Neuroplasticity and Mental Health: BDNF, Synaptic Pruning, Exercise Effects, and Treatment Implications

For decades, psychiatric disorders were conceptualized primarily through the lens of neurochemical imbalance — too little serotonin in depression, too much dopamine in psychosis. While these frameworks generated effective treatments, they were always incomplete. The neuroplasticity hypothesis offers a more nuanced, mechanistically rich framework: psychiatric illness frequently involves disrupted capacity for adaptive neural remodeling, and effective treatments — whether pharmacological, psychotherapeutic, or behavioral — share a common downstream pathway of restoring or enhancing plasticity.

Neuroplasticity refers to the brain's ability to reorganize its structure and function in response to experience, injury, or pathological processes. This encompasses multiple levels of analysis: synaptic plasticity (changes in the strength and number of synaptic connections), structural plasticity (dendritic remodeling, neurogenesis, gliogenesis), network-level plasticity (reorganization of functional connectivity between brain regions), and epigenetic plasticity (experience-dependent changes in gene expression). Each of these levels is relevant to understanding the pathophysiology and treatment of psychiatric conditions.

The scope of this framework is vast. Impaired neuroplasticity has been implicated in major depressive disorder (MDD), which affects approximately 280 million people worldwide (WHO, 2023), with a lifetime prevalence of 16–20% in high-income countries. It is also central to understanding post-traumatic stress disorder (PTSD; lifetime prevalence ~6.1% in U.S. adults per the National Comorbidity Survey Replication), anxiety disorders (global prevalence ~4%, with substantial undercount), schizophrenia (lifetime prevalence ~0.7–1.0%), and substance use disorders. The neuroplasticity framework does not replace neurotransmitter models but rather subsumes them: serotonin, dopamine, glutamate, and GABA all exert their psychiatric relevance in large part through their effects on plasticity mechanisms.

This article provides a detailed examination of the key molecular and cellular mediators of neuroplasticity — with particular focus on brain-derived neurotrophic factor (BDNF) signaling, synaptic pruning and its dysregulation, exercise-induced neuroplastic changes, and the implications of this science for treatment selection and development. The goal is to bridge the gap between basic neuroscience and clinical practice, offering clinicians a mechanistically grounded understanding of why certain interventions work, for whom, and through what pathways.

Brain-derived neurotrophic factor (BDNF) is a member of the neurotrophin family of growth factors and is the most extensively studied molecular mediator of neuroplasticity in psychiatry. BDNF is synthesized as a precursor protein (proBDNF) and cleaved into its mature form (mBDNF), each of which has distinct — and often opposing — biological effects. Mature BDNF binds to the tropomyosin receptor kinase B (TrkB) receptor, activating intracellular signaling cascades including the Ras-MAPK/ERK pathway, the PI3K-Akt pathway, and the PLCγ-CaMKII pathway. These cascades collectively promote synaptic potentiation, dendritic growth, spine formation, and cell survival. In contrast, proBDNF preferentially binds the p75 neurotrophin receptor (p75NTR), which promotes long-term depression (LTD), synaptic retraction, and — in some contexts — apoptosis.

The balance between proBDNF/p75NTR signaling and mBDNF/TrkB signaling is therefore a critical determinant of whether neural circuits undergo strengthening or weakening. This balance is disrupted in multiple psychiatric conditions:

• Major Depressive Disorder: A landmark meta-analysis by Molendijk et al. (2014), pooling data from 55 studies (n > 9,000), found that serum BDNF levels are significantly reduced in patients with MDD compared to healthy controls, with a moderate effect size (Cohen's d ≈ 0.71). Critically, BDNF levels normalize with successful antidepressant treatment, suggesting BDNF as both a state marker and a mediator of treatment response. Postmortem studies have confirmed reduced BDNF mRNA and protein levels in the hippocampus and prefrontal cortex of depressed individuals who died by suicide.
• Schizophrenia: Reduced serum BDNF has been documented across illness stages, including first-episode psychosis, with meta-analytic evidence showing a small-to-moderate effect (d ≈ 0.38). BDNF reductions correlate with cognitive impairment, which is the strongest predictor of functional outcome in schizophrenia.
• PTSD: A meta-analysis by Angelucci et al. (2014) found decreased peripheral BDNF in PTSD, with the extent of reduction correlating with symptom severity. Animal models of stress-enhanced fear conditioning show that BDNF infusion into the infralimbic cortex facilitates fear extinction — a finding with direct relevance to exposure-based therapies.

The Val66Met polymorphism (rs6265) of the BDNF gene is one of the most studied genetic variants in psychiatric neuroscience. The Met allele, carried by approximately 20–30% of Caucasians and up to 50% of East Asian populations, impairs activity-dependent secretion of BDNF from neurons. Met allele carriers show reduced hippocampal volume (~4% reduction in some imaging studies), impaired episodic memory performance, and — in some studies — altered susceptibility to stress-related psychopathology. The landmark study by Egan et al. (2003) at the NIMH first characterized the functional impact of this variant, demonstrating impaired intracellular trafficking and activity-dependent secretion of BDNF in Met-carrying neurons.

However, the clinical impact of Val66Met is complex and context-dependent. Gene-by-environment interaction studies suggest the Met allele confers vulnerability specifically when combined with early life stress, consistent with a diathesis-stress model. A meta-analysis by Hosang et al. (2014) found that the Val66Met polymorphism moderates the relationship between stressful life events and depression, but the overall effect size is small, and many candidate gene studies in psychiatry have not replicated robustly. Clinicians should be cautious about over-interpreting single-gene findings in what are fundamentally polygenic disorders.

Synaptic pruning is the experience-dependent elimination of weaker or redundant synaptic connections, a process essential for refining neural circuits during development. The human brain reaches peak synaptic density at approximately age 1–2 years in the visual cortex and somewhat later in the prefrontal cortex (PFC), after which approximately 40–50% of synapses are eliminated by adulthood. This pruning is not random; it follows Hebbian principles — connections that are repeatedly co-activated are strengthened, while those that are inactive are tagged for elimination. The result is a more efficient, specialized neural architecture.

The molecular machinery of synaptic pruning involves several key players: complement proteins C1q, C3, and C4, which opsonize (tag) weak synapses for phagocytic removal by microglia; classical and non-classical MHC class I molecules, which regulate synapse elimination through immune-like signaling; and astrocytic signals, including the MEGF10 and MERTK phagocytic pathways. Microglia, the brain's resident immune cells, physically engulf tagged synapses in a process that depends on fractalkine (CX3CL1-CX3CR1) signaling and is regulated by neuronal activity.

Disrupted synaptic pruning has been implicated in at least two major categories of psychiatric illness:

Excessive Pruning: Schizophrenia

The synaptic pruning hypothesis of schizophrenia, first proposed by Feinberg (1982), posits that the illness arises in part from excessive elimination of cortical synapses during adolescence — precisely the period when schizophrenia typically emerges (median age of onset: 18–25 years for men, 25–35 years for women). Postmortem studies consistently show reduced dendritic spine density in the dorsolateral prefrontal cortex (DLPFC) of patients with schizophrenia, with reductions of ~15–25% compared to age-matched controls (Glantz & Lewis, 2000).

The most significant genetic evidence linking pruning to schizophrenia came from the landmark study by Sekar et al. (2016), published in Nature. This study identified structurally distinct alleles of the complement component 4 (C4) gene — located within the major histocompatibility complex (MHC) region on chromosome 6, the strongest genetic association signal in schizophrenia GWAS — and demonstrated that schizophrenia-associated C4 alleles are associated with greater C4A expression in the brain. In a mouse model, C4 mediated synaptic elimination during postnatal development. This study provided the first mechanistic link between the strongest genetic risk locus for schizophrenia and a specific neurobiological process: complement-mediated synaptic pruning.

Insufficient Pruning: Autism Spectrum Disorder

In contrast, autism spectrum disorder (ASD; prevalence now estimated at 1 in 36 children per CDC 2023 estimates) has been associated with excessive synaptic density, suggesting insufficient pruning. Postmortem analyses by Tang et al. (2014) found ~18% more dendritic spines in the temporal cortex of children with ASD and ~50% more in adolescents compared to controls, with the difference widening with age — consistent with a pruning deficit rather than overproduction. This excess connectivity may underlie the sensory hypersensitivity and reduced signal-to-noise ratio in cortical processing characteristic of ASD. The mTOR signaling pathway, which inhibits autophagy-mediated pruning, has been implicated, and mTOR inhibitors (e.g., rapamycin) have shown promise in animal models of tuberous sclerosis complex with ASD features.

Stress, Microglia, and Aberrant Pruning in Depression

Emerging evidence suggests that chronic stress dysregulates microglial pruning in a manner relevant to depression. Chronic stress in rodent models activates microglia in the medial PFC and hippocampus, leading to excessive synaptic stripping — loss of dendritic spines and reduced synaptic density in regions critical for top-down emotion regulation. PET imaging studies in humans using TSPO (translocator protein) ligands — a marker of microglial activation — have shown increased neuroinflammation in the PFC and anterior cingulate cortex of patients with MDD, though findings are inconsistent and TSPO imaging has significant methodological limitations.

Physical exercise is the most robustly supported non-pharmacological intervention for enhancing neuroplasticity across the lifespan. Its effects are mediated through multiple converging mechanisms:

• BDNF upregulation: Aerobic exercise reliably increases peripheral BDNF levels, with meta-analytic evidence showing a small-to-moderate acute effect (d ≈ 0.46) following a single exercise bout and cumulative effects with regular training (Szuhany et al., 2015). The source of exercise-induced BDNF is primarily the brain (~70–80% of circulating BDNF during exercise originates centrally), with the hippocampus being a major contributor. BDNF crosses the blood-brain barrier, making peripheral levels a reasonable — if imperfect — proxy for central changes.
• Hippocampal neurogenesis: Exercise robustly stimulates adult neurogenesis in the dentate gyrus of the hippocampus in animal models, an effect mediated through BDNF, VEGF (vascular endothelial growth factor), and IGF-1 (insulin-like growth factor 1). The landmark randomized controlled trial by Erickson et al. (2011) demonstrated that 12 months of moderate-intensity aerobic exercise (walking 40 minutes, 3 days/week) produced a ~2% increase in hippocampal volume in older adults (n = 120), effectively reversing 1–2 years of age-related atrophy. The control group (stretching) showed ~1.4% volume decline over the same period. Hippocampal volume increases correlated with improved spatial memory performance and elevated serum BDNF.
• Anti-inflammatory effects: Exercise reduces systemic levels of pro-inflammatory cytokines (IL-6, TNF-α, CRP) and increases anti-inflammatory mediators (IL-10), counteracting the neuroinflammation implicated in depression and neurodegeneration. This is particularly relevant given that approximately 25–30% of MDD patients show elevated inflammatory markers.
• Glutamatergic and monoaminergic modulation: Exercise increases synaptic availability of serotonin, norepinephrine, and dopamine, and modulates glutamatergic signaling in cortical and limbic circuits.
• Lactate as a signaling molecule: Exercise-produced lactate crosses the blood-brain barrier and induces BDNF expression through the SIRT1-PGC1α-FNDC5 pathway. The myokine irisin (cleaved from FNDC5) has emerged as a key mediator of exercise-brain crosstalk.

Clinical Evidence for Exercise in Psychiatric Disorders

The clinical evidence for exercise as a psychiatric intervention has matured considerably:

Depression: The most comprehensive meta-analysis to date (Singh et al., 2023, British Journal of Sports Medicine) analyzed 218 RCTs (n = 14,170) and found that exercise significantly reduced depressive symptoms across all types compared to active controls, with a moderate-to-large effect size (SMD = −0.43 to −0.67) depending on exercise modality. Vigorous-intensity exercise showed the largest effects. The number needed to treat (NNT) for clinically significant improvement is estimated at approximately 4–5 for moderate-intensity aerobic exercise, comparable to pharmacotherapy in mild-to-moderate depression. The landmark SMILE trial (Blumenthal et al., 1999; 2007 follow-up) demonstrated that supervised aerobic exercise (3 sessions/week, 30 minutes at 70–85% heart rate reserve) was as effective as sertraline in reducing depressive symptoms in adults with MDD over 16 weeks, with remission rates of ~46% for exercise versus ~47% for sertraline. At 10-month follow-up, the exercise group had significantly lower relapse rates.

Anxiety disorders: Meta-analytic evidence indicates a moderate effect of exercise on anxiety symptoms (SMD ≈ −0.33 to −0.48), with both aerobic and resistance training showing benefits. Effects are most consistent for generalized anxiety and panic disorder.

Schizophrenia: A Cochrane review and subsequent meta-analyses show that aerobic exercise improves global cognition (d ≈ 0.33), working memory, and total symptom severity in schizophrenia, with 12 or more weeks of moderate-to-vigorous exercise showing the most robust effects. Given that no pharmacological intervention has demonstrated consistent efficacy for cognitive deficits in schizophrenia — the largest driver of functional disability — exercise represents one of the only evidence-based interventions for this domain.

Dose-Response Considerations

Evidence converges on a minimum effective dose of approximately 150 minutes per week of moderate-intensity aerobic exercise, consistent with WHO physical activity guidelines. However, dose-response analyses suggest that even subthreshold levels confer some benefit, and higher doses (beyond 150 minutes) show additional gains with diminishing returns. Resistance training appears to have independent effects on mood (SMD ≈ −0.66 for depression in a meta-analysis by Gordon et al., 2018), likely through partially distinct mechanisms including IGF-1 signaling and hypothalamic-pituitary-adrenal (HPA) axis modulation.

Reconceptualizing psychiatric pharmacotherapy through the lens of neuroplasticity provides a more coherent explanation of treatment latency, individual variability, and differential effectiveness than traditional monoamine models alone.

SSRIs and Serotonergic Antidepressants

The 2–4 week latency of SSRI antidepressant response has long been an anomaly under the serotonin hypothesis, since serotonin reuptake inhibition occurs within hours. The neuroplasticity framework resolves this: SSRIs increase BDNF expression and activate TrkB signaling, promote hippocampal neurogenesis, and enhance synaptic plasticity in prefrontal-limbic circuits — processes that require days to weeks. Castrén's neuroplasticity theory of antidepressant action (2005, updated 2013) proposes that SSRIs reopen "critical period"-like plasticity in the adult brain, allowing new learning (including the corrective emotional learning facilitated by psychotherapy) to reshape maladaptive neural patterns. A groundbreaking 2023 study by Bhatt et al. in Science and by Casarotto et al. in Cell (2021) demonstrated that fluoxetine and all tested antidepressants — including SSRIs, SNRIs, tricyclics, and ketamine — bind directly to TrkB receptors, promoting BDNF signaling. This provides a unifying mechanism independent of monoamine effects.

Response rates for SSRIs in MDD are approximately 50–60%, with remission rates of 30–35% in first-line treatment. The STAR*D trial (Sequenced Treatment Alternatives to Relieve Depression), the largest clinical trial in depression, demonstrated cumulative remission rates of approximately 67% after four treatment steps, but with significantly poorer outcomes at each subsequent step (37% at step 1, 31% at step 2, 14% at step 3, 13% at step 4), underscoring the importance of treatment selection and the diminishing returns of monoaminergic augmentation strategies alone.

Ketamine and Rapid-Acting Antidepressants

Ketamine, an NMDA receptor antagonist, produces antidepressant effects within hours to days — a timeline that maps onto rapid synaptic plasticity mechanisms rather than the slower neurogenesis pathway. The leading mechanistic model proposes that ketamine's blockade of NMDA receptors on GABAergic interneurons produces a transient disinhibition of glutamatergic pyramidal neurons, leading to a surge in glutamate release that activates AMPA receptors. AMPA receptor activation triggers BDNF release and mTOR-dependent signaling, resulting in rapid synaptogenesis — a burst of new synaptic connections in the PFC within 2–24 hours in rodent models (Duman & Aghajanian, 2012). This synaptic remodeling restores PFC output to limbic structures, normalizing the top-down regulatory deficits seen in depression.

Clinical data for IV ketamine and intranasal esketamine (Spravato) in treatment-resistant depression (TRD) show response rates of 50–70% and remission rates of 25–35% within 24–72 hours. However, durability is limited without repeated dosing (median time to relapse: ~18 days after a single infusion), raising questions about whether ketamine-induced plasticity is consolidated into lasting circuit changes without adjunctive interventions. The SUSTAIN trials for esketamine demonstrated maintained efficacy with twice-weekly to weekly dosing, but with relapse rates of ~25% over 16 weeks after randomized withdrawal.

Psychedelics: Psilocybin and Neuroplastic Remodeling

Psilocybin, a 5-HT2A receptor agonist, has shown remarkable antidepressant effects in phase II trials, with response rates of 50–70% and remission rates of 25–40% in TRD after just 1–2 dosing sessions (Carhart-Harris et al., 2021; Davis et al., 2021). Preclinical data show that psilocybin and other psychedelics promote rapid dendritic growth, spinogenesis, and enhanced functional connectivity — effects that persist for at least one month after a single dose. Shao et al. (2021) demonstrated in mice that a single dose of psilocybin increased dendritic spine density in the frontal cortex by approximately 10% within 24 hours, with spines remaining elevated at one month. The "REBUS" (relaxed beliefs under psychedelics) model by Carhart-Harris and Friston proposes that psychedelics temporarily flatten the brain's prior expectations (reduce the precision weighting of top-down predictions), enabling a window of heightened plasticity during which new patterns of thought and behavior can crystallize — a mechanism reminiscent of Castrén's critical-period reopening theory for SSRIs.

Lithium and Mood Stabilizers

Lithium, the gold-standard mood stabilizer (NNT for suicide prevention: ~12–15 per Cipriani et al., 2013 meta-analysis), has well-characterized neuroprotective and neuroplastic effects. Lithium inhibits glycogen synthase kinase 3β (GSK-3β), a kinase that promotes apoptosis and inhibits BDNF signaling. Chronic lithium treatment increases hippocampal volume (~2–3%), BDNF expression, Bcl-2 (anti-apoptotic protein) levels, and gray matter volume in the PFC. These neuroprotective effects may explain lithium's unique efficacy in preventing both manic and depressive episodes — and its anti-suicidal properties — through mechanisms that transcend acute mood stabilization.

If pharmacotherapy enhances the brain's capacity for plasticity, psychotherapy provides the structured experience that directs plastic changes toward adaptive outcomes. This framing — articulated by Castrén and others — positions medication and psychotherapy not as competing treatments but as synergistic interventions operating on complementary components of the same neuroplastic process.

Cognitive Behavioral Therapy (CBT)

CBT, the most extensively researched psychotherapy for depression and anxiety, has been associated with measurable neurobiological changes in neuroimaging studies. Pre-post treatment neuroimaging of CBT responders shows reduced amygdala reactivity to negative stimuli, increased prefrontal cortical activation during emotion regulation tasks, and normalized resting-state connectivity between the PFC and amygdala. These changes are consistent with strengthened top-down regulatory circuits — a form of experience-dependent plasticity. Meta-analytic data indicate CBT response rates of approximately 50–60% for MDD, with an NNT of approximately 4–5 for depression remission compared to waitlist controls.

Exposure-Based Therapies and Fear Extinction

Prolonged Exposure (PE) and Cognitive Processing Therapy (CPT) for PTSD rely on fear extinction — the formation of new inhibitory memories that compete with and suppress conditioned fear responses. Fear extinction depends critically on BDNF-TrkB signaling in the infralimbic cortex (homologous to the human ventromedial PFC) and NMDA receptor-dependent plasticity in the basolateral amygdala. D-cycloserine (DCS), a partial NMDA receptor agonist that enhances extinction learning, has been studied as an adjunct to exposure therapy, with a meta-analysis by Mataix-Cols et al. (2017) showing a small but significant augmentation effect (d ≈ 0.25) when administered before exposure sessions, particularly in specific phobia and social anxiety disorder.

EMDR and Memory Reconsolidation

Eye Movement Desensitization and Reprocessing (EMDR) may engage memory reconsolidation — the process by which retrieved memories become transiently labile and modifiable before restabilizing. Reconsolidation requires protein synthesis-dependent plasticity in the amygdala and hippocampus. While the specific role of bilateral stimulation remains debated, EMDR achieves comparable efficacy to PE for PTSD (response rates: ~55–65%), suggesting convergent plasticity-dependent mechanisms.

The critical clinical implication of this framework is that the timing and pairing of plasticity-enhancing interventions with structured therapeutic experiences may be more important than either intervention alone. This is the theoretical rationale behind combining ketamine or psilocybin administration with psychotherapy sessions — creating a window of enhanced plasticity during which therapeutic learning can be maximally consolidated.

Not all patients respond equally to neuroplasticity-promoting interventions. Identifying predictors of neuroplastic responsiveness is a priority for precision psychiatry.

Age

Neuroplastic capacity declines with age, though it is never entirely lost. Older adults show reduced exercise-induced BDNF elevations, slower rates of hippocampal neurogenesis (though the extent of adult neurogenesis in humans remains debated), and reduced cortical thickness gains from cognitive training. However, the Erickson et al. (2011) exercise trial demonstrated meaningful hippocampal volume increases even in adults aged 55–80, suggesting that the aging brain retains clinically relevant plasticity.

Chronic Stress and Allostatic Load

Chronic stress is a potent suppressor of neuroplasticity. Glucocorticoid-mediated dendrite retraction in the hippocampus and PFC, microglial activation, and reduced BDNF expression create a state of "plasticity resistance" that may explain the lower treatment response rates seen in patients with high adverse childhood experience (ACE) scores. Patients with ≥4 ACEs have approximately 2–3 times the risk of treatment-resistant depression compared to those with no ACEs (Williams et al., 2016).

Inflammatory Status

Patients with elevated baseline inflammatory markers (CRP > 3 mg/L, elevated IL-6) show differential treatment response patterns. In the BIODEP study and related work, high-inflammation depressed patients showed poorer response to SSRIs but potentially better response to anti-inflammatory augmentation strategies. Inflammation directly impairs BDNF signaling and promotes neurotoxic kynurenine pathway metabolites (quinolinic acid) at the expense of neuroprotective metabolites (kynurenic acid), creating a mechanistic link between immune dysregulation and impaired plasticity.

Genetic Factors

As noted, the BDNF Val66Met polymorphism affects activity-dependent BDNF secretion and may moderate treatment response, though findings are inconsistent. Polygenic risk scores for psychiatric disorders capture variance in plasticity-related pathways, but their clinical utility for treatment selection remains limited. Pharmacogenomic panels (e.g., CYP450 genotyping) affect drug metabolism rather than plasticity per se, and current evidence does not strongly support their routine use for improving outcomes (NNT for guided vs. unguided prescribing in the GUIDED trial: ~14).

Physical Fitness and Cardiometabolic Health

Baseline physical fitness moderates the neuroplastic effects of exercise interventions. Paradoxically, less-fit individuals often show larger relative BDNF increases and greater cognitive gains from exercise, suggesting that the plasticity-promoting effects are greatest where the deficit is largest. Comorbid metabolic syndrome — present in approximately 30–40% of patients with schizophrenia and 20–25% of patients with MDD — impairs BDNF signaling through insulin resistance and chronic inflammation, representing a modifiable barrier to neuroplastic recovery.

Psychiatric comorbidity is the rule rather than the exception, and understanding comorbidity through the neuroplasticity lens illuminates shared mechanisms and treatment challenges.

• MDD + Anxiety Disorders: Comorbidity rates of 50–60% reflect shared dysfunction in PFC-amygdala circuits and overlapping BDNF deficits. Combined conditions are associated with reduced hippocampal volumes, greater treatment resistance, and lower remission rates (~25–30% vs. ~35% for depression alone).
• MDD + Substance Use Disorders: Approximately 30–40% of individuals with MDD have a co-occurring substance use disorder. Chronic substance use (particularly alcohol and psychostimulants) produces lasting alterations in striatal dopaminergic plasticity and PFC-striatal connectivity, compounding the plasticity deficits of depression and creating a neurobiological substrate for treatment resistance.
• PTSD + TBI: Traumatic brain injury and PTSD co-occur in approximately 30–50% of military and civilian trauma populations. TBI directly disrupts neuroplastic mechanisms through axonal injury, neuroinflammation, and impaired BDNF signaling, while PTSD involves maladaptive plasticity in fear circuits. This combination is associated with poorer response to both psychotherapy and pharmacotherapy.
• Schizophrenia + Metabolic Syndrome: The metabolic comorbidity burden in schizophrenia, driven partly by antipsychotic medication effects and partly by shared genetic risk, creates a dual hit to neuroplasticity: antipsychotics (particularly second-generation agents) can impair BDNF signaling (though clozapine may be an exception), while metabolic dysfunction provides an independent route to neuroplastic impairment through insulin resistance and inflammation.

These comorbidity patterns underscore the need for integrated treatment approaches that address multiple plasticity-impairing factors simultaneously — combining pharmacotherapy with exercise, psychotherapy, and metabolic management rather than relying on any single intervention.

Several research frontiers are advancing the clinical translation of neuroplasticity science, though important limitations must be acknowledged.

Psychedelic-Assisted Therapy

Phase II trials of psilocybin for TRD (COMPASS Pathways) and MDD (Johns Hopkins, NYU) have yielded promising results, but phase III data are still emerging. Critical unanswered questions include optimal dosing, the necessity of the subjective psychedelic experience for efficacy, long-term durability, and safety in patients with psychotic spectrum vulnerability. The FDA has designated psilocybin a "Breakthrough Therapy" for TRD, but regulatory approval decisions are pending.

Non-Invasive Brain Stimulation

Transcranial magnetic stimulation (TMS) — particularly intermittent theta-burst stimulation (iTBS) — enhances cortical plasticity by mimicking endogenous plasticity-inducing stimulation patterns. The THREE-D trial (Blumberger et al., 2018) demonstrated that iTBS (3-minute sessions) was non-inferior to standard 10 Hz rTMS (37-minute sessions) for TRD, with remission rates of ~32% for both protocols. The accelerated Stanford Neuromodulation Therapy (SAINT) protocol — delivering multiple iTBS sessions per day over five days targeting the left DLPFC — achieved ~79% remission rates in an open-label study (Cole et al., 2020) and ~55% remission in a subsequent sham-controlled trial. If replicated, this represents a potential paradigm shift in how brain stimulation is delivered.

Biomarker Development

Serum BDNF, while correlating with depression severity at the group level, lacks the sensitivity and specificity for individual-level diagnostic or treatment-selection utility. More promising biomarkers under investigation include neuroimaging-based markers (e.g., hippocampal subfield volumes, default mode network connectivity as measured by fMRI), EEG-based plasticity indices (e.g., TMS-evoked potentials measuring cortical excitability), and multi-omics approaches integrating genetic, transcriptomic, proteomic, and metabolomic data.

Critical Limitations

Several caveats apply to the neuroplasticity framework:

• Human adult neurogenesis remains debated. Landmark studies (Spalding et al., 2013; Boldrini et al., 2018) provided evidence for ongoing hippocampal neurogenesis in adult humans, but Sorrells et al. (2018) reported that neurogenesis drops to undetectable levels after childhood. The discrepancy may reflect methodological differences in tissue processing and marker detection, but it introduces uncertainty about the human relevance of rodent neurogenesis findings.
• Peripheral BDNF is an imperfect proxy for central BDNF. Platelets store and release BDNF, and peripheral levels are influenced by physical activity, diet, time of day, and metabolic status.
• Publication bias likely inflates effect sizes for exercise, psychotherapy, and other behavioral interventions in psychiatric populations, a concern supported by asymmetric funnel plots in several meta-analyses.
• The neuroplasticity framework is broad enough to accommodate almost any finding, raising concerns about unfalsifiability. Specifically, both too much plasticity (e.g., maladaptive fear conditioning in PTSD) and too little plasticity (e.g., failure to extinguish fear) can be invoked as pathological, making the framework descriptive rather than predictive in some contexts.

Integrating neuroplasticity science into clinical practice does not require abandoning existing evidence-based frameworks. Rather, it provides a unifying rationale for multimodal treatment and informs sequencing, combination, and personalization decisions:

1. Assess plasticity-impairing factors: Chronic stress, inflammation (screen with hs-CRP), metabolic syndrome, sedentary behavior, sleep disruption (which impairs overnight memory consolidation and synaptic homeostasis), and substance use all impair neuroplasticity. Addressing these modifiable factors may enhance treatment response regardless of the primary intervention.
2. Combine plasticity-enhancing pharmacotherapy with structured therapeutic experience: The theoretical rationale for combining antidepressants with psychotherapy is strengthened by the plasticity framework — medication opens a window; therapy directs the remodeling. Meta-analytic data consistently show that combined treatment outperforms monotherapy for moderate-to-severe depression (remission rate advantage of approximately 10–15 percentage points).
3. Prescribe exercise as a first-line adjunct: Exercise should be recommended with the same specificity as medication — type (aerobic, resistance, or combined), frequency (at least 3 sessions per week), intensity (moderate to vigorous), and duration (minimum 30 minutes per session). The evidence base now supports exercise as a standalone treatment for mild-to-moderate depression and as an adjunct for moderate-to-severe depression, anxiety disorders, PTSD, and cognitive impairment in schizophrenia.
4. Consider neuroplasticity-targeted interventions for treatment resistance: For patients who have not responded to first-line treatments, interventions that directly enhance plasticity — ketamine/esketamine, TMS (particularly accelerated protocols), and, when available, psychedelic-assisted therapy — offer mechanistically distinct approaches that bypass monoaminergic treatment resistance.
5. Attend to timing and consolidation: Sleep supports synaptic homeostasis (Tononi and Cirelli's synaptic homeostasis hypothesis) and memory consolidation. Ensuring adequate sleep during active treatment — whether after a psychotherapy session, ketamine infusion, or exercise bout — may enhance the consolidation of treatment-induced plasticity.

The neuroplasticity framework offers clinicians not just a deeper understanding of why treatments work but a principled basis for optimizing how, when, and in what combination they are delivered. As the field moves toward precision psychiatry, neuroplastic biomarkers and genotyping may eventually guide individualized treatment selection — but even now, the existing evidence supports a plasticity-informed approach to care that is more integrative, more mechanistically coherent, and more attuned to the modifiable factors that determine who responds to treatment and who does not.