BOSTON — Beneath the vaulted ceilings of the Sheraton Hotel, where clinicians, researchers, and mental health advocates from around the world gathered for the 2025 Boston International Trauma Conference, organized by the world-renowned trauma expert Bessel van der Kolk, a seismic shift unfolded. It was not loud. It did not come with the familiar buzz of celebrity scientists or viral TED-style announcements. Instead, it came quietly, clinically—even maternally.
Standing before a packed room and thousands more tuning in from across the globe, a prominent Dr. Karlen Lyons-Ruth is a Professor of Psychology at Harvard Medical School opened her keynote with a gesture of gratitude to Dr. Bessel van der Kolk—the host of the conference and longtime advocate for trauma science—whom she first met in a dimly lit Cambridge room 36 years ago. “We came to share our emerging thoughts about how important trauma was,” she recalled. “And how could we get the word out?”
Now, that word has spread. And it’s transforming how we think about the origins of human suffering.
The speaker’s scientifically meticulous yet emotionally resonant presentation was built around a provocative question: Is neglect the first form of threat? Her answer, backed by a landmark neurodevelopmental study, may redefine early intervention policy, reshape attachment theory, and challenge decades of assumptions about what truly harms the developing brain.
A Village of Scientists, a Generation at Risk
The research presented came from the MIND Study—a large-scale, longitudinal investigation into how a mother’s childhood maltreatment might alter the stress response and brain development of her infant in the first year of life.
The speaker collaborated with some of the most respected names in the field, including Ellen Grant of Boston Children’s Hospital and Marty Teicher of McLean Hospital. The team employed cutting-edge MRI and salivary cortisol testing in over 180 mother-infant dyads, examining how adverse experiences reverberate across generations, not through overt violence alone, but through the subtle and often invisible forms of disconnection.
“This is the first study we know of,” she explained, “that includes both cortisol levels and amygdala volume in the same infant sample.”
And the findings were striking.
The Weather Forecast Model of Trauma
Drawing on evolutionary biology, the study introduced what researchers dubbed the weather forecast model of intergenerational adaptation. A mother’s early experience of maltreatment—particularly neglect—may act as a forecast signal to her developing infant. That signal, biologically embedded in maternal stress physiology, “prepares” the infant’s brain and hormonal systems for an unpredictable, threatening world.
This isn’t merely a metaphor. The researchers found that maternal childhood neglect—not abuse—was positively associated with elevated infant cortisol levels, a biomarker of stress. Elevated cortisol was also associated with increased amygdala volume, a neurological hallmark of heightened fear vigilance.
Importantly, the mother’s history of abuse, while traumatic, was not related to increased cortisol in the infant. Infants of mothers who had experienced high levels of childhood abuse showed lower cortisol responses to stress at four months of age.
“It was the opposite of what we expected,” the speaker admitted, noting that abuse in human studies of older children tends to shrink the right amygdala. “But we’re seeing something quite different in the earliest phases of life.”
The Neuroscience of Disconnection
Another key focus of the research was whether neglect leaves an imprint on the infant’s stress response system and how that imprint is delivered postnatally.
The team used a modified “still-face” paradigm, a mild stressor, during which maternal behaviors were carefully coded using a five-domain system. These included harmful intrusion (mocking, rough handling), role confusion, and physical withdrawal. But the most predictive dimension—by far—was one often overlooked: disorientation.
Disoriented mothers, described as emotionally detached, affectively odd, or frenetic in their interaction with the infant, raised red flags. Their babies showed significantly higher cortisol output and—again through indirect pathways—larger amygdala volumes.
“When you see ghostly whispers, robotic vocal rhythms, or wooden affect, it doesn’t jump out at you like abuse might,” she explained. “But the baby picks it up. Their stress system responds as if they are under threat.”
The conclusion: lack of affective connection is not a passive absence. It is an active stressor.
The Two-Hit Model: Forecast and Weather
Perhaps the most elegant finding was statistical and deeply human. When both prenatal risk (maternal childhood neglect) and postnatal risk (maternal disorientation) were present, the infant’s cortisol response soared. In statistical terms, this interaction formed a “two-hit model,” suggesting that it takes both a forecast of adversity and real-world confirmation for the infant’s stress response system to be recalibrated for a world of instability.
The infant’s system remains more stable if only one factor is present. “Forecasts can be wrong,” the speaker said. “So biology builds in a kind of insurance.”
This adaptive mechanism echoes findings from other labs using different methods, strengthening the reliability of the results. More chillingly, the pattern reversed in cases of maternal abuse paired with negative, intrusive parenting. Those infants showed suppressed cortisol responses, hinting at early disengagement from overwhelming environments, or what the researcher called “pre-adaptations” to survive a hostile world.
The Silent Tragedy of Withdrawal
As the keynote transitioned to long-term outcomes, the audience was drawn into a more intimate realm: video clips of mother-infant separations coded decades ago. In one, a toddler rushed toward her mother after a brief absence. The mother stepped back. The infant stopped in her tracks, crouched on the floor, and tried to comfort herself on the rug.
What looks, to the untrained eye, like “mild” parenting failure is, in fact, predictive of lifelong patterns. The child had already internalized the need to not burden her caregiver.
Statistical models showed that maternal withdrawal in infancy predicted borderline features, suicidality, dissociation, and even parentification behaviors, in which the child assumes the emotional labor of caring for the parent. These patterns held across multiple large datasets, including the NIH’s SECCYD study of 1,000 families.
Hostile or intrusive parenting, while attention-grabbing, was less predictive of these outcomes.
Redefining Threat in Trauma Science
In concluding her keynote, the speaker proposed revising trauma’s classical model.
Traditionally, trauma is defined as a threat of attack or injury, activating fight, flight, or freeze responses. But the findings at the conference suggest an equally potent early-life danger: the threat of abandonment.
For an entirely dependent infant, neglect is more life-threatening than a predator. The infant’s biological response is to seek contact, cry louder, cling harder, and—if that fails—dissociate or assume the role of caregiver.
“Our models of threat must be developmentally sequenced,” she urged. “Infants fear abandonment. Later, they fear an attack. And our systems need to reflect that.”
A Call to Action
The speaker turned to policy and ethics in her final moments at the podium.
“We have the data. We have the treatments. We have ten or more randomized controlled trials showing we can intervene early. But we don’t use them.”
Why not? Because, she argued, “they don’t make money for the health care system.”
Her voice was calm, but her words seared the room with urgency: “We are watching suffering infants and families fall through the cracks—when we already know how to help.”
And yet, she closed with hope. “The majority of the world’s parents form secure attachments. We know how to do this. We have to transmit it.”
