT Base, a chat interface similar to ChatGPT-4 that helps during practice sessions by providing correct answers; or GPT Tutor, an interface similar to ChatGPT-4 but with safeguards, such as teacher-designed hints rather than directly giving away answers. A control group had no technology assistance and relied only on textbooks and notes. During the AI-assisted practice session, the GPT Base group performed 48% better than the control group. But when the GTP Base group was given an exam on the material without access to AI, they performed 17% worse than the control group. The GPT Tutor group performed 127% better in the practice session but scored about the same as the control group on the exam.
DOI: 10.1073/pnas.2422633122opens in new window

Although trigger warnings given to students ahead of potentially controversial course material have no significant impact, safe space notifications positively affect students’ sense of emotional security and their perception of instructors as more liberal, according to research in the Journal of Experimental Psychology: Applied. Researchers randomly assigned 738 undergraduate students in the United States to watch videos of instructors lecturing on trauma after providing either a trigger warning, a safe space notification, both, or neither. Participants rated each instructor’s trustworthiness, concern for student well-being, political orientation, and left-wing authoritarianism tendencies, as well as their feelings of psychological safety and willingness to discuss controversial topics in that classroom. Trigger warnings were found to have no overall impact on students’ perceptions. In contrast, safe space notifications increased their feelings of psychological safety and inclinations to discuss controversial topics. Safe spaces did increase students’ perceptions of instructors as caring and trustworthy but also liberal and as having left-wing authoritarian tendencies, including a desire to engage in censorship.
DOI: 10.1037/xap0000541

Children 24 months and older can form associations between words they hear and objects they see, even when the speaker’s facial expressions and other visual cues are partially obscured, according to research in Developmental Psychology. Researchers recorded eye gaze in 153 toddlers ages 17 to 42 months in France while the toddlers performed a word-learning task. In the task, a speaker used a novel word to name and occasionally look at one of two novel objects flanking them. The children performed the task in three situations: the speaker’s face fully visible, the speaker’s eyes covered with dark glasses, or the speaker’s mouth covered by a surgical mask. Toddlers aged 24 months and older learned the new word-object associations regardless of the situation. Moreover, toddlers’ word-learning performance was linked with gaze shifting between the objects and the speaker’s face, but not with attention on the speaker’s eyes or mouth. The findings suggest toddlers’ optimal attentional strategy around 24 months involves social understanding and object exploration rather than focusing on cues from a speaker’s face.
DOI: 10.1037/dev0002016

Consistently feeling close to the same set of friends on a daily basis is associated with greater emotional well-being in adolescents, according to research in Developmental Psychology. Researchers asked 195 11th graders in the United States to record the names of up to three friends they felt closest to and to assess their mood every day for 2 weeks. While noting considerable fluctuation in friendship closeness across time, the researchers found that greater daily continuity in close friendships was associated with higher daily positive mood in adolescents and not negative mood, at both the within-person and between-person levels.
DOI: 10.1037/dev0002014