Bridging the Synaptic Gap: From Neuroplasticity to Cognitive Renewal

Abstract

Neurodegenerative diseases, particularly Alzheimer's disease (AD) and Parkinson's disease (PD), have a debilitating impact on memory, perception, and overall cognitive function. They are among the leading causes of dementia and cognitive decline: an estimated 57 million people were living with dementia worldwide in 2021, with nearly 10 million new cases every year, and over 60% of those affected living in low- and middle-income countries (WHO, 2021). While neuron death is a known factor, a more immediate and significant driver of decline is the loss of synaptic connections and the subsequent disruption of crucial neural networks. This paper addresses three core areas:

• The fundamental role of neuroplasticity in memory formation and cognitive recovery.
• The direct link between synaptic degradation and memory loss.
• The mechanisms of memory retrieval and the breakdown of long-term memory networks in neurodegeneration.

Building on this, the paper proposes a multifaceted approach to mitigation, unifying pharmacological, technological, and lifestyle-based elements into a comprehensive strategy for cognitive renewal. Crucially, it goes one step further than a survey of the biology: it ranks the leading intervention classes by the strength of the human evidence behind them, and confronts the widening gap between what the science can do and what patients can actually access. That access gap is not abstract, India alone is home to an estimated 8.8 million people aged 60 and over living with dementia (Lee et al., 2023), and the one recently approved disease-modifying drug costs roughly USD 26,500 per year (Eisai, 2023). The paper therefore closes with a concrete, prevention-first policy agenda anchored in existing Indian public-health infrastructure.

Background

This section covers several key terms used throughout the paper and gives a clearer picture of the underlying problem.

Neuroplasticity

The human brain is not a static organ; it is a dynamic system that constantly remodels and rewires itself in response to experience, injury, learning, and environmental stimuli. Neuroplasticity refers to the ability of the nervous system to change its structure and function by forming, modifying, or removing neural connections. These connections can be weakened through a process known as long-term depression (LTD) or strengthened through long-term potentiation (LTP). LTP represents a persistent increase in synaptic strength triggered by coordinated cellular activity, and forms the basis for memory encoding. LTD, conversely, involves a decrease in synaptic strength that is crucial for pruning unnecessary or incorrect connections. A prime example of this dynamic capability is adult hippocampal neurogenesis, a process that occurs in the dentate gyrus and contributes to learning and memory throughout life (Bliss & Collingridge, 1993; Kandel, 2001).

Memory Systems

Human memory is not a single entity but comprises multiple systems. Episodic memory, the recollection of specific episodes or experiences, relies heavily on the hippocampus and the related medial temporal lobes. Long-term memory storage, by contrast, involves vast cortical networks, including the hippocampus, medial prefrontal cortex, posterior cingulate, and angular gyrus, which together form the default mode network (DMN). These networks are profoundly disrupted in disorders like AD and PD. The pathological hallmarks of AD, namely amyloid-β plaques and tau tangles, specifically target the hippocampus and compromise DMN connectivity, severely affecting the ability to form and retain memories.

Synaptic Loss

A core unifying factor in the progression of neurodegenerative diseases is the degree of synaptic loss. In both AD and PD, post-mortem studies reveal that the severity of memory loss correlates more strongly with the degree of synaptic loss than with the outright death of neurons (Terry et al., 1991). Soluble oligomeric forms of amyloid-β (Aβ) show a high degree of synaptotoxicity, actively inhibiting LTP and enhancing LTD. This toxic activity leads to the shrinkage of dendritic spines, effectively erasing the physical points of synaptic contact (Spires-Jones & Hyman, 2014). In rodent models of PD, for instance, synaptic connections are disrupted even before memory deficits become apparent, indicating that the loss of synaptic potentiation may be the primary cause of cognitive decline. This is the paper's central claim, and the thread that runs through everything below: neurodegeneration is, first and foremost, a disorder of synapses and the networks they build, and synapses, unlike dead neurons, can in principle be protected and rebuilt.

Literature Review

Past Studies

Early work by researchers such as Bliss & Collingridge (1993) and Kandel (2001) established LTP and LTD as the foundational building blocks of neuroplasticity and memory. Seminal studies in neuropathology later linked dementia more directly to synapse loss than to overall neuron loss (Terry et al., 1991; Scheff & DeKosky, 1990). The devastating effect of Aβ oligomers on dendritic spine stability in AD was subsequently demonstrated in detail, confirming their role as primary synaptotoxic agents (Spires-Jones & Hyman, 2014; Mecca et al., 2020). In the context of PD, dopamine has been shown to gate plasticity, and its depletion degrades LTP and impairs certain memory-related behaviors, damage that is sometimes partly reversible with treatments like L-DOPA (Picconi & Calabresi, 2012). Modern PET imaging that binds the synaptic vesicle protein SV2A now allows synaptic density to be measured in living patients, confirming in vivo the synapse loss first seen at autopsy (Mecca et al., 2020).

Synaptic Plasticity and Memory Loss in Parkinson's and Alzheimer's Disease

A commonality of both AD and PD pathologies is synaptic dysfunction, albeit through different molecular pathways. In AD, amyloid peptides and hyperphosphorylated tau proteins disrupt synaptic function from the outset. They bind to synaptic sites, disrupt calcium homeostasis, and activate destructive pathways involving enzymes like calcineurin and caspases. The end result is a significant loss of synaptic receptors, which correlates directly with cognitive impairment, especially in the encoding of new memories. This has led many researchers to describe AD as a "synaptopathy", a disease primarily of synapses, even though neuron death becomes significant in its later stages. Furthermore, misfolded tau proteins, which normally help stabilize the cytoskeleton in neurons, accumulate in dendrites and synapses, creating a physical barrier to synaptic transport. The one break in the clouds, perhaps, is that synapses, in contrast to dead neurons, are potentially rescuable. Therapies such as monoclonal antibodies that clear amyloid peptides have shown measurable clinical effects, and immunotherapy remains an active area of testing.

In Parkinson's disease, the initial pathology involves the degeneration of dopamine-producing neurons in the midbrain, which primarily affects motor skills. However, the hippocampus and the prefrontal cortex also receive dopaminergic projections that are vital for facilitating neuroplasticity and learning. In vivo studies of Parkinson's rodent models confirm this, showing that hippocampal LTP is reduced early in the disease's progression. As mentioned, L-DOPA, a precursor to dopamine, can help restore LTP and improve memory, likely by modulating NMDA receptor function, which is critical for excitatory neurotransmission. Alpha-synuclein, the protein implicated in PD, normally plays a role in neurotransmitter release; when misfolded, however, it can directly harm synapses. Investigations of PD patient brains show that alpha-synuclein aggregates are present even in the absence of neuron loss and can cause structural changes in the hippocampus. Advanced PD can also lead to Parkinson's disease dementia (PDD), which resembles the dementia caused by AD but includes the presence of Lewy bodies formed by alpha-synuclein. A potential aid in certain cases is the use of cholinesterase inhibitor drugs, which raise levels of acetylcholine, another key neurotransmitter.

Although the two diseases involve different molecules (Aβ/tau proteins vs. α-synuclein), in both cases the pathology converges on weakened synaptic connections. Early detection of this synaptic decline could serve as a valuable diagnostic indicator, and new imaging techniques, such as PET tracers that target synaptic proteins, represent a significant step in that direction.

Neural Networks and How They Can Be Restored

Networks are crucial for the preservation of memory, because memory is not an isolated phenomenon involving a single brain element. Long-term storage and retrieval involve a coordinated dance between the hippocampus (acting as an index or pointer) and the cortex (where memories are stored). Over time, memories become less dependent on the hippocampus and more reliant on distributed cortical connections. Both AD and PD have devastating effects on this delicate interplay.

Network Disintegration in Alzheimer's Disease

In Alzheimer's disease, the most significant display of network disintegration occurs within the default mode network (DMN), which is deeply involved in memory retrieval and self-referential thought. One study found a decrease in DMN functional connectivity between the posterior cingulate cortex and the right hippocampus that correlated with cognitive decline and lower memory performance in AD patients (Greicius et al., 2004). Amyloid plaques are also heavily concentrated in DMN regions, extending the dysconnectivity. Some researchers therefore describe AD as a "progressive disconnection syndrome," in which the normal hippocampal-cortical dialogue fails (Brier et al., 2012). This is why, along with having trouble forming new memories, AD patients also lose old ones as these connections disintegrate. The Papez circuit (involving the fornix, mammillary bodies, hippocampus, cingulate cortex, and thalamus), which is vital for memory consolidation, is likewise affected as its components atrophy in AD.

Network Disruption in Parkinson's Disease

In Parkinson's disease, the early indications differ from those of AD, but the pathology begins to resemble AD in later stages. Sub-cortical loops are heavily involved in the initial pathology: the basal-ganglia-thalamo-cortical circuits are disrupted by the dopamine loss described earlier, leading to both motor and memory-retention problems. At rest, PD patients show reduced connectivity to the posterior DMN, similar to AD patients. A unique feature of PD, however, is the emergence of compensatory network activity, MRI shows the activation of additional brain regions during memory tasks, representing the brain's attempt to engage as many resources as possible to cope with reduced conventional memory circuits. As the disease progresses, this compensation becomes ineffective and memory becomes increasingly difficult. Unlike AD, where the memory circuits themselves are the primary target, in PD the neurotransmitter systems (dopamine, acetylcholine) are typically compromised first, and only then do the memory circuits degenerate.

Beyond structural integrity, the brain's memory network relies on dynamic coordination, the precise timing of neural activity, which is also affected. AD is associated with altered theta and gamma oscillations in the hippocampus, while in PD the normal pattern of cortical oscillations is disturbed by an excess of beta waves in the cortical-hippocampal network. The interconnection between networks such as the Papez circuit also suffers, leading to impairment in switching between different tasks and memories.

These findings suggest that therapeutic interventions for both AD and PD need to be network-informed. In AD, deep brain stimulation (DBS) of the fornix has been shown in early trials to improve memory retrieval by enhancing network activity. In PD, the effectiveness of DBS in increasing network connectivity is debated, but it is widely used to improve motor function. Individuals with high cognitive reserve can often use their compensatory networks more effectively, delaying the onset of symptoms, which suggests that cognitive training and intellectual engagement can themselves be a powerful remedy.

Proposed Solutions

A one-sided solution will not solve this problem. To be effective, the attack has to be waged on multiple fronts, pharmacological therapies, gene and molecular techniques, neurostimulation, and lifestyle change. As stated in the introduction, the shared goal of every one of these solutions is the same: to protect and enhance neuroplasticity so that synapses and networks survive.

Synapse-Protective and Synapse-Loss-Preventive Therapies

Because synapse loss is the feature that best tracks memory decline in both AD and PD, measures must be taken to protect existing synapses and promote the formation of new ones.

• In AD, anti-amyloid monoclonal antibodies such as lecanemab clear amyloid pathology and slow decline. In the CLARITY-AD phase 3 randomized trial (1,795 participants with early AD), lecanemab slowed decline on the CDR-SB scale by 27% over 18 months versus placebo, a statistically robust but modest absolute difference of 0.45 points, accompanied by amyloid-related imaging abnormalities (ARIA) in a meaningful minority of patients (van Dyck et al., 2023). Neurotrophic factors such as brain-derived neurotrophic factor (BDNF) support synaptic density, and nerve growth factor (NGF) has been delivered to the basal forebrain by gene therapy in small AD trials, showing trophic effects on neurons. Drugs such as ampakines, which prolong AMPA-receptor opening, and NMDA-receptor modulators are being explored to enhance cognition; in the earliest stages, such agents may help delay symptom onset by supporting LTP, though much of this evidence remains preclinical.
• In PD, since dopamine loss is central, optimizing dopaminergic therapy is a major treatment, though care is needed, as excess dopaminergic stimulation can disturb the LTP/LTD balance. Preliminary studies of molecules that stabilize dendritic spines have shown some prevention of amyloid-plaque-induced spine loss. Cholinesterase inhibitors are already licensed for PDD and AD. Low-dose psychedelics such as psilocybin have been proposed as potential promoters of synaptogenesis, but here clinical evidence is essentially absent and several risks must be weighed.

Enhancing Neurogenesis and Brain Repair

Since the cortical-hippocampal connection is impaired early in AD, approaches that boost neurogenesis are attractive. Strategies include lifestyle interventions such as exercise, neurotransmitter-based approaches, transplantation of stem cells, and administration of neurotrophic factors like BDNF and NGF. Most of these remain at the preclinical or early-trial stage.

Gene-Editing Approaches

Technologies like CRISPR could, in principle, enhance plasticity, for example by knocking down genetic inhibitors of axonal growth or building cellular resistance to toxic proteins. Proof-of-concept studies have already used CRISPR to reduce amyloid production in cells. It is important to be clear about the maturity of this evidence: these are cellular and animal proof-of-concept results, not human trials, and they sit at the far, experimental end of the ladder described below.

Lifestyle Interventions

This class of solutions is by far the most benign and risk-free, and, importantly, it rests on some of the strongest human evidence in the whole field. It includes physical exercise, cognitive training, and dietary optimization.

• Aerobic exercise. In a one-year randomized controlled trial of 120 older adults, a moderate aerobic-walking program (three days a week) increased hippocampal volume by about 2% (roughly 2.1% on the left and 2.0% on the right), effectively reversing one to two years of age-related shrinkage, with larger hippocampi linked to higher serum BDNF (Erickson et al., 2011). (The original draft attributed this "2%" figure to "older women"; the trial in fact enrolled both sexes.)
• Resistance training. In a six-month randomized controlled trial, twice-weekly weight (resistance) training significantly improved executive function and associative memory in older women with probable mild cognitive impairment, relative to a balance-and-toning control group (Nagamatsu et al., 2012). (The original draft's specific "15% memory improvement" figure could not be traced to a source and has been replaced with the study's verified qualitative finding.)
• Multidomain programs. The landmark FINGER trial randomized 1,260 at-risk older adults (aged 60 to 77) to a two-year program combining diet, exercise, cognitive training, and vascular-risk monitoring. The intervention group improved on the overall cognitive battery by roughly 25% more than controls, with even larger relative gains in executive function and processing speed (Ngandu et al., 2015).

Network Re-tuning (Neurostimulation)

As noted earlier, deep brain stimulation (DBS) has improved memory in early trials for both AD and PD patients. Non-invasive procedures such as transcranial magnetic stimulation (TMS) and transcranial direct current stimulation (tDCS) have also been explored: theta-burst TMS can enhance LTP formation, and tDCS applied to the prefrontal cortex has improved memory in some AD patients. This evidence base, however, rests on small and often mixed trials.

Ranking the Interventions by Evidence Strength

The claim that "neurodegeneration is a plasticity disorder" is now textbook consensus, and simply listing therapies risks implying they are equally proven. They are not. The more useful, and more contestable, question is: if synapse protection is the goal, which levers actually have the human evidence to pull today, and which are still promises? Ordering the intervention classes by the strength of their evidence turns a catalogue into a decision tool. From strongest human evidence to most speculative:

1. Lifestyle interventions (exercise, cognitive training, multidomain programs), strongest human RCT evidence. Aerobic exercise, resistance training, and the multidomain FINGER model are each backed by randomized controlled trials in older adults (Erickson et al., 2011; Nagamatsu et al., 2012; Ngandu et al., 2015). Effect sizes are modest and the benefit is prevention and slowing rather than reversal, but the risk profile is near-zero and the cost is low, which makes this the highest-value class overall.
2. Anti-amyloid immunotherapy (lecanemab, donanemab), proven in phase 3, but small and costly. Lecanemab is the first disease-modifying drug shown in a large RCT to slow decline (27% on CDR-SB; van Dyck et al., 2023) and is approved in several countries. But the absolute benefit is small, it carries an ARIA safety signal, it demands biweekly infusions and MRI monitoring, and, as the policy section shows, its cost has already led NICE to reject it for the UK's NHS (NICE, 2024).
3. Symptomatic neurotransmitter drugs (cholinesterase inhibitors, dopaminergic therapy), licensed, but symptomatic only. These are established and widely used, yet they treat the symptoms of neurotransmitter loss rather than halting the underlying synaptic and network decay.
4. Neurostimulation (DBS of the fornix, TMS, tDCS), promising but unproven at scale. Early trials show signals of benefit, but results are mixed, samples are small, and the invasive options carry surgical risk.
5. The molecular and regenerative frontier (NGF/BDNF gene therapy, stem cells, CRISPR editing, psilocybin), largely preclinical. These are the most exciting mechanistically and the least proven clinically; most evidence is from cells and animals or tiny early trials, and none should yet be presented to patients as established therapy.

The synthesis is deliberately falsifiable, a future head-to-head trial could, for instance, show immunotherapy outperforming exercise on hard clinical endpoints and reorder this list. But as of today the ladder makes an uncomfortable point clear: the interventions with the best evidence are also the cheapest and most widely available, while the newest and most expensive drug sits second, not first.

From Bench to Bedside: The Access Gap and a Policy Agenda

A white paper on cognitive renewal that stops at the biology leaves the most consequential question unanswered: who actually gets renewed? The synaptic gap this paper is named for has a second meaning, the gap between what laboratories can do and what patients can reach.

Consider the flagship of modern AD therapy. Eisai set the U.S. price of lecanemab at roughly USD 26,500 per year (Eisai, 2023), before the cost of the infusions, MRI monitoring, and specialist time it requires. When Britain's National Institute for Health and Care Excellence (NICE) weighed that price against a benefit it described as real but small, it declined to recommend lecanemab (and donanemab) for routine NHS use, concluding the benefits were "too small to justify the additional costs" (NICE, 2024). If one of the world's better-resourced public health systems judges this drug unaffordable, its relevance to a country like India, where an estimated 8.8 million people aged 60 and over already live with dementia (Lee et al., 2023) and most never receive a formal diagnosis, is close to zero.

This is where the evidence ladder becomes a policy instrument rather than an academic exercise. The single most powerful, most equitable lever is not the most expensive drug but prevention. The 2024 Lancet Commission on dementia estimates that up to 45% of dementia cases could be prevented or delayed by acting on 14 modifiable risk factors across the life course, including hypertension, hearing loss, untreated vision loss, diabetes, physical inactivity, smoking, high LDL cholesterol, air pollution, and low education (Livingston et al., 2024). These are precisely the same plasticity-protective, synapse-sparing levers this paper has argued for on biological grounds, now quantified as a population-level opportunity.

A concrete, actionable agenda for India, using institutions that already exist rather than proposing new ones, would be:

1. Embed brief cognitive screening and risk-factor management in primary care. India already runs the National Programme for the Health Care of the Elderly (NPHCE) and, under Ayushman Bharat, a national network of Health and Wellness Centres delivering comprehensive primary care and non-communicable-disease screening (Government of India, NPHCE). Adding a validated short cognitive screen plus systematic management of the Lancet Commission's 14 risk factors, starting with blood-pressure control, subsidized hearing aids, and cataract/vision correction, would convert existing infrastructure into a dementia-prevention engine at marginal extra cost.
2. Adopt cost-effectiveness gatekeeping before any public funding of anti-amyloid drugs. Rather than approving high-cost immunotherapies by default, India's public payers should apply an explicit NICE-style value threshold (cost per quality-adjusted life year), directing scarce funds toward interventions that buy more cognition per rupee.
3. Fund the cheap, high-evidence levers directly. Public reimbursement for hearing aids and vision correction, and community exercise and cognitive-engagement programs modelled on the FINGER trial (Ngandu et al., 2015), target the strongest rungs of the evidence ladder and reach the largest number of people, including the rural and low-income majority who will never see an infusion clinic.

Framed this way, "cognitive renewal" is not only a laboratory ambition but a distributional choice: whether a nation spends its limited health budget on a small, expensive benefit for a few, or a modest, well-evidenced benefit for millions.

Conclusion

Neurodegenerative diseases are fundamentally diseases of neuroplasticity. Toxic proteins such as amyloid-β, α-synuclein, and tau disrupt this vital process, leading to the disintegration of critical brain networks like the hippocampal-cortical system. But the brain's capacity for change also offers a path toward recovery. A future, personalized approach that uses biomarkers to combine the right drugs, aerobic and resistance exercise, cognitive training, and targeted brain stimulation, paired with advanced imaging, can help strengthen synapses, re-synchronize networks, and provide meaningful relief to patients with AD and PD. Yet the biology is only half the answer. Because the best-evidenced interventions are also the cheapest and most preventive, the decisive challenge is no longer only whether we can protect synapses, but who we choose to protect. Closing the synaptic gap will require closing the access gap alongside it, and, for the millions of people ageing into risk in countries like India, prevention embedded in everyday primary care is where cognitive renewal must begin.

Sources

1. World Health Organization, Dementia fact sheet, 57 million people living with dementia worldwide (2021), ~10 million new cases/year, >60% in low- and middle-income countries, US$1.3 trillion annual cost.
2. Lee et al. (2023), "Prevalence of dementia in India," Alzheimer's & Dementia, ~7.4% prevalence in adults aged 60+, ≈8.8 million people. (Figure corroborated by the accessible analysis below.)
3. Estimating the Prevalence of Dementia in India (PMC), restates the 7.4% (≈8.8 million) national estimate for adults 60+ and reports a comparable ~8.4% figure.
4. Livingston et al. (2024), Lancet Commission on dementia prevention, intervention and care, up to 45% of dementia potentially preventable via 14 modifiable risk factors.
5. Alzheimer Europe summary of the 2024 Lancet Commission, confirms 14 risk factors and the 45% prevention figure (LDL cholesterol +7%, vision loss +2%).
6. NBC News, Leqembi (lecanemab) priced at $26,500 a year, Eisai's U.S. list price and Medicare coverage limits.
7. Eisai, U.S. pricing announcement for Leqembi (lecanemab), wholesale acquisition cost of $26,500/year.
8. Alzheimer Europe, NICE final guidance recommending against routine NHS use of donanemab and lecanemab, benefits judged too small to justify the cost.
9. Eisai, full CLARITY-AD phase 3 results (van Dyck et al., 2023, NEJM), 1,795 participants, 27% slowing of decline on CDR-SB at 18 months (difference 0.45), ARIA rates reported.
10. Erickson et al. (2011), PNAS, Exercise training increases size of hippocampus and improves memory, RCT, 120 older adults, ~2% hippocampal-volume increase after one year of aerobic walking.
11. Review confirming Nagamatsu et al. (2012) resistance-training RCT, twice-weekly resistance training over six months improved executive function and associative memory in older women with probable MCI.
12. Ngandu et al. (2015), FINGER trial (via review, PMC), 1,260 at-risk older adults, 2-year multidomain RCT; ~25% greater improvement on the cognitive battery vs. control.
13. Government of India, National Programme for the Health Care of the Elderly (NPHCE), official MoHFW programme delivering preventive and primary elderly care, integrated with Ayushman Bharat Health and Wellness Centres.
14. "Adding health to years: A review of the NPHCE in India" (PMC), confirms NPHCE was launched by MoHFW in 2011 and now delivers primary elderly care and NCD screening through Ayushman Bharat Health and Wellness Centres.