A post claiming that "Stanford scientists just successfully reversed autism symptoms" is spreading widely - and for once, there really is a Stanford study behind the headline. The catch: it was done in mice, using techniques that are not yet available for people. Here is what the researchers actually found.
What the study actually did
Researchers at Stanford Medicine, led by postdoctoral scholar Sung-Soo Jang and senior author John Huguenard, studied mice genetically engineered to lack Cntnap2, a gene strongly linked to autism in humans. These mice show classic autism-like traits: avoiding other mice, repetitive grooming, hyperactivity, heightened sensitivity to stimuli, and a higher susceptibility to seizures. Published in Science Advances in August 2025, the study traced these behaviors to overactive neurons in the reticular thalamic nucleus, a brain region that acts as a gatekeeper for sensory information passing between the thalamus and the cortex.
How the "reversal" worked
When the researchers calmed those overactive neurons - either with an experimental anti-seizure compound called Z944, or with a lab technique called DREADD that lets scientists genetically switch specific neurons on or off with a designer drug - the mice's hyperactivity, repetitive grooming, and social avoidance improved, and their seizure susceptibility dropped. That is a genuinely exciting result: it points to a specific brain circuit, not just a gene, that may drive several autism-associated behaviors at once.
Why "reversed autism" overstates it
Two things get lost when a headline says Stanford "reversed autism." First, this is a mouse model of one specific genetic cause of autism (Cntnap2 mutations); autism in humans involves many different genetic and environmental contributors, and this circuit may not be the driver in every case. Second, DREADD is a research tool that requires genetically modifying neurons in advance - it is not a treatment that could be given to a person today. Z944 is closer to real-world use since it is an actual drug candidate, but it has not been tested in autistic humans for this purpose, and mouse behavior is not the same thing as the lived experience of autism. The researchers themselves frame this as a new therapeutic pathway worth exploring, not a cure.
Why it is still worth paying attention to
Unlike some of the supplement and diet claims we have fact-checked, this one reflects real, peer-reviewed neuroscience with a plausible mechanism, published in a serious journal. It adds to a growing picture of autism as involving specific, identifiable brain circuits - which is genuinely useful for future research, even if a human treatment is likely years away, if it arrives at all.
For more on the genetics and brain biology researchers are studying in autism, see our articles on autism genes and brain pathways and AuDHD, or take our free Autism AQ-10 screening.
Sources: Jang, Huguenard et al., Stanford Medicine, published in Science Advances (August 2025); reporting via ScienceDaily. Educational summary - not a diagnosis or medical advice.