The first of the two studies set out to answer a question that has long interested developmental scientists: what actually happens inside a baby when a joyful exchange with a parent unfolds? To find out, the researchers measured changes in the heart rates of nearly 1,000 four-month-old infants as they engaged in positive interactions with their mothers. Electrodes were attached to the babies’ chests and stomachs, while trained observers watched the mother-infant pairs and recorded, moment by moment, whether each member of the pair displayed positive, neutral, or negative engagement. This combination of continuous physiological recording and fine-grained behavioral observation allowed the team to link what was happening between the mother and baby with what was happening inside the baby’s body.

The physiological measure at the center of the study was respiratory sinus arrhythmia, a form of heart-rate variability that is coupled with breathing. Scientists understand this variability to be a biological indicator tied to emotion regulation and to a person’s adaptability to different environments. In adults and in children, higher heart-rate variability in this range is generally associated with a more flexible capacity to respond to social and emotional demands. By calculating this measure from the infants’ heart-rate data, the researchers could ask whether moments of shared positivity were accompanied by shifts in the biological systems that support emotional self-regulation.

The answer was striking. The researchers found increases in the babies’ heart-rate variability precisely when the infants and their mothers were positively engaged at the same time. Crucially, the same physiological changes did not appear when only the baby was smiling, or when only the mother was showing positive emotion. Both partners had to be on the same positive wavelength for the effect to emerge. Daniel Messinger, a professor in the Department of Psychology at the University of Miami College of Arts and Sciences and a senior author on the paper, put it plainly: “They both have to be positive. When they’re both positive, particularly for longer periods of time, you see this increase in the baby’s heart rate.”

That mutual requirement may be the study’s most important insight. It suggests that emotional development is not simply a matter of a happy baby or an affectionate parent acting alone. Instead, the physiological states associated with emotion regulation appear to be recruited during genuine two-way exchanges, in which both the caregiver and the infant are sharing positive affect simultaneously. “Play is not just fun and games. It’s influencing the baby’s physiology,” said Messinger, who is also a professor of pediatrics. “There’s something about positive sociality that’s associated with this internal baby state. The baby’s physiology is linked to what’s going on in the interaction.” In other words, the interaction itself, not the mood of either partner in isolation, is what seems to matter.

The scale and design of the study lend weight to its conclusions. The research was part of a national study on drug exposure conducted at four sites across the country, including Miami, and involved several universities. When the team analyzed the heart-rate data, they controlled for prenatal drug exposure and social adversity, helping to ensure that the observed link between shared positivity and heart-rate variability was not simply an artifact of those factors. Among the members of the research team was Samantha G. Mitsven, a University of Miami alumna who worked on the study while completing her Ph.D. in developmental and child psychology. With nearly 1,000 infants contributing data, the study represents one of the larger physiological investigations of early mother-infant interaction.

The second study, also published in Philosophical Transactions of the Royal Society B, approached infant smiling from a different angle by asking how babies and adults influence each other’s smiling over time. The researchers observed 58 infants at two ages, 4 months and 8 months, while the babies interacted with their mothers, their fathers, and a stranger. This design allowed the team to compare smiling dynamics across different relationship partners within the same infants, revealing patterns that would be invisible in studies focused on mothers alone.

The results showed a clear division of labor in early smiling exchanges. Adults initiated and reciprocated smiling interactions at higher rates than infants did, and seeing an adult smile significantly increased the rate at which babies began smiling themselves. Babies, on the other hand, were more likely to be the ones who ended shared smiling moments, a pattern the researchers interpret as a possible sign that infants are learning to moderate their own excitement. The team also found developmental change: at 8 months, the babies were both more likely to start a smiling interaction on their own and more likely to reciprocate an adult’s smile compared with when they were 4 months old. “That suggests that the dyad becomes more dynamically responsive to one another at older ages,” said Messinger, who was a first author on the paper along with Eliala Alice Salvadori, who conducted the research for her dissertation at the University of Amsterdam, and Cristina Colonnesi, an associate professor at the University of Amsterdam. Another first author, Alan B. Cobo-Lewis, completed his postdoctoral fellowship and undergraduate studies at the University of Miami and is now an associate professor at the University of Maine.

Perhaps the most socially revealing finding came from the comparison across partners. At both ages, babies were most likely to return their mother’s smiles, compared with smiling interactions involving their fathers or an unknown adult. The adults who were strangers actually initiated and sustained smiles more often than the mothers did, yet the infants responded preferentially to their mothers. “This was, importantly, the first paper that involved not only moms and dads, but also strangers,” Messinger noted. “The strangers initiate and sustain smiles more than moms, but infants are more likely to smile back at the mom. So, the relationship really counts there.” The finding underscores that infant smiling is not a reflexive response to any friendly face, but a selectively tuned behavior shaped by the specific relationship between the baby and the person in front of them.

Together, the two studies sketch a picture of early emotional development as a genuinely interactive process. The heart-rate study shows that mutual positive engagement, sustained over time, is associated with increases in a physiological indicator linked to emotion regulation, while the smiling study shows that these exchanges become more fluid, reciprocal, and relationship-specific as infants grow from 4 to 8 months of age. For parents, the practical message is encouraging rather than demanding: ordinary moments of shared laughter and smiling are not idle pleasures but may be part of the biological work of early development. For scientists, the results open a window onto how social experience and physiology intertwine in the first months of life, suggesting that the foundations of emotional well-being are built not in solitude, but in the back-and-forth of a shared smile.

Subject of Research: The role of shared positive caregiver-infant interactions in infants' physiological and emotional development

Article Title: Shared smiles may shape infants’ emotional development

Article References: Shared smiles may shape infants’ emotional development. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: infant development, emotion regulation, heart-rate variability, respiratory sinus arrhythmia, mother-infant interaction, smiling, caregivers, developmental psychology, physiology, social interaction, University of Miami, Philosophical Transactions of the Royal Society B

The research, published in Physiological Reports, focused on high-frequency heart rate variability, or HF-HRV, a measure derived from electrocardiogram recordings that reflects beat-to-beat fluctuations in heart rhythm within a specific frequency band. HF-HRV is closely associated with respiratory sinus arrhythmia, the natural speeding and slowing of the heart with breathing, which in turn is linked to parasympathetic, or vagal, nervous system activity. Because this branch of the autonomic nervous system is thought to support calmness, social engagement, and emotional regulation, HF-HRV has become a popular window onto the physiological states of caregivers and children during interaction.

Previous studies of physiological synchrony have typically averaged heart rate variability over fixed time windows and compared those averages between partners. Such approaches can detect gross similarity but may miss a crucial feature: the possibility that one person’s physiological rhythm precedes or follows the other’s by a meaningful interval. Eye contact, vocal exchanges, and physical touch between parents and children can change within seconds, and the physiological consequences of those moments may ripple forward or backward in time. The Japanese team, led by Aoi Sakurai, set out to build and test a method that explicitly accounts for these temporal shifts.

The study recruited 15 mother-infant dyads with healthy infants aged five to seven months, recruited between December 2022 and March 2023 through local childcare newspapers and social media in Sendai City. Each pair was scheduled for two recording sessions roughly two weeks apart. Mothers were asked to feed and change their infants beforehand to ensure calm, and then both wore small wireless ECG sensors, positioned at the upper sternum for mothers and near the clavicular region for infants, while spending five minutes in unrestricted free play on a yoga mat, talking, touching, and playing as they naturally would.

Turning raw infant ECG into usable data proved to be the study’s greatest technical hurdle. Of the 30 recorded sessions, 14 were excluded because of device failures, session interruptions, or electrode detachment and movement artifacts, leaving 16 analyzable sessions from 10 unique dyads. The researchers processed the surviving signals with a custom MATLAB pipeline that combined automated QRS detection with careful visual inspection and manual correction, flagging suspicious beats as missing values rather than interpolating them. Artifact rates in the retained data were remarkably low, at roughly 0.29 percent for mothers and 0.02 percent for infants, a testament to the labor-intensive cleaning process.

The core of the analysis was a lag-sensitive cross-correlation technique. The team computed HF-HRV power in 30-second windows advanced in half-second steps, using frequency bands of 0.15 to 0.40 hertz for mothers and 0.24 to 1.04 hertz for infants, reflecting developmental differences in breathing rate. For each maternal reference window, the infant’s signal was shifted across a search range spanning minus 30 to plus 30 seconds, and the correlation between the two traces was calculated at every offset. When a positive correlation coefficient of 0.8 or higher emerged as a local peak, that moment was marked as a synchronization point, classified as Lag-plus if the infant’s pattern aligned after a positive offset, Lag-minus if the mother’s pattern aligned after a negative offset, or Lag-zero for exact alignment.

The results were striking. Highly correlated intervals accounted for a mean of 74.5 percent of the total analyzed interaction time, a figure well above the variable or lower synchrony levels often reported in previous mother-infant studies. More intriguing still was the distribution of lags: within the highly correlated time, Lag-minus intervals made up about 54.4 percent and Lag-plus intervals about 44.6 percent, while exact zero-lag alignment accounted for a mere 1.1 percent. In other words, under this analytical definition, mother-infant heart rhythm synchrony was almost always offset in time, and the direction of the offset varied both across sessions and within them, with some dyads showing a bias toward one direction over limited periods.

Because high correlations between two slowly fluctuating biological signals can arise by chance from autocorrelation and shared measurement conditions, the team also compared their observations against a null model. For each session, they generated 100 surrogate datasets by randomly permuting the order of 30-second segments within the maternal and infant time series, destroying the original correspondence while preserving local structure. Observed synchrony exceeded the session-specific surrogate distribution in 13 of the 16 sessions, with a mean observed-minus-surrogate difference of about 0.205, and the positive difference held across all 10 dyads in a sensitivity analysis. This suggests that the alignment was more than coincidental temporal structure, though the authors are careful to frame the surrogate test as exploratory, since it relied on a single null model and a modest number of iterations.

The theoretical backdrop is Polyvagal Theory, which positions the vagal system as central to physiological co-regulation in close relationships. The findings are consistent with the idea that caregivers and infants mutually regulate one another’s autonomic states, and they echo earlier reports of heart rhythm synchrony between pregnant women and their fetuses, between marital partners, and even between dogs and their owners. Yet the authors caution against overinterpretation. The lag categories should not be read as evidence that one partner behaviorally leads or regulates the other, and the study measured HF-HRV without recording respiration, so the results reflect temporal physiological coordination rather than respiration-controlled estimates of vagal activity.

The limitations are candidly acknowledged. The sample was small and limited to mothers and five- to seven-month-old infants, with no fathers, no baseline or still-face conditions, and no non-partner dyad pairings to rule out shared environment or common movement. No synchronized behavioral coding was performed, so the physiological alignment cannot yet be tied to specific events like eye contact, vocalization, or touch. The predefined plus-or-minus 30-second lag range may itself have shaped the findings. Still, the study demonstrates the feasibility of a lag-sensitive approach that could, if replicated in larger and more diverse samples, eventually help researchers probe whether physiological synchrony differs in populations facing psychosocial risk, such as mothers with postpartum depression, offering an objective, minimally invasive complement to the behavioral observations that have long dominated attachment research.

Subject of Research: Lag-sensitive analysis of mother-infant heart rate variability synchrony during free interaction

Article Title: Lag‐sensitive cross‐correlation analysis of mother–infant HF‐HRV synchrony during free interaction: An exploratory feasibility study

Article References: Sakurai, A., Koide, K., Kasahara, Y., Nakamura, Y., & Yoshizawa, T. (2026). Lag‐sensitive cross‐correlation analysis of mother–infant HF ‐ HRV synchrony during free interaction: An exploratory feasibility study. Physiological Reports, 14(18), Article e71116. https://doi.org/10.14814/phy2.71116

Image Credits: AI Generated

DOI: 10.14814/phy2.71116

Keywords: heart rate variability, mother-infant synchrony, HF-HRV, cross-correlation, autonomic nervous system, respiratory sinus arrhythmia, polyvagal theory, attachment, physiological co-regulation, surrogate data, free play, infant development

The study, led by Qian Wu and Wenbin Fu of the Second Affiliated Hospital of Guangzhou University of Chinese Medicine, together with Jiajin Yuan of Sichuan Normal University and colleagues, was designed around a deceptively simple question: what happens to the brain’s rapid emotional alarm system when the body is forced to run on nothing but water? Prolonged fasting exposes a person to a sustained energy deficit and a chronic psychological stress load, both of which are known in the broader literature to influence how people process negative and threatening information. Yet most previous work has relied on single snapshots, comparing fasted individuals with non-fasted controls at one moment in time. By measuring the same participants repeatedly across four phases, the team could watch the adaptation unfold within each brain rather than inferring it from differences between people.

The measurement tool at the heart of the study is the event-related potential, or ERP, a technique that averages the brain’s electrical responses to repeated stimuli recorded from the scalp. Because ERPs unfold over tens to hundreds of milliseconds, they reveal stages of processing that unfold far too quickly for conscious report. The researchers used a negative emotional picture Oddball task, a classic paradigm in which participants view a stream of images and occasionally encounter an unexpected, emotionally charged deviant. The deviant automatically pulls attention, and the brain’s response to it can be decomposed into well-characterized components. The occipital N1, arising roughly 100 milliseconds after stimulus onset over visual cortex, reflects early attentional resource recruitment. The fronto-central N2, peaking around 200 to 300 milliseconds, indexes attentional orienting to unexpected or conflicting information. The parietal P3, the latest of the three, is widely read as a marker of motivational salience and evaluative processing, the brain’s judgment of how much an event matters.

Alongside the EEG recordings, the team collected supplementary measures of autonomic flexibility. Before each Oddball session, participants rested while the researchers recorded photoplethysmograms, an optical measure of blood volume pulse, and respiration traces. From these they computed respiratory sinus arrhythmia, or RSA, the natural speeding and slowing of the heart that is synchronized with the breathing cycle. RSA is commonly interpreted as an index of parasympathetic vagal influence and, in the emotional domain, of the flexibility with which a person’s physiology can respond to changing emotional contexts. Lower RSA during a challenging period is generally taken to reflect reduced capacity for adaptive emotional engagement. Including this autonomic layer gave the researchers a second, independent line of evidence against which to check the patterns emerging from the scalp recordings.

Nineteen healthy adults carried the study through its demanding schedule. Each completed the emotional Oddball task at four points: at baseline before fasting began, on day seven of complete fasting, on day fourteen of complete fasting, and on day two of the recovery period, when normal eating had resumed. The entire experimental process was conducted under full medical supervision, with a professional medical team monitoring participants’ physical and mental condition throughout, and the protocol was approved by a biomedical research ethics committee in accordance with the Declaration of Helsinki. All participants provided informed consent. The longitudinal design allowed the researchers to analyze the data with linear mixed-effect models, a statistical framework that respects the repeated-measures structure of the data and estimates how each phase differs from baseline within the same individuals.

The ERP results revealed a phase-dependent pattern that the authors describe as a shift in the architecture of emotional reactivity. Early in the fast, at day seven, the dominant change was an amplification of the parietal P3, the component associated with salience evaluation. In other words, one week into total food deprivation, the participants’ brains appeared to assign heightened motivational significance to the negative emotional deviants, treating the disturbing images as more consequential than they had at baseline. This heightened evaluative response is consistent with the intuition that an energized alarm system accompanies the early stages of a metabolic challenge, when the body’s stress physiology is mobilized and the brain may be primed to prioritize threat-relevant information.

By day fourteen, however, the picture had changed. The later phase of the fast was characterized not by amplified evaluation but by an attenuated fronto-central N2, the component reflecting attentional orienting to unexpected information. After two weeks without food, the participants’ brains appeared less responsive to the surprise value of the emotional deviants, suggesting a blunting of the orienting response rather than a continued amplification of it. Meanwhile, across both fasting phases, at day seven and day fourteen, the participants showed increased occipital N1 responses to the explicit task demands, indicating greater recruitment of early attentional resources whenever the task required it. The brain, in this reading, was still capable of mobilizing visual attention when instructed, but its automatic, involuntary reactions to emotional surprise had shifted from amplification to dampening as the fast wore on.

Crucially, the story did not end with the fast. On day two of recovery, after participants had resumed eating, all of the observed ERP changes had returned to baseline levels. The heightened salience evaluation of the early phase and the attenuated orienting of the late phase were both reversible, disappearing once normal energy balance was restored. The autonomic data told a complementary story: RSA estimates supported a reduction in emotional context flexibility during the complete fasting period, again consistent with the idea that sustained metabolic stress constrains the physiological flexibility that supports adaptive emotional responding. The convergence of a cortical measure and a peripheral autonomic measure on the same conclusion strengthens the overall pattern, even though the two signals were collected in different states, one during active task performance and one at rest.

The authors are careful to frame the work as hypothesis-generating rather than definitive. With nineteen participants and no mention of a control group consuming a normal diet across the same schedule, the findings require replication in larger controlled studies before the identified neural and autonomic patterns can be considered for practical use. The researchers themselves emphasize that the ERP and RSA readouts should not yet be treated as validated monitoring tools. Still, the study points toward an intriguing application. If the phase of a fast can be read out from a person’s neural and autonomic responses, then neurobehavioral monitoring strategies could in principle be timed to the metabolic state of the individual, deploying countermeasures, whether nutritional, psychological, or pharmacological, at the moments when the brain’s emotional machinery is known to shift gear. Such considerations matter in contexts ranging from clinical fasting protocols to extreme occupational settings, though the authors stop short of claiming readiness for any of these uses.

What makes the study compelling beyond its immediate numbers is the conceptual shift it forces. Emotional reactivity under starvation is often assumed to move in a single direction, toward either irritability and hypervigilance or apathy and blunting. This work suggests the reality is a choreographed sequence: an early phase in which the brain magnifies the significance of negative events, followed by a later phase in which automatic orienting to the unexpected fades, all while the capacity for deliberate attentional effort is preserved and all of it unwinding within days of refeeding. The brain, it seems, does not merely degrade under metabolic stress; it adapts in stages, trading one configuration of emotional processing for another as the energy crisis deepens, and then quietly reverting when the crisis passes. Disentangling the mechanisms behind that choreography, from ketone metabolism to stress hormones to sleep disruption, is the task the authors now hand to the field.

Subject of Research: Neural and autonomic adaptation of emotional reactivity during prolonged complete fasting, measured longitudinally with event-related potentials

Article Title: Phase-dependent adaptation of emotional reactivity during prolonged complete fasting: longitudinal ERP evidence from a human metabolic challenge

Article References: Wu, Q., Fan, Z., Lin, D., Wang, S., Sun, L., Liang, X., Wen, X., Wu, B., Lin, W., Wang, T., Yuan, J., & Fu, W. (2026). Phase-dependent adaptation of emotional reactivity during prolonged complete fasting: longitudinal ERP evidence from a human metabolic challenge. Journal of Translational Medicine. https://doi.org/10.1186/s12967-026-09011-2

Image Credits: AI Generated

DOI: 10.1186/s12967-026-09011-2

Keywords: complete fasting, emotional reactivity, event-related potentials, P3, N2, N1, respiratory sinus arrhythmia, EEG, metabolic stress, neurobehavioral monitoring, longitudinal study, Journal of Translational Medicine