Neurogenesis in Adults: Current Evidence

For most of the twentieth century, neuroscience operated under a fixed assumption: the adult human brain does not make new neurons. The doctrine, attributed largely to Santiago Ramón y Cajal, held that neuronal populations were established early in life and remained static thereafter. Any loss was permanent.

That view began to fracture in the 1960s with evidence from rodent studies, then more decisively in the 1990s when researchers identified dividing cells in the human hippocampus that expressed markers consistent with newborn neurons. The term for this process is adult neurogenesis—the birth of functional neurons in the mature brain.

The phenomenon is now well-documented in other mammals. In humans, the evidence remains more contested. Some postmortem studies report thousands of new neurons generated daily in the dentate gyrus of the hippocampus well into older age (Boldrini et al., 2018). Others, using different tissue-preparation methods, report that neurogenesis drops precipitously after childhood and is undetectable in adults (Sorrells et al., 2018). The contradiction is not trivial. It reflects methodological challenges inherent in studying fragile, recently born cells in postmortem tissue, as well as differences in how those cells are identified and counted.

What is no longer in dispute is that the adult hippocampus retains a population of neural progenitor cells. Whether those cells reliably differentiate into mature, synaptically integrated neurons in humans—and under what conditions—is the live question.

The question of whether the adult human brain generates new neurons is not academic. It bears directly on how we understand recovery, adaptation, and the revisability of neural systems across the lifespan.

If the hippocampus—a structure central to memory encoding, spatial navigation, and emotional regulation—can incorporate new cells, then its circuitry is not fixed. That opens the door to interventions that might enhance neurogenesis in the service of learning, mood stabilization, or cognitive resilience after injury. Conversely, if neurogenesis is absent or negligible in human adults, then therapeutic models built on the assumption of cellular renewal may need recalibration.

The stakes are especially high in psychiatry. A longstanding hypothesis holds that major depressive disorder may involve impaired hippocampal neurogenesis, and that antidepressant medications—particularly selective serotonin reuptake inhibitors—work in part by restoring it (Santarelli et al., 2003). This theory, derived largely from rodent models, has shaped drug development and clinical reasoning for two decades. If adult human neurogenesis is rare or absent, the explanatory framework weakens.

The controversy also matters because it illustrates a broader tension in translational neuroscience: animal models reliably demonstrate a phenomenon, but human confirmation remains elusive due to technical and ethical constraints. We cannot biopsy living hippocampal tissue. We rely on postmortem samples, often collected hours after death, processed under varying conditions, and analyzed with techniques that may or may not preserve the fragile markers of newborn neurons.

For clinicians, the uncertainty complicates patient communication. Patients ask whether exercise, sleep, or stress reduction can "grow new brain cells." The honest answer is that we have strong evidence these factors promote neurogenesis in rodents, provisional evidence in non-human primates, and indirect evidence in humans—largely from neuroimaging studies showing hippocampal volume changes. That is not the same as direct proof of new neuron formation, but neither is it grounds for dismissing the possibility.

The modern era of adult neurogenesis research began with the work of Joseph Altman in the 1960s, who used tritiated thymidine to label dividing cells in the rodent brain and identified new neurons in the hippocampus and olfactory bulb. The findings were largely ignored. It was not until the 1990s, when Elizabeth Gould and others replicated the phenomenon in primates, that the field gained traction (Gould et al., 1999).

In humans, the first direct evidence came from a study of cancer patients who had received the thymidine analog BrdU as part of diagnostic imaging. Postmortem analysis of their hippocampi revealed BrdU-labeled cells expressing neuronal markers, indicating that new neurons had been born during the patients' final months of life (Eriksson et al., 1998). The study was small but paradigm-shifting.

Since then, the field has bifurcated. Some groups, using immunohistochemical staining for markers such as doublecortin (DCX)—a protein transiently expressed in immature neurons—report robust neurogenesis continuing into the eighth decade of life. Boldrini and colleagues (2018), examining hippocampal tissue from 28 individuals aged 14 to 79, found comparable numbers of DCX-positive cells across age groups and estimated that the human dentate gyrus generates approximately 700 new neurons per day per hippocampus, even in older adults.

Other groups have reached opposite conclusions. Sorrells et al. (2018) analyzed tissue from 59 individuals and reported a sharp decline in DCX-positive cells after infancy, with no detectable neurogenesis in adults. They attributed prior positive findings to methodological artifacts, including antibody cross-reactivity and tissue degradation.

A 2019 study attempted reconciliation. Moreno-Jiménez et al. examined "fresh" postmortem tissue—processed within hours of death—and reported thousands of immature neurons in the dentate gyrus of older adults, with numbers declining in patients with Alzheimer disease. The authors argued that rapid tissue processing is essential to preserve DCX immunoreactivity, and that delayed fixation leads to false negatives (Moreno-Jiménez et al., 2019).

Methodological differences thus account for much of the controversy. Tissue quality, postmortem interval, fixation protocols, antibody selection, and cell-counting criteria all vary across studies. A 2021 review in Nature Medicine noted that the field lacks standardized methods and called for multi-site validation studies using uniform protocols (Kempermann et al., 2018).

Beyond the hippocampus, there is no credible evidence for widespread neurogenesis in the adult human cortex or striatum under normal conditions. Some studies have suggested limited neurogenesis in the striatum following stroke (Arvidsson et al., 2002), but these findings have not been consistently replicated in humans.

Functional integration is another open question. Even if new neurons are born, do they form synapses and contribute to network activity? In rodents, optogenetic silencing of adult-born neurons impairs pattern separation—the ability to distinguish similar memories—suggesting functional relevance (Sahay et al., 2011). In humans, indirect evidence comes from MRI studies showing that aerobic exercise increases hippocampal volume and improves memory performance, changes that correlate with increased serum levels of brain-derived neurotrophic factor (BDNF), a molecule known to promote neurogenesis in animal models (Erickson et al., 2011).

A 2022 meta-analysis in JAMA Psychiatry reviewed 18 studies examining hippocampal volume changes in response to antidepressant treatment. The analysis found modest but significant volume increases in patients who responded to treatment, though the authors cautioned that volume change does not confirm neurogenesis and could reflect synaptogenesis, angiogenesis, or glial proliferation (Kraus et al., 2022).

The most rigorous human evidence to date comes from carbon-14 dating of hippocampal neurons, a technique that exploits the spike in atmospheric carbon-14 from Cold War nuclear testing. By measuring carbon-14 incorporation into neuronal DNA, researchers estimated that approximately one-third of hippocampal neurons in the dentate gyrus are subject to exchange over a human lifetime, consistent with ongoing neurogenesis (Spalding et al., 2013). This method bypasses the pitfalls of immunohistochemistry but cannot resolve the rate or functional significance of neuron turnover.

In sum, the weight of evidence supports the existence of neural progenitor cells in the adult human hippocampus and suggests that neurogenesis occurs, at least under some conditions. The rate, persistence across the lifespan, and functional contribution remain unresolved.

Within the Nervous System Intelligence framework, neurogenesis is not merely a curiosity of cellular biology. It is a structural substrate for prediction revision—a mechanism by which the brain updates its internal models in response to new information.

The hippocampus is a prediction engine. It encodes spatial and episodic information, compares incoming sensory data against stored representations, and signals prediction error when the two diverge. If the hippocampus were static, its predictions would calcify. The addition of new neurons—each with a distinct integration window and heightened plasticity—allows the network to remain revisable.

This is most evident in pattern separation, the process by which the brain distinguishes between similar but non-identical experiences. Adult-born neurons in the dentate gyrus are thought to enhance this capacity by providing a pool of highly excitable cells that respond preferentially to novel or ambiguous inputs (Aimone et al., 2011). In computational terms, they increase the dimensionality of hippocampal representations, reducing interference between overlapping memories.

From an NSI perspective, this is a form of structural validation. The nervous system does not merely adjust synaptic weights in response to error; it can, under certain conditions, recruit new computational units. This is not limitless plasticity—neurogenesis is tightly regulated and declines with age—but it is a reminder that the system retains generative capacity.

The NIRVA Method's Validate movement is implicated here. Validation, in the NSI sense, is the process by which the nervous system confirms or disconfirms its predictions through embodied experience. Neurogenesis may be one mechanism by which the system prepares itself for validation—by expanding its representational repertoire in anticipation of novelty.

Exercise, environmental enrichment, and learning all promote neurogenesis in animal models. These are not passive interventions. They are active engagements with the environment that generate prediction error and demand model revision. The nervous system responds by building new infrastructure.

Conversely, chronic stress and glucocorticoid exposure suppress neurogenesis. Stress, in NSI terms, is often a state of sustained prediction error without resolution—the system is repeatedly surprised but cannot revise its models effectively. The suppression of neurogenesis under these conditions may reflect a defensive posture: the system withdraws from plasticity to preserve existing structure.

This is speculative but consistent with the broader NSI thesis: the nervous system is not a passive receiver of inputs but an active modeler of the world, and its structural plasticity is governed by the same prediction-error dynamics that govern synaptic plasticity. Neurogenesis, where it occurs, is one tool in the system's revisability toolkit.

For clinicians, the neurogenesis literature offers provisional guidance rather than definitive protocol. The evidence is strong enough to inform lifestyle counseling but not strong enough to anchor pharmacological claims.

Patients frequently ask whether they can "regrow" neurons lost to depression, stress, or aging. The honest answer is that we do not know whether clinically significant neurogenesis occurs in adult humans, but we do know that behaviors associated with neurogenesis in animal models—aerobic exercise, adequate sleep, cognitive challenge, and stress reduction—are independently beneficial for mood, cognition, and metabolic health. The recommendation stands regardless of whether new neurons are born.

In psychiatry, the neurogenesis hypothesis of depression has been influential but remains unproven in humans. SSRIs increase hippocampal neurogenesis in rodents, and this effect is necessary for some behavioral responses to antidepressants in animal models (Santarelli et al., 2003). But we lack direct evidence that SSRIs promote neurogenesis in human patients, and we lack evidence that neurogenesis is required for antidepressant efficacy in humans. Hippocampal volume increases observed on MRI after SSRI treatment could reflect other forms of plasticity.

Clinicians should avoid overstating the neurogenesis mechanism when discussing antidepressants with patients. It is more accurate to say that antidepressants likely promote various forms of neural plasticity, of which neurogenesis may be one component in some individuals.

In neurology, the question arises in the context of recovery after stroke or traumatic brain injury. There is no evidence that the adult human brain generates new cortical neurons to replace those lost to injury. Rehabilitation likely works through synaptic reorganization, unmasking of latent pathways, and recruitment of perilesional tissue—not through neurogenesis. Overpromising regeneration can undermine realistic goal-setting.

Where neurogenesis may have clinical relevance is in the design of cognitive interventions for older adults. If the hippocampus retains neurogenic capacity into late life, then interventions that combine aerobic exercise, novelty, and cognitive challenge may support hippocampal function more effectively than passive cognitive training. A 2023 randomized trial in Neurology found that a 12-week multimodal intervention combining aerobic exercise, cognitive training, and social engagement improved episodic memory in older adults and was associated with increased hippocampal volume on MRI, though the study did not measure neurogenesis directly (Ten Brinke et al., 2023).

Clinicians should also be alert to factors that suppress neurogenesis in animal models: chronic sleep deprivation, high-dose glucocorticoids, heavy alcohol use, and social isolation. While we cannot confirm that these factors suppress neurogenesis in humans, they are modifiable risk factors for cognitive decline and mood disorders, and addressing them is sound clinical practice.

You cannot measure your own neurogenesis. You cannot feel new neurons being born. But you can structure your behavior in ways that, in other mammals, reliably promote it.

Start with movement. Aerobic exercise is the most robust pro-neurogenic intervention identified in animal studies. In humans, a moderate-intensity routine—30 minutes of brisk walking, cycling, or swimming most days of the week—is associated with preserved hippocampal volume and improved memory performance in older adults. The mechanism may or may not involve neurogenesis, but the outcome is consistent.

Sleep is non-negotiable. Sleep deprivation suppresses neurogenesis in rodents, and chronic short sleep is associated with hippocampal atrophy in humans. Aim for seven to nine hours of sleep per night, with attention to sleep quality, not just duration. If you wake unrefreshed despite adequate time in bed, investigate further.

Seek novelty, but not chaos. Environmental enrichment—exposure to new environments, learning new skills, social interaction—promotes neurogenesis in animal models. In humans, this translates to activities that combine physical, cognitive, and social challenge. Learning a language, taking up a musical instrument, or engaging in complex social games may be more effective than passive consumption of information.

Manage stress, but do not avoid it entirely. Acute stress can enhance learning and memory. Chronic, uncontrollable stress suppresses neurogenesis. The distinction matters. A challenging project with a clear endpoint is different from ongoing uncertainty without resolution. Where possible, convert chronic stressors into bounded challenges.

Nutrition likely plays a role, though the evidence is less direct. Omega-3 fatty acids, flavonoids, and caloric restriction have all been shown to promote neurogenesis in animal models. In humans, adherence to a Mediterranean-style diet is associated with larger hippocampal volumes and lower dementia risk, though causality is not established.

Finally, consider what you are asking of your hippocampus. If your days are repetitive, your environment unchanging, and your cognitive demands minimal, you are not creating the conditions under which neurogenesis—if it occurs—would be advantageous. The system builds what it needs. Give it a reason to build.