We do not always fully understand our emotions, yet everyday life would be impossible without them. Emotions guide decisions and actions, helping people navigate the world. But when emotional responses are poorly matched to a situation—or persist for too long—they can become harmful.
Researchers still do not have a complete understanding of the brain activity that generates emotional states, how those states shape behavior, or how emotional processing breaks down in mental illness. A Stanford Medicine team has now identified a brainwide pattern of neural activity that follows a mildly unpleasant sensory event, revealing a timing-based mechanism that appears to be shared by both humans and mice.
The findings, published in Science, suggest that emotions may depend on a brief period of sustained communication between distant brain regions. According to the researchers, this neural “persistence” could represent a fundamental feature of healthy emotional processing—and a possible point of failure in psychiatric disorders involving emotional symptoms.
The study was led by Karl Deisseroth, professor of bioengineering and of psychiatry and behavioral sciences, together with senior co-authors Carolyn Rodriguez, Vivek Buch, and Paul Nuyujukian. Lead authors included Isaac Kauvar, Ethan Richman, and Tony Liu. The work was conducted within Stanford Medicine’s Human Neural Circuitry program, which combines clinical hospital settings with highly precise measurements of brain activity and behavior.
Why A Large Brain Needs Time
Across mammals, brain size expanded dramatically throughout evolution. A mouse brain contains roughly 100 million neurons, while the human brain contains about 90 billion. This expansion enables richer and more flexible mental experiences, but it also creates a challenge: signals must travel much farther, and integrating information across the brain takes time.
The researchers propose that emotions may help solve this integration problem. Emotional states can compress many different inputs—sensory information, goals, context, and bodily needs—into a unified mode that guides behavior over a meaningful period of time. To achieve this, the brain may require short-lived stability in communication patterns spread across multiple regions.
One of the study’s co-authors compared this process to a piano sustain pedal: a note is played briefly, but the sound lingers long enough to shape the musical phrase. Similarly, a short sensory event may trigger an extended neural “tail” that helps the brain integrate information and generate an emotional response. If this persistence becomes too short or too prolonged, it may contribute to psychiatric symptoms.
A Safe Trigger: The “Eye Puff”
To study emotion in a controlled and repeatable way, the researchers needed a stimulus that was safe, precisely timed, and usable in both humans and mice. They chose a familiar tool from ophthalmology: a brief puff of air directed at the eye, similar to the one used during routine eye-pressure tests.
The air puff is not painful, but most people find it mildly unpleasant. Participants described the sensation as “annoying,” “unpleasant,” or “uncomfortable.” When the puffs were delivered repeatedly in rapid succession, the unpleasant feeling tended to build and persist even after the stimulus ended—an important hallmark of emotional processing rather than a simple reflex.
Rare Human Recordings Reveal A Two-Phase Brain Signature
High-resolution recordings across the entire human brain are rarely possible. In this study, the researchers worked with Stanford Hospital patients undergoing clinical monitoring for severe epilepsy. These patients already had intracranial electrodes implanted for medical reasons so clinicians could identify where seizures began. During their hospital stay, many volunteered to participate in the research.
Behaviorally, responses to the random eye puffs were highly consistent. Each puff triggered an immediate reflexive blink. In the following seconds, participants often continued blinking or squinting more than usual—an additional protective behavior that researchers could measure and connect to the lingering unpleasant emotional state.
At the same time, the implanted electrodes recorded a distinctive two-part neural pattern.
Phase 1
A strong and rapid burst of activity occurred within approximately 200 milliseconds, spreading information about the puff across the brain.
Phase 2
A slower and longer-lasting response unfolded over the following several hundred milliseconds, becoming more concentrated within circuits linked to emotional processing.
The researchers argue that this second phase creates a brief window during which widely separated brain regions remain connected long enough to integrate information and support an emotional state.
The Same Pattern Appears In Mice
To determine whether these mechanisms are conserved across evolution, the researchers repeated the same eye-puff experiment in mice. They observed a remarkably similar two-phase pattern of brain activity.
When mice received a rapid series of puffs, the slower second phase became stronger and longer-lasting, pushing the animals into a generalized negative emotional state. This was reflected in a reduced willingness to pursue rewards even after the puffs stopped.
The researchers note that persistence beyond the original stimulus and its influence on unrelated behaviors are considered classic signatures of emotion.
Ketamine Selectively Removes The “Emotional Tail”
The team next investigated whether the slower, persistent second phase was truly connected to emotional experience by using ketamine, a drug known to produce dissociation. Ketamine is used as an anesthetic at high doses and is also approved as an antidepressant at lower doses. One of its notable effects is that people may remain aware of sensory events while feeling emotionally detached from them.
Under carefully controlled hospital conditions and with informed consent, participants received a single dose of ketamine. Afterward, repeated eye puffs no longer felt strongly annoying to many participants. Some even described the sensation playfully. Importantly, reflexive blinking remained intact, but the longer-lasting protective behavior—keeping the eyes more closed between puffs—mostly disappeared.
Researchers observed the same selective behavioral effect in mice. Reflexive responses stayed normal, while the prolonged self-protective behavior was reduced.
Neurally, ketamine produced a similarly selective effect. The initial fast, brainwide burst of activity (Phase 1) remained largely unchanged. However, the slower second phase faded much more quickly, shortening the duration of puff-related activity—similar to releasing a piano’s sustain pedal and cutting off the lingering note.
Even without eye puffs, ketamine shortened the brain’s intrinsic timing, meaning activity patterns remained correlated for less time. This suggests a broader acceleration of brain dynamics during dissociation. The effect reversed once the drug wore off.
The team also found that ketamine temporarily reduced synchrony across the brain in both humans and mice. The researchers propose that when this stabilizing and integrative phase becomes too brief, the brain may fail to combine distributed information into a unified emotional experience.
What This Could Mean For Mental Health
The authors suggest that the timing of this integrative brain activity—how quickly it rises and how slowly it fades—may become an important measurable feature in psychiatric disorders.
If the activity fades too quickly, the brain may struggle to coordinate information across regions, potentially contributing to symptoms in which experiences feel disconnected from the self. If the activity becomes too persistent or overly intense, it could contribute to intrusive or long-lasting emotions and thoughts, such as those seen in post-traumatic stress disorder, obsessive-compulsive disorder, depression, or eating disorders.
The researchers also note that persistent brain activity could affect the speed of information processing more broadly. They raise the possibility that overly stable brain states may make it harder to adapt to rapidly changing input, an idea they suggest could relate to some challenges observed in autism spectrum disorder.
Beyond these hypotheses, the study’s central message is methodological: direct brainwide comparisons between humans and mice can reveal conserved neural dynamics that may underlie fundamental aspects of emotion. The team says future work will explore whether similar principles also apply to positive emotional experiences.
The study included contributions from researchers at the Veterans Affairs Palo Alto Health Care System and Weill Cornell Medicine and was supported by several grants from the National Institutes of Health and other funding sources.
