US lab turns to AI and virtual reality to probe the biology of human memory
Researchers at the Marine Biological Laboratory are combining high-performance computing and immersive visualisation to map molecular structures linked to how the brain stores and predicts information
The Marine Biological Laboratory in Woods Hole, Massachusetts, is using artificial intelligence, high-performance workstations and virtual reality to investigate the molecular foundations of human memory.
In a statement, the lab said researchers led by Andre Fenton, a professor of neural science at New York University, and Abhishek Kumar, an assistant professor of cell and regenerative biology at the University of Wisconsin–Madison, have brought together advanced computing and immersive tools to analyse vast three-dimensional datasets from the brain. The work is supported by the National Institute of Mental Health and the Chan Zuckerberg Initiative.
The project focuses on the hippocampus, a brain region central to learning and memory. The team is studying protein markers that trace dendritic structures around a micrometre long, which account for only about 1% of markers in the area but are thought to play an outsized role in how memories are formed and organised.
To handle the scale and complexity of the data, the researchers integrated NVIDIA RTX graphics processing units, HP Z workstations and the syGlass virtual reality platform. Using this setup, the team captured roughly 10 terabytes of volumetric data and carried out detailed visual inspections at human resolution.
“My life’s work is to understand how minds operate, and especially to understand memory; not merely as a trace of the past in the brain but as an estimate of the future that the brain is afforded,” Fenton said.
The workflow was turned into an interactive experience using syGlass running on an HP Z6 desktop workstation equipped with multiple NVIDIA RTX GPUs. Researchers can step inside the data using VR headsets, exploring complex neural structures in three dimensions rather than relying on flat images or automated analysis alone.
That approach also opened the door to an unusual pilot programme. Three high-school interns were brought into the lab and trained to use the VR system to identify and label memory-related protein markers across billions of neurons. According to the lab, the immersive environment made it possible for non-experts to contribute meaningfully to the research.
“If we can understand how something is built, then if there’s a problem, we can dissect that and get to the bottom of it,” Kumar said.
The team is now examining how proteins become mislocalised in the hippocampus, a process believed to be linked to memory disorders. The lab said it is also exploring ways to expand the high-school research programme beyond the initial group of interns, using the combination of AI and virtual reality to broaden access to advanced neuroscience research.
The Recap
- MBL uses AI, VR and workstations to study memory.
- Captured 10 terabytes of high-resolution 3D image data.
- Team plans to expand high-school participation after pilot.
