![[Adobe Stock]](https://sp-ao.shortpixel.ai/client/to_webp,q_glossy,ret_img,w_750,h_375/https://www.psypost.org/wp-content/uploads/2025/09/happy-couple-750x375.jpg)
Sharing a room with a romantic partner or caregiver may help the human body regulate basic internal functions, including blood sugar levels, skin temperature, and stress arousal, with less overall effort than when alone.
Across four experiments with adults and infants published in Science Advances, participants experienced smaller bodily disruptions and faster recovery during physical challenges when accompanied by a close companion. The findings suggest that social connection does not merely influence emotions and behavior, but also supports core biological maintenance from moment to moment.
Rethinking homeostasis as a social process
Every living organism must keep its internal environment within a narrow, healthy range despite changing conditions, a biological balancing act known as homeostasis. Managing body temperature, digesting food, and regulating nervous system arousal all consume energy, so meeting those demands with smaller spikes and quicker recovery, in the authors’ framework, leaves more resources for other vital bodily functions.
Evolutionary biologists and psychologists typically view group living through external benefits such as protection from threats, cooperation, reproduction, and emotional comfort under stress. Study lead author Monia Masalha and senior author Shir Atzil, an associate professor of psychology at The Hebrew University of Jerusalem who leads the Atzil Lab, worked with their colleagues to test a broader principle they call social physiology. Under this view, the physical presence of a close companion directly reduces the bodily effort needed to maintain internal stability across distinct regulatory systems.
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.
“Humans are profoundly social, yet physiology is typically studied as if the body regulates itself in isolation,” Atzil told PsyPost. “We came to realize that this view overlooks something fundamental: social animals begin life completely dependent on others to regulate their physiological needs. Their physiology is socially regulated from the start, and the brain systems responsible for maintaining physiological balance develop within this social context.”
“This led us to ask whether social regulation remains a basic feature of human physiology in adulthood: do close social relationships actually change the efficiency of homeostatic regulation?” she added. “The question also has important implications for health. Social relationships are strongly associated with health and longevity, but the physiological mechanisms linking sociality to these outcomes remain incompletely understood.”
Testing that hypothesis required examining an automatic, energy-demanding bodily system that people cannot consciously control and that is not traditionally thought of as sensitive to social company. Blood sugar regulation after eating carbohydrates fits those criteria, as removing glucose from the bloodstream depends on involuntary metabolic feedback rather than deliberate behavior.
If social proximity improves regulatory efficiency across the body as a general principle, the researchers predicted that this advantage should emerge not only during stress, but also while digesting a carbohydrate meal and regulating temperature.
Testing blood sugar regulation alongside a partner
To test this prediction across systems and age groups, the researchers conducted an exploratory pilot study with mothers and infants, followed by three preregistered laboratory experiments with adults. The datasets and analysis code were posted publicly on the Open Science Framework.
In every experiment, the team quantified regulatory efficiency by measuring both how far a bodily signal deviated from its starting baseline during a challenge and how long it took to return toward normal. Combining the height and duration of that shift into a single cumulative response score allowed the authors to compare overall physiological effort during identical challenges completed alone versus alongside a close partner. Participants underwent both conditions on the same day in randomized order, and the team analyzed the results both within the same individuals and between separate groups using only each participant’s first session to avoid carryover effects. As preregistered, the team used one-directional statistical tests, which look only for an effect in the predicted direction.
Guided by the infant pilot results, the adult experiments were statistically powered to detect a small effect and drew from 180 healthy volunteers aged 18 to 37 who had been in monogamous heterosexual relationships for at least nine months. Because some couples participated in more than one experiment, the team collected 339 paired adult observations in a comfortable, living-room-style laboratory where couples could talk or touch naturally during social sessions.
In the first preregistered adult experiment, 108 participants (54 couples) who passed blood screening for normal three-month average blood sugar levels fasted overnight and ate a small plain biscuit two hours beforehand to standardize starting glucose. In two sessions separated by at least four hours, once alone and once with their partner, each person ate 102 grams of white wheat bread providing 50 grams of carbohydrates while finger-stick blood tests tracked glucose over two hours.
Baseline blood sugar levels were comparable across conditions, yet glucose spikes were smaller and subsided faster when couples digested the meal together. Within the same individuals, cumulative elevated blood sugar above normal fasting levels was an average of 14.55% lower in the partner session than in the alone session. Comparing only first sessions across groups, the 54 participants tested with their partner showed a 23.9% lower cumulative glucose response than the 54 tested alone.
“For glucose, the effect was sizable for a single standardized meal,” Atzil remarked. “That is an average effect; some individuals benefited considerably more, and some less.”
To examine whether partner presence simply reduced stress during the meal challenge, the researchers continuously tracked electrodermal activity, which measures tiny changes in skin electrical conductance caused by sweat glands. Those glands are controlled by the sympathetic nervous system, the branch governing automatic fight-or-flight arousal. Skin conductance actually dipped below baseline during the two-hour bread test rather than rising, and the difference between the alone and partner sessions was not statistically significant. Skin conductance was also unrelated to blood sugar levels, and accounting for either skin conductance or video-coded physical touch left the partner advantage in glucose regulation intact.
Effects across temperature control, stress, and infancy
To determine whether this efficiency extended to a separate homeostatic pathway, a second preregistered experiment shifted to temperature regulation in 116 adults (57 couples and two individuals whose partners were excluded). Wearing lightweight clothing and temperature-sensing wristbands on both hands, participants spent five minutes at room temperature, five minutes in a cold room set to 12 degrees Celsius (about 54 degrees Fahrenheit), and 15 minutes warming back up.
Skin temperature dropped less and rebounded faster when couples entered the cold room together. The cumulative temperature drop was 16.17% smaller between first-session groups and 6.76% smaller within individuals, though the within-person difference passed only the one-directional test. Video coding showed a small association between greater physical contact and milder temperature drops, and adjusting for touch left a modest social effect that likewise passed only the one-directional test. Room thermometers confirmed that having a second person present did not materially warm the cold room itself.
A third preregistered experiment examined rapid sympathetic arousal in 115 adults (57 couples and one individual). Because the team used a single electrical stimulator, partners took turns rather than undergoing the test simultaneously, receiving two brief, mild electric shocks to the forearm during a one-minute stress period while skin conductance was recorded. Comparing first sessions across groups, cumulative skin conductance arousal was 5.98% lower with a partner present than when alone, though a smaller 1.89% reduction within the same individuals across repeated sessions was not statistically significant.
When the authors standardized and pooled all 339 adult observations across glucose, temperature, and stress, cumulative physiological expenditure was 7.59% lower within individuals and 12.07% lower between groups in the social condition. That pattern persisted after accounting for relationship satisfaction, body mass index, time of day, and session order, though individual responses varied and a subset of participants showed less efficient regulation with their partner present.
The consistency of the results surprised Atzil. “I actually hypothesized the glucose effects in a paper back in 2018,” Atzil noted. “But seeing Social Physiology consistently across every homeostatic system we tested was overwhelming.”
“What makes this especially interesting is that the same general pattern appeared across temperature regulation and stress,” she explained. “If being with a close partner repeatedly improves regulation across several physiological systems, day after day, those gains could accumulate into a meaningful reduction in the body’s overall regulatory burden over time.”
The researchers also observed a parallel stress-buffering pattern early in life during the exploratory pilot study, conducted in family homes with 56 infants aged 2.5 to 13 months and their mothers (after excluding 14 pairs for incomplete sensor recordings and two babies who grew too distressed while alone to continue). Wearing wrist sensors, infants heard a two-second, 82-decibel sound of a falling metal pipe, once while their mothers stayed close and could hold or comfort them, and once while their mothers waited out of sight and earshot in another room. Cumulative skin conductance responses were 2.55% lower within infants when mothers were nearby (meeting the team’s one-directional threshold, though not statistically significant on a two-sided test) and 4.24% lower between first-session groups, with no link to infant age.
How the findings fit with prior research
“My main takeaway is that the presence of a close social partner may directly influence how an individual’s body maintains physiological balance, or homeostasis,” Atesh Koul, a postdoctoral researcher at the University of Ferrara and the Italian Institute of Technology who was not involved in the study, told PsyPost. “Research in psychology and neuroscience has traditionally focused on one person at a time. This study extends the idea that physiological regulation is shaped by the presence of another person.”
Martin Picard, a Science of Health professor in the Departments of Psychiatry and Neurology at Columbia University Irving Medical Center, the Robert N. Butler Columbia Aging Center, and the Energy and Healing Institute, leads the Picard Lab and was not involved in the research. “Social factors that influence how we feel—not only the intrinsic biological processes—change the way energy flows in our bodies,” Picard told PsyPost. He added that the study “extends previous findings that mental stress can exacerbate glycemic control—how well the body handles sugar.” “The replication of the main finding across multiple cohorts adds strength to the finding,” he said.
These stress findings in adults and babies align with an earlier virtual reality experiment with women, which found that a virtual character dampened fear-induced skin conductance regardless of social concern, whereas a live stranger reduced arousal only among women with low social concern. That earlier study, however, assessed acute fear reactivity beside a stranger or digital avatar rather than homeostatic efficiency alongside a close partner.
At the same time, the sympathetic arousal results stand in some tension with an earlier study of everyday social interactions among college students, where higher depression and loneliness were associated with a lack of physiological arousal reduction during intimate time with romantic partners or close friends. That study tracked unstructured daily socializing among students with elevated distress rather than controlled physical challenges in healthy volunteers.
“Just as cognitive psychology and neuroscience have increasingly turned toward studying social processes in realistic, multiperson settings, this interesting study brings a similar perspective to physiological homeostasis,” Giacomo Novembre, principal investigator of the Neuroscience of Perception and Action Lab at the Italian Institute of Technology in Rome, who was not involved in the study, told PsyPost. “The finding that homeostatic regulation can be enhanced by social proximity is consistent with extensive evidence linking social connection to health in humans and other social species. It also offers a physiological explanation for some of the evolutionary benefits of sociality.”
“Its main contribution is to move from long-term associations to immediate physiological effects,” Koul added. “Rather than asking whether social connectedness predicts health over months or years, the study asks whether the presence of a close partner changes how the body responds to a specific challenge in the short term.”
Possible mechanisms and methodological strengths
In the authors’ interpretation, social proximity may improve bodily efficiency through several overlapping routes, including physical heat sharing, threat buffering, and predictive regulation in the brain. Under this predictive process, known as allostasis, the brain may use familiar social cues to anticipate upcoming bodily needs and curb physiological swings early. The researchers propose that this immediate energetic benefit could help explain both the evolution of social bonds and links between social connection and metabolic health.
“The main message is that human physiology is social,” Atzil said. “Social context is not merely psychological; it can influence basic physiological regulation, including glucose homeostasis.”
She sees implications for health. “We have identified a completely new dimension that influences basic bodily regulation: our bodies do not regulate themselves entirely in isolation,” Atzil said. “If social proximity changes how efficiently we regulate processes such as glucose, temperature and stress, this could have important implications for our health and longevity.”
“Social Physiology may help explain why we are so strongly motivated to form close bonds,” she continued. “Social Physiology provides an immediate benefit to being together: being with close others can make physiological regulation easier and less costly, and those gains may themselves reinforce social closeness.”
Outside experts who commented on the study pointed to its multi-experiment scope and extensive checks against confounding variables as methodological strengths, while noting that naturalistic partner interactions leave open questions about which specific cues matter most.
“I have moderately high confidence in the findings,” Koul said. “The experiments were preregistered, used reasonably large samples, and examined three physiological systems. The authors also performed a valuable series of control analyses addressing alternative explanations, including condition order, physical touch, body mass, room temperature, and relationship satisfaction. However, because several aspects of social interaction varied simultaneously, further research is needed to identify which aspects of a partner’s presence produce these effects and through what mechanisms.”
“A particular strength of the study is that partners were free to talk and touch at their discretion, capturing features of everyday interaction that more constrained experimental designs often exclude,” Novembre said. “This openness, however, makes it difficult to identify which channels of communication—or combinations of channels—contributed to the observed effects. Although the authors examined touch, it did not conclusively account for their findings. Investigating the contributions of other channels and their interplay is therefore a promising next step, connecting these results with ongoing research in social neuroscience and psychology.”
Caveats, clinical boundaries, and next steps
There are a few things to keep in mind when reading these findings. According to the paper’s competing-interests statement, Atzil is exploring intellectual property protection related to applications of social physiology. Each test measured only a brief window of regulation on a single day, so the study cannot establish whether modest short-term efficiencies of roughly 2% to 15% within individuals persist over time or prevent chronic conditions such as type 2 diabetes. The adult sample was also restricted to healthy young volunteers in established heterosexual relationships, leaving open how bodies respond around acquaintances, friends, or conflicted partners, especially since some participants regulated less efficiently with their partner in the room.
“It is too early to translate these findings into specific clinical recommendations,” Atzil cautioned. “We are not suggesting that eating with someone close is a treatment for diabetes or that patients should change their glucose management. Rather, the study identifies a new dimension that can affect glucose metabolism, and now needs to be tested in people with metabolic disease.”
“These findings should not yet be given a clinical interpretation,” Koul noted. “The study examined short-term responses under controlled conditions. It did not test whether these effects improve health, prevent disease, or accumulate over time. The proposed links to health and evolution are promising and theoretically interesting, but they remain empirically distant.”
Having another person nearby also adds complexity to the setting, which the authors note could either help the body resolve a challenge or compete with it for regulatory resources. Future longitudinal studies examining wider age ranges, diverse relationships, and hormonal markers will be needed to trace the exact biological pathways behind social physiology.
“The individual differences may be especially important,” Atzil added. “Some relationships may provide greater physiological gains than others. It is interesting to think: how do different relationships affect us metabolically?”
“Future studies should determine whether the effect depends on the identity of the partner, the quality of the relationship, or the type of interaction,” Koul said. “For example, do strangers produce similar effects, and is physical presence necessary, or is seeing someone, for instance, through a webcam, sufficient?”
Picard raised the possibility that social proximity could reach down to the cellular level, including mitochondria, the tiny structures inside cells that convert nutrients into usable energy.
“What is it about being with someone you love that changes your state of being and the energetic state of your body?” Picard asked. “Could this also influence the state of your mitochondria? Your ability to heal?”
On the developmental side, because the infant pilot study measured only nervous system arousal during noise, the study authors note that future work will need to test whether babies also show blood sugar and temperature gains alongside caregivers.
“Our long-term goal is to test the evolutionary hypothesis that Social Metabolic Gains shape our social preferences, influencing whom we choose to bond with,” Atzil said. “We also want to test whether these physiological gains accumulate over time and contribute to better health, both in healthy individuals and in people with metabolic disease. Social Physiology offers a candidate mechanism for a longstanding observation: social bonds are associated with better health and longevity. If social proximity repeatedly reduces the physiological cost of maintaining homeostasis, these small gains could have lasting health benefits.”
“Humans are extremely social animals,” she concluded. “It is exciting to discover a previously unknown benefit of social bonds that may have helped drive the evolution of our social nature.”
The study, “Social Physiology: Human homeostasis is more efficient in social proximity,” was authored by Monia Masalha, Matan Cohen, Maayan Aloni, Einav Toledo, Ella Stahl, Diane Shachar, Or Eliav, Shai Fuchs, and Shir Atzil.
