Neuroscience, Vaccines, and Autism: What Science Actually Says and Doesn’t Say

An honest look at vaccine biology, autism research, and a hypothesis that may offer real hope for families of nonspeaking children.

Comments prepared in support of my upcoming presentation at the Autism Health Conference 2026

You can register to attend here.

---

If you are raising a severely autistic child, especially a nonspeaking one, you have probably spent years navigating a medical and scientific establishment that often felt dismissive of your observations, your instincts, and your questions. You have watched your child struggle. And many of you watched something change around the time of a vaccination appointment. Your child was developing. Then, in the days or weeks that followed a vaccine visit, something shifted. Language that was emerging went quiet. Eye contact that was growing became rare. A child who had been turning toward the world seemed to turn away from it.

You reported this to your pediatrician and were told it was a coincidence. You found other parents who described the same sequence of events and were told you were part of a community built on fear. You were handed studies and told the question was settled. But the question did not feel settled, because you were there. You saw what you saw.

This piece takes that observation seriously. Not because parental observation is infallible, and not because the science supports every conclusion drawn from it, but because a pattern reported independently by thousands of families across decades and across countries is not nothing. It is a signal that deserves honest engagement rather than dismissal. The goal here is to examine what science actually knows, what it does not know, and where the honest uncertainties lie.

* * *

The Microstroke Hypothesis: Where It Comes From

The microstroke hypothesis proposes that vaccine-induced inflammation could, in some infants, cause tiny subclinical injuries to blood vessels in the brain. These would be too small to show up on standard imaging but potentially significant enough to affect neurodevelopment. It is not a mainstream scientific position. No major medical body endorses it, and there is no direct imaging evidence that it occurs.

But the hypothesis is not invented from nothing. It draws on real biology.

THE ALUMINUM QUESTION

Researchers at INSERM (Institut National de la Santé et de la Recherche Médicale, France’s national health research agency), led by Romain Gherardi, have published peer-reviewed work showing that aluminum from vaccine adjuvants can be taken up by immune cells and transported to brain tissue in animal models [1,2]. Once there, it triggers microglial activation and neuroinflammation. Critics note that the animal doses differ from human vaccine schedules, and that human brain accumulation has not been directly demonstrated. The research is real, the debate about its relevance is legitimate, and the question has not been fully resolved.

THE INFLAMMATORY PATHWAY

Vaccines work by triggering an immune response. That response includes a transient surge of inflammatory cytokines. This is not a side effect; it is the mechanism. What is genuinely uncertain is whether, in a small subset of infants with particular genetic vulnerabilities, this cytokine surge could be exaggerated enough to temporarily compromise the blood-brain barrier. The blood-brain barrier in infants is less mature than in adults. The biology here is plausible at a theoretical level. It has not been demonstrated at vaccine doses.

VITT: PROOF THAT THE IMMUNE SYSTEM CAN AFFECT CEREBRAL VESSELS

One of the more important developments in recent years was the documented discovery of VITT (Vaccine-Induced Immune Thrombocytopenia and Thrombosis), a rare thrombotic complication of adenoviral vector COVID-19 (coronavirus disease 2019) vaccines [3]. This condition involves immune-mediated clotting, including in cerebral blood vessels. It was initially missed, then denied, then confirmed. It matters not because it is directly analogous to childhood vaccine injury, but because it established beyond doubt that vaccine-induced immune mechanisms can, in rare cases, affect brain vasculature. The possibility space is not zero.

WHAT THE EVIDENCE ACTUALLY SUPPORTS The most defensible version of the microstroke hypothesis is not a general claim about all vaccines in all children. It is a narrower claim: that children with specific genetic vulnerabilities, including immune dysregulation or mitochondrial disorders, might respond atypically to immune activation, potentially experiencing neurological effects that most children do not. This subgroup model is unproven but not biologically incoherent.

* * *

The Autism Connection: What Is Plausible and What Is Not

The microstroke hypothesis, if it has any relevance to autism, operates through a specific neurobiological mechanism and deserves to be evaluated on its own terms, separate from the broader and long-contested public debate over vaccines and autism. The question here is not about that debate. It is about whether inflammation-mediated cerebrovascular injury is a plausible pathway to neurodevelopmental harm in a susceptible subset of children.

The dominant scientific hypothesis holds that autism’s origins are largely prenatal. Prospective studies of infants at high genetic risk for autism, watched from birth, show measurable differences in brain development and social attention within the first months of life, well before vaccination could play any role. This represents the mainstream position and poses a genuine challenge to broad postnatal vaccine-injury models of autism. It is worth noting, however, that a hypothesis being dominant is not the same as it being complete. The prenatal origin framework accounts well for many autistic individuals but has more difficulty explaining regressive presentations, and researchers continue to investigate what additional factors may shape outcomes after birth.

WHERE THE QUESTION REMAINS GENUINELY OPEN

Regressive autism is different. Roughly 20 to 30 percent of autistic children appear to develop typically and then lose skills, often language and social engagement, somewhere between 12 and 24 months of age. This regression is real, documented, and poorly understood. Its timing does overlap with the vaccine schedule. And it is worth saying plainly: the parents who report watching their child change in the days and weeks following vaccination are not, as a group, confused or misremembering. Regression happens. The timing they describe is real. The unresolved scientific question is not whether the regression occurred but what caused it.

This distinction matters. For too long, the response from the medical establishment has been to challenge the parents’ account rather than to engage the mechanism. That is not good science, and it has damaged trust in ways that will take a long time to repair. The honest position is to acknowledge the reported pattern, take it seriously as an observation, and invest in the research needed to understand it.

Autistic brains, examined postmortem, show neuroinflammatory signatures, activated microglia, elevated cytokines, and white matter abnormalities [4,5]. These findings are consistent with what you might expect from microvascular injury, though they almost certainly have other explanations as well. The point is not that vaccines caused these findings, but that the brain biology of autism involves immune and vascular components that researchers are still working to understand.

We do not fully understand regressive autism. Ruling out postnatal contributions on political rather than scientific grounds would itself be a failure of honest inquiry.

The most credible narrow claim, and it is narrow, is that in a subgroup of children with underlying immune or metabolic vulnerabilities, a strong immune activation event, possibly including vaccination, could act as a trigger for neurological regression in children already on a susceptible developmental trajectory. This has not been demonstrated. It has not been ruled out.

* * *

The Locked-In Hypothesis: A Different Kind of Hope

Whatever one believes about the causes of autism, there is a question that matters enormously for families right now: for nonspeaking autistic children, is the absence of communication a reflection of absent thought, or is it a reflection of a motor system that cannot reliably carry thought into the world?

This is the locked-in hypothesis. And here, the neuroscience offers something genuinely encouraging.

WHAT THE BRAIN RESEARCH SHOWS

Apraxia of speech is a well-documented condition in which a person knows what they want to say but cannot reliably execute the motor sequence required to say it. The brain regions implicated in autism, the supplementary motor area, the basal ganglia, the cerebellum, are precisely the regions involved in motor initiation and sequencing. Postmortem studies of autistic brains consistently show Purkinje cell loss in the cerebellum, a region critical for the timing and coordination of voluntary movement, including speech [6,7].

Eye-tracking studies, which assess comprehension without requiring any motor output, have found language understanding in some nonspeaking autistic individuals that far exceeds what behavioral testing would predict. Neuroimaging during language tasks has shown activation patterns inconsistent with global cognitive impairment. These findings do not prove that every nonspeaking autistic person has intact cognition. They do suggest that standard assessments, which rely heavily on verbal and fine motor responses, may systematically underestimate what is happening inside these children’s minds.

Some nonspeaking autistic individuals who later gained communication access have described years of awareness without any means of expressing it. These accounts are internally consistent, sophisticated, and in some cases verifiable. They deserve to be taken seriously.

LETTER BOARDS AND EMERGING COMMUNICATION METHODS

It is important to be honest about the controversies here, because parents deserve accurate information.

Facilitated communication, in which a facilitator physically supports a person’s hand while they point to letters, has been shown in controlled research to reflect the facilitator’s output rather than the communicator’s [8]. The facilitator is not consciously fabricating; the mechanism is the ideomotor effect, the same phenomenon behind Ouija boards. This is not a fringe finding. It is a well-replicated result, and FC (facilitated communication) has caused real harm to families through false accusations and misdirected hope.

But there are newer approaches that operate differently in principle. The Rapid Prompting Method (RPM) and Spelling to Communicate (S2C) explicitly work toward independent pointing, with the goal of fading physical support entirely. Some practitioners conduct message-passing tests to verify independence. Some individuals using these methods have transitioned to fully independent communication and gone on to write books, give public talks, and advocate for themselves in ways that could not plausibly be attributed to a facilitator.

These methods are not yet validated by the kind of large controlled trials that would satisfy a skeptical scientific audience. That is a real limitation and families should know it. But the methods are not equivalent to FC, the evidence is not uniformly negative, and the stakes for getting this wrong in either direction are enormous.

THE CORE UNANSWERED QUESTION We do not know how many nonspeaking autistic individuals have intact cognition that standard assessment cannot reach. This is one of the most important unanswered questions in autism research, and it has been inadequately studied, partly because the political environment around these questions has made honest inquiry more difficult than it should be.

* * *

New Medicines on the Horizon

The communication barrier is not the only frontier where the outlook is changing. For the first time in the history of autism medicine, researchers are developing compounds that target the core social symptoms of the disorder, not just the behavioral symptoms that have always been easiest to manage pharmacologically. This is a meaningful shift, and families deserve to know about it.

Currently approved medications for autism, risperidone and aripiprazole, address irritability and agitation. They carry significant side effect burdens and do nothing for the social and communicative dimensions of the condition that most affect quality of life. For decades, this was considered the ceiling. It no longer is.

A PROMISING EXAMPLE: L1-79

One of the most advanced compounds in this new generation is L1-79, developed by Yamo Pharmaceuticals. It works through a mechanism that would have seemed counterintuitive even a decade ago: rather than adding to neurotransmitter activity, it reduces it. Specifically, L1-79 inhibits tyrosine hydroxylase, the enzyme that controls the rate of dopamine and norepinephrine production in the brain. The scientific rationale is that many autistic individuals, particularly those with more severe presentations, have an overactive catecholamine system, and that excess signaling through these pathways contributes to the social communication deficits and excitatory imbalances that define the disorder.

The drug is not without precedent. The L-isomer of the same molecule, metyrosine, is already FDA (Food and Drug Administration)-approved for pheochromocytoma, a condition involving catecholamine excess. L1-79 uses a racemic formulation adapted for use in autism, giving it a pharmacological foundation that is established even if the application is new.

The clinical results, presented in full at the International Society for Autism Research (INSAR) conference in 2025, are the most compelling data yet seen for a compound targeting core autism symptoms [9]. In a rigorous 12-week, double-blind, placebo-controlled crossover study of 58 adolescents and young adults, L1-79 produced a 7.94-point advantage over placebo on the Vineland Adaptive Behavior Scales socialization score. The minimal clinically important difference for that instrument is 4 points. The drug nearly doubled it. Families participating in the trial also reported meaningful improvements in what they identified as the three most bothersome symptoms of their child’s autism. There were no serious adverse events. No one dropped out because of side effects.

The FDA granted L1-79 Fast Track Designation in 2018, reflecting the agency’s recognition that there is a serious unmet need in this population [9]. Phase 3 trials are now in preparation. That is not a guarantee of approval, and the path from Phase 2 to the pharmacy shelf is long and often disappointing. But the direction of travel is real, and it points somewhere new.

L1-79 is one example among several compounds now in development that approach autism differently, targeting neurobiological mechanisms rather than suppressing surface behaviors. The field is not where families deserve it to be. But it is moving, and it is moving in a direction shaped by a growing scientific understanding that the brains of autistic individuals are not simply broken versions of neurotypical brains, but are organized differently in ways that specific, targeted interventions may be able to address.

* * *

What We Owe These Children

Parents of nonspeaking autistic children have often been told, implicitly or explicitly, that their child is not “in there” in the way they sense. They have been handed devastating prognoses based on assessments that may not capture what these children actually know and feel. Many have spent years grieving a connection they were told was not possible.

The science does not yet tell us with certainty how many of these children have rich inner lives waiting for a reliable path out. But it tells us enough to know that presuming they do not is not the conservative scientific position. It is its own kind of risk. A child with intact cognition who is treated as cognitively absent suffers a harm that is profound and largely invisible.

The emerging evidence around motor-speech dissociation, the documented cases of nonspeaking individuals who found their voice through letter boards and augmentative communication, and the neuroimaging data suggesting preserved language networks in some nonverbal autistic individuals all point in the same direction: the story is not over. The barrier may be at the output, not at the source.

For families, the practical implication is worth sitting with. If your nonspeaking child is in there, aware, understanding more than they can show, then every interaction you have with them, every book you read aloud, every conversation you have in their presence, every moment you treat them as a full person, matters in ways you may not be able to measure yet. And the work being done to build reliable, independently verifiable communication pathways for this population is among the most important work in autism today.

A Note to Parents Who Have Questions

If you have watched your child and felt that the official answers did not fit what you observed, you are not irrational. Some of the most important scientific questions about autism remain genuinely open. The research community is not infallible, and the history of medicine includes many cases where parental observation preceded scientific confirmation by years or decades.

At the same time, the most important thing for your child right now is not resolving debates about causation. It is finding every possible way to presume their competence, to offer them tools for expression, and to meet them where they are.

The locked-in hypothesis, if it holds for your child, means that what you have always sensed is true: they are there. They have been there all along. And the distance between you may be shorter than anyone has told you.

Science is slowly learning to find them. Do not stop looking.

References

[1] Khan Z, Combadière C, Authier FJ, et al. Slow CCL2-dependent translocation of biopersistent particles from muscle to brain. BMC Medicine. 2013;11:99. https://link.springer.com/article/10.1186/1741-7015-11-99

[2] Gherardi RK, Eidi H, Crépeaux G, Authier FJ, Cadusseau J. Biopersistence and Brain Translocation of Aluminum Adjuvants of Vaccines. Frontiers in Neurology. 2015;6:4. https://www.frontiersin.org/articles/10.3389/fneur.2015.00004/full

[3] Pavord S, Scully M, Hunt BJ, et al. Clinical Features of Vaccine-Induced Immune Thrombocytopenia and Thrombosis. New England Journal of Medicine. 2021;385:1680-1689. https://www.nejm.org/doi/full/10.1056/NEJMoa2109908

[4] Vargas DL, Nascimbene C, Krishnan C, Zimmerman AW, Pardo CA. Neuroglial activation and neuroinflammation in the brain of patients with autism. Annals of Neurology. 2005;57(1):67-81. https://pubmed.ncbi.nlm.nih.gov/15546155/

[5] Morgan JT, Chana G, Pardo CA, et al. Microglial activation and increased microglial density observed in the dorsolateral prefrontal cortex in autism. Biological Psychiatry. 2010;68(4):368-376. https://pubmed.ncbi.nlm.nih.gov/20674603/

[6] Bauman ML, Kemper TL. Histoanatomic observations of the brain in early infantile autism. Neurology. 1985;35(6):866-874. https://pubmed.ncbi.nlm.nih.gov/4000488/

[7] Whitney ER, Kemper TL, Bauman ML, Rosene DL, Blatt GJ. Cerebellar Purkinje cells are reduced in a subpopulation of autistic brains: a stereological experiment using calbindin-D28k. Cerebellum. 2008;7(3):406-416. https://pubmed.ncbi.nlm.nih.gov/18701892/

[8] American Psychological Association. Resolution on Facilitated Communication. 1994; reaffirmed 2016. https://www.apa.org/about/policy/facilitated-communication

[9] Yamo Pharmaceuticals. Yamo Pharma Presents Statistically Significant Phase 2 Autism Results for L1-79 at INSAR 2025. GlobeNewswire. May 8, 2025. https://www.globenewswire.com/news-release/2025/05/08/3077235/0/en/Yamo-Pharma-Presents-Statistically-Significant-Phase-2-Autism-Results-for-L1-79-at-INSAR-2025.html