Brain Memory: Star Cells’ Role Revealed
- Challenging decades of scientific thought,new research indicates that astrocytes,a type of brain cell often overshadowed by neurons,could be central to memory storage.
- astrocytes, characterized by their star-like shape, have long been recognized for their maintenance roles, such as clearing debris and supplying nutrients to neurons.
- "you can imagine an astrocyte as an octopus with millions of tentacles," said Leo Kozachkov, lead author and postdoctoral fellow at IBM Research.He explained that these "tentacles" or...
Astrocytes, not just neurons, may be key to memory storage, according to groundbreaking research. A new study suggests these star-shaped brain cells, which communicate through calcium signaling, could explain the brain’s remarkable capacity. Researchers propose a novel model where astrocytes, with their unique structure, facilitate vast facts storage by connecting to thousands of synapses. This challenges long-held beliefs about the brain’s mechanisms, opening new avenues for treating memory disorders adn potentially impacting artificial intelligence. The model introduces a new outlook on memory formation, suggesting that memory capacity scales with astrocyte-synapse interactions. This understanding could reveal therapeutic targets for neurodegenerative diseases,such as Alzheimer’s. News Directory 3 is following the story. Discover what’s next for this revolutionary study.
Astrocytes May Play Key Role in Brain’s Memory Storage
Updated June 9,2025
Challenging decades of scientific thought,new research indicates that astrocytes,a type of brain cell often overshadowed by neurons,could be central to memory storage. The study, published in PNAS, introduces a novel network architecture suggesting astrocytes are responsible for the brain’s impressive memory capacity.
astrocytes, characterized by their star-like shape, have long been recognized for their maintenance roles, such as clearing debris and supplying nutrients to neurons. They also feature processes that form tripartite synapses, involving two neurons and the astrocyte.
“you can imagine an astrocyte as an octopus with millions of tentacles,” said Leo Kozachkov, lead author and postdoctoral fellow at IBM Research.He explained that these “tentacles” or processes wrap around nearby synapses.
Unlike neurons, astrocytes communicate through calcium signaling. They respond to synaptic activity by altering their internal calcium levels, which can trigger the release of chemical messengers into the synapse.
Kozachkov described these processes as “tiny calcium computers,” sensing and passing information, then receiving feedback to adjust neuronal activity. However, the precise computational functions of astrocytes remain under investigation.
To understand this function, Kozachkov and colleagues employed machine learning architectures capable of representing complex interactions.
Dmitry Krotov, a research staff member at the MIT-IBM Watson AI Lab and IBM Research, noted that conventional networks linking only pairs of neurons might limit information encoding. The team hypothesized that astrocytes, connecting to thousands of synapses, could mediate communication across these connections, explaining the brain’s storage capabilities.
Kozachkov said the unique structure of astrocytes provides a natural way to design large information storage systems.
The researchers also proposed that astrocytes store memories through gradual changes in calcium patterns, translated into signals for neurons. In this model, each astrocyte process functions as a distinct computational unit.
Kozachkov stated that their model doesn’t require many neurons to store numerous memories, offering an energy-efficient advantage.
Maurio de Pittà, an assistant professor at the Krembil Research Institute, who was not involved in the study, called the model a “biologically grounded clarification” for memory storage. He noted that studies with high-resolution microscopes support the view of astrocyte processes interwoven throughout the brain.
De Pittà cautioned that models are approximations and that current technologies cannot fully capture real-time brain dynamics needed to validate the hypothesis.
While the role of astrocytes in memory formation is emerging, de Pittà said there is no clear proof that calcium-based interactions help create, store, or recall memories. If the model is correct,it could offer a new outlook on brain storage,suggesting memory capacity scales with astrocyte-synapse interactions.
The study authors suggest the model offers potential therapeutic targets for neurodegenerative diseases.
Kozachkov said that since astrocytes are implicated in Alzheimer’s and other memory disorders, their model provides a computational view of potential issues. He added that the model may inspire the search for new therapeutic targets,such as modulating astrocyte process connectivity or signaling to restore lost memory function.
However, more research is needed to translate this work into clinical treatments.
beyond neuroscience, the model may have applications in artificial intelligence. De Pittà suggested it could help create brain-like hardware systems for efficient information storage and recall, possibly benefiting voice recognition, robotics, AI assistants, and brain-machine interfaces.
What’s next
Future research will focus on validating the astrocyte memory model through real-time observation of brain activity and exploring potential therapeutic applications for neurodegenerative diseases.