How Does Combined Audiovisual Flicker Stimulation at Gamma Frequency More Effectively Modulate Alzheimer's Disease?

I. Overview of the Pathophysiology of MCI/AD

Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by hallmark pathologies: deposition of β- amyloid (Aβ) forming senile plaques, hyperphosphorylation of tau leading to neurofibrillary tangles, neuroinflammation (activation of microglia and astrocytes), synaptic dysfunction, and abnormal neural network activity. In the early stage, patients often present with mild cognitive impairment (MCI), where significant Aβ and tau pathology is already present in the brain but cognitive impairment remains relatively mild — representing a "golden window" for intervention. The default mode network (DMN) is the most active network during resting states, with core nodes including the posterior cingulate cortex (PCC), precuneus (PCu), and medial prefrontal cortex (mPFC). Numerous studies have shown that functional connectivity between the PCC and PCu is significantly reduced in AD and MCI patients, and this reduction is closely correlated with cognitive decline. Additionally, neural oscillations (especially in the gamma band, around 40 Hz) are abnormal in the AD brain, manifesting as reduced power or impaired synchrony, which may affect the integration and processing of neural information.[1]

 

II. Mechanism of Visual Gamma Stimulation in Improving MCI/AD

Both this study and previous animal experiments indicate that 40 Hz visual flicker stimulation (delivered via wearable light goggles) can drive gamma- band neural activity in the visual cortex. This driving effect is not a simple sensory response; rather, continuous, rhythmic visual input induces neurons to fire synchronously at 40 Hz. In AD mouse models, such visual gamma stimulation significantly reduces Aβ deposition and alters the morphology and function of microglia – shifting them from a proinflammatory state toward a phenotype with enhanced phagocytic clearance capacity. Although the clinical trial did not use visual stimulation alone (it was always combined with auditory stimulation), prior mechanistic studies suggest that visual gamma stimulation, by modulating neuroplasticity in sensory cortices and activating local immune cells, may delay or ameliorate early AD pathology. In the human trial reported here, after combined audiovisual stimulation, participants exhibited clear 40 Hz neural entrainment on EEG (Figure 1A shows that all participants at all time points showed 40 Hz entrainment in multiple channels), with the visual pathway contributing a major driving source. The core purpose of achieving entrainment was to confirm that sensory stimulation successfully drove the intended neural activity (i.e., target engagement), serving as the “trigger” and biomarker for subsequent biological effects (such as enhanced functional connectivity and altered immune factors). It also allows assessment of neuroplastic changes (e.g., a mild power decrease after 8 weeks suggesting homeostatic adaptation) and provides direct evidence for clinical translation feasibility and optimization of stimulation parameters.[2]

Figure 1 evaluates the effects of different durations (4 weeks no stimulation, 4 weeks stimulation, 8 weeks stimulation) on 40 Hz neural entrainment.

Figure 1 – Changes in Gamma Entrainment with Duration of Stimulation

ExposureFigure 1A shows the number of EEG channels (out of 64) exhibiting 40 Hz entrainment pre vs. post each intervention period. Each dot represents one subject; side black lines indicate mean ± SEM. No significant differences were seen between pre and post in any group (P>0.11), indicating stable entrainment channel count.
Figure 1B displays the change in average 40 Hz relative power (i.e., percentage of 40 Hz power relative to total 150 Hz power) across all channels pre vs. post intervention. After 8 weeks of stimulation, a small but statistically significant decrease in 40 Hz relative power was observed (P=0.04).
Figure 1C presents channelwise analysis of changes in 40 Hz relative power (tvalue heatmap). Cool colors indicate power decrease; the most significant decreases were in the left occipital region (P7 and O1), suggesting that longterm stimulation may induce local homeostatic plasticity adaptation.
Figure 1D shows power spectra (logtransformed) from 150 Hz; blue = preintervention, orange = postintervention. Only a slight decrease at 40 Hz is seen, with no significant changes in other bands (delta, theta, alpha, beta).
Overall interpretation of Figure 1: All participants showed stable 40 Hz entrainment, but longterm stimulation led to a mild power decrease in the left occipital region.

 

III. Mechanism of Auditory Gamma Stimulation in Improving MCI/AD

Similarly, 40 Hz auditory stimulation (delivered via headphones) can drive gamma oscillations in the auditory cortex and related brain regions. The auditory steady- state response (ASSR) is a wellestablished measure in neuroscience; this study used the same principle, presenting 40 Hz sound synchronized with light.[6] In mouse studies, auditory gamma stimulation alone also reduces Aβ burden, modulates microglial activity, and improves spatial memory.

Its core mechanisms may include:

Enhancing longterm potentiation (LTP) signaling pathways, promoting synaptic plasticity;

Modulating the cytokine network, e.g., reducing proinflammatory factors such as TWEAK (TNF- like weak inducer of apoptosis). In the present study, although the effect of auditory stimulation cannot be isolated, all enrolled participants were required to show stable entrainment to 40 Hz audiovisual stimulation at screening, suggesting that both auditory and visual pathways contribute to maintaining gamma oscillations (as shown in Figure 1B, 40 Hz power proportion remained stable during stimulation).

 

IV. Mechanism of Combined Audiovisual 40 Hz Flicker Stimulation in Improving MCI/AD

The mechanism of combined audiovisual 40 Hz flicker stimulation is not simply additive; rather, it enhances effects on wholebrain networks through multisensory integration. Animal studies show that simultaneous light and sound stimulation drives a broader range of brain regions (including higher cognitive centers such as the hippocampus and prefrontal cortex) into gamma oscillations, outperforming singlemodality stimulation.[3] Its pathways include:

1. Neural entrainment and network reshaping[4]
Continuous audiovisual stimulation forces sensory cortices and DMN nodes to synchronously fire at 40 Hz. In this study, restingstate fMRI showed that after 8 weeks of daily one- hour audiovisual flicker stimulation, functional connectivity between the PCC and PCu was significantly enhanced (Figure 3B, p=0.016), a connection that is typically weakened in AD patients. This change suggests that combined audiovisual stimulation may repair impaired brain network communication.

2. Neuroimmune modulation
After 8 weeks of stimulation, multiple immune factors in the CSF changed. A latent variable (LV1) composed of several cytokines significantly increased (Figure 4C, p=0.02), with TWEAK significantly decreased (Figure 4D, p=0.04), and TGFα, MIP1β, etc., showing decreasing trends (Figure 4E). These factors are involved in microglial/astrocyte activation and migration. Downregulating these proinflammatory signals may reduce neuroinflammation, thereby creating a healthier microenvironment for neurons.

 

V. Clinical Study

1. Methods

Design: Delayed- start, nonblinded, single- arm feasibility trial (ClinicalTrials gov: NCT03543878).

Participants: 10 patients with prodromal AD (MCI stage) confirmed by biomarkers (CSF Aβ₄₂, p- tau).

Intervention: Daily one- hour home- based 40 Hz audiovisual flicker (light goggles + headphones), divided into a 4week group (no stimulation first 4 weeks, then 4 weeks stimulation) and an 8- week group (direct stimulation for 8 weeks).

Assessments: Baseline, midpoint, endpoint.

Primary outcomes: Safety, tolerability, adherence. Exploratory outcomes: EEG entrainment, resting- state fMRI functional connectivity, CSF Aβ₄₂/tau/immune factors.

 

2. Results
Safety, tolerability, and adherence
No serious related adverse events; common mild events were dizziness, tinnitus, headache. Of 17 screened individuals, only 1 withdrew due to intolerance and 1 due to worsening tinnitus. After enrollment, adherence was excellent: mean adherence 95.5%, all participants >89%, and weekly adherence remained above 88% (Figure 2D). Nine out of 10 participants voluntarily entered an openlabel extension, further supporting the intervention's acceptability.

Figure 2 – Tolerability and Adherence of Gamma Flicker Stimulation

 

Figure 2 overall shows participants' tolerance to 40 Hz audiovisual flicker and home- use adherence. Figure 2A depicts the study design overview: horizontal axis = weeks, gray bars = no- stimulation periods, yellow/black striped bars = daily 1- hour stimulation periods, vertical marks indicate clinical visits (baseline, midpoint, endpoint, including EEG, MRI, lumbar puncture). Figure 2B (top) shows the stimulation devices (light goggles and headphones), and (bottom) shows the scale used to assess tolerance (from “no sensation” to “intolerable”). Figure 2C shows each participant's maximum tolerated intensity for visual- only, auditory- only, and combined stimulation at baseline (expressed as percentage of device maximum output). Nine participants tolerated >70% of maximum; only one lower. Figure 2D (left) shows average adherence for each participant during the 4- 8 week intervention (all >89%, overall mean 95.5%); (right) shows weekly adherence, with dark and light gray representing the 8- week stimulation group (F/F) and the delayed- start group (N/F), respectively. Weekly adherence remained stable above 88%. Conclusion of Figure 2: Gamma flicker stimulation is highly tolerable and has good home adherence in MCI patients.

3. Neural entrainment
All participants showed robust 40 Hz EEG entrainment at each assessment. On average, 49/64 channels were entrained (Figure 1A). Notably, after 8 weeks of stimulation, 40 Hz relative power showed a small but statistically significant decrease (Figure 1B right, p=0.04), primarily in the left occipital region (Figure 1C right, P7/O1 channels). The researchers speculate this may represent homeostatic plasticity adaptation rather than loss of efficacy.

4. Functional connectivity changes
Resting- state fMRI analysis showed that 8 weeks of stimulation significantly increased functional connectivity between the PCC and PCu (Figure 3B right, p=0.016), while connectivity between the PCC and mPFC showed no significant change (Figure 3C). Because PCC- PCu connectivity is often weakened in AD, this change suggests normalization toward the healthier state.

Figure 3 – Changes in Resting-State Functional Connectivity with Duration of Stimulation Exposure

 

Figure 3 assesses the effect of gamma flicker stimulation on default mode network functional connectivity. Figure 3A shows the locations of three ROIs: PCu (blue), PCC (red), and mPFC (green), core nodes of the DMN. Figure 3B compares functional connectivity strength (Fisher's Z) between PCC and PCu pre vs. post intervention. Only after 8 weeks of daily stimulation did connectivity significantly increase (P=0.016); no significant change occurred during the 4- week no- stimulation period or the 4- week stimulation period. Figure 3C shows PCC- mPFC connectivity, with no statistically significant change in any intervention phase (P>0.13). Conclusion of Figure 3: Long- term (8- week) gamma flicker stimulation selectively enhances the AD- vulnerable PCC- PCu connection, while PCC- mPFC connectivity is unaffected, suggesting network- specific restorative effects.

5. AD pathology proteins
CSF Aβ₄₂, p- tau, t- tau, and their ratios showed no statistically significant changes after 4 or 8 weeks of stimulation.

6. Immune factor changes
To control for individual differences in Aβ₄₂ and p- tau levels, linear- corrected partial least squares discriminant analysis (PLSDA) was performed. A latent variable, LV1, was extracted, which significantly increased after 8 weeks of stimulation compared to baseline (Figure 4C, p=0.02).

 

TWEAK significantly decreased (p=0.04, Figure 4D)

TGF- α, MIP- 1β, DNER, IL- 5 showed decreasing trends (Figure 4E).
These factors are largely involved in glial activation, chemotaxis, and proliferation; their downregulation suggests mitigation of neuroinflammation.

Figure 4 – Changes in Cytokines and Immune Factors in Cerebrospinal Fluid After 8 Weeks of Flicker Stimulation

 

Figure 4 comprehensively displays the modulation of the CSF immune profile after 8 weeks of gamma flicker stimulation.
Figure 4A is a data matrix of CSF immune factors across all participants at different time points (corrected for Aβ₄₂ and p- tau). Each row represents one sample (baseline, 4- week, 8- week time points for the same participant), each column a factor. A green box highlights the most downregulated factors in LV1.
Figure 4B shows the weighted composition of LV1. Negative weights indicate that the factor decreases when LV1 increases; positive weights indicate increase. Most immune factors (e.g., TWEAK, MIP- 1β, TGF- α) have negative weights.
Figure 4C compares LV1 values across time points (paired t- tests). No significant change after 4 weeks (P=0.11), but after 8 weeks LV1 significantly increased (P=0.02), indicating that long- term stimulation is required to produce a significant shift in the immune profile.
Figure 4D shows the change in TWEAK (TNF- like weak inducer of apoptosis) levels. After 8 weeks, TWEAK significantly decreased (P=0.04).
Figure 4E lists other factors with decreasing trends (TGF- α, MIP- 1β, DNER, IL- 5); although not statistically significant, the trends are consistent.
Overall interpretation of Figure 4: Eight weeks of gamma flicker stimulation induces downregulation of multiple pro- inflammatory factors in CSF, with the most clear- cut decrease in TWEAK, indicating effective modulation of the neuroimmune system.

 

VI. Summary

This study is the first to systematically evaluate the feasibility of daily home- based 40 Hz audiovisual gamma flicker stimulation in biomarker- confirmed prodromal AD patients. Key conclusions:

Dimension	Conclusion
Safety	No serious adverse events; low incidence of mild adverse reactions; individuals with epilepsy/migraine/severe tinnitus were excluded.
Tolerability and Adherence	Extremely high (mean adherence rate 95.5%; 90% of participants chose to continue for 1 year).
Target Engagement	Stable 40 Hz neural entrainment achieved in all participants (confirmed by EEG).
Network Effect	After 8 weeks of stimulation, AD- vulnerable PCC- PCu functional connectivity significantly increased (fMRI).
Immune Effect	Pro- inflammatory factors such as TWEAK were downregulated in CSF, indicating modulation of the neuroimmune system.[5]
Pathological Protein	No changes in Aβ or tau levels observed within 4- 8 weeks; longer treatment may be required.

 

Limitations: small sample size, no sham control, relatively short intervention duration, exclusion of some common comorbidities. Nonetheless, these findings strongly support that gamma sensory flicker is a safe, feasible, non- pharmacological, non- invasive home intervention with preliminary biological effects. Future larger, longer, placebo- controlled trials are needed to validate clinical efficacy.

Final conclusion: This study takes a critical first step toward translating gamma sensory stimulation into a novel therapeutic approach for Alzheimer's disease.

 

 

 

[1] Hunter F. Iaccarino, Annabelle C. Singer, et al. Gamma frequency entrainment attenuates amyloid load and modifies microglia. Nature. 540, pages230–235 (2016)

https://www.nature.com/articles/nature20587

 

[2] Annabelle C. Singer, Anthony J. Martorell,et al. Noninvasive 40-Hz light flicker to recruit microglia and reduce amyloid beta load. Nature. 13, pages1850–1868 (2018)

https://www.nature.com/articles/s41596-018-0021-x

 

[3] Anthony J. Martorell, Abigail L. Paulson, et al. Multi-sensory Gamma Stimulation Ameliorates Alzheimer’s-Associated Pathology and Improves Cognition. Sciencedirect. Volume 177, Issue 2, 4 April 2019, Pages 256-271.e22

https://www.sciencedirect.com/science/article/pii/S0092867419301631

 

[4] Chinnakkaruppan Adaikkan, Steven J. Middleton, et al. Gamma Entrainment Binds Higher-Order Brain Regions and Offers Neuroprotection. Sciencedirect. Volume 102, Issue 5, 5 June 2019, Pages 929-943.e8

https://www.sciencedirect.com/science/article/pii/S0896627319303460

 

[5] kristie M. Garza, Lu Zhang, et al. Gamma Visual Stimulation Induces a Neuroimmune Signaling Profile Distinct from Acute Neuroinflammation. jneurosci. Journal of Neuroscience 5 February 2020, 40 (6) 1211-1225

https://www.jneurosci.org/content/40/6/1211

 

[6] R Galambos, S Makeig, et al. A 40-Hz auditory potential recorded from the human scalp. Pnas. April 15, 1981,78 (4) 2643-2647

https://www.pnas.org/doi/10.1073/pnas.78.4.2643