Scientists Uncover Brain Mechanism for Intuition Flashes

NYU Langone Health researchers pinpoint high-level visual cortex as the source of 'one-shot learning' insights.

Published on Feb. 5, 2026

Got story updates? Submit your updates here. ›

A new study led by researchers at NYU Langone Health has identified the brain region called the high-level visual cortex (HLVC) as the place where 'priors' - images seen in the past and stored - are accessed to enable one-shot perceptual learning, a mysterious type of one-shot learning that dramatically alters our ability to recognize something after seeing it just once. The researchers combined fMRI, EEG, and machine learning to locate priors in the HLVC and reveal the brain computations involved in this process.

Why it matters

Understanding the brain mechanisms behind one-shot learning, or 'aha moments' that change how we perceive the world, could shed light on abnormal perceptual experiences like hallucinations seen in conditions like schizophrenia and Parkinson's disease. The findings also have implications for developing AI models with human-like one-shot learning capabilities.

The details

The researchers explored changes in brain activity when people are shown blurred 'Mooney' images and then clear versions, a process that forces them to use stored priors to recognize the objects. Using fMRI, EEG, and machine learning, the team found that the HLVC is where these priors are accessed and used to enable one-shot perceptual learning. They also built an AI model with a prior storage module that achieved human-like one-shot learning, better than other leading AI models.

  • The study was published online on February 4, 2026 in Nature Communications.
  • The researchers' previous 2018 study on this process showed that after seeing a clear version of a blurred image, subjects became twice as good at recognizing it.

The players

Biyu He

Co-senior study author, associate professor in the departments of Neurology, Neuroscience, and Radiology at NYU Grossman School of Medicine.

Eric Oermann

Co-senior author, assistant professor in the Departments of Neurosurgery and Radiology at NYU Langone.

Ayaka Hachisuka

First author, from the NYU Langone Institute for Translational Neuroscience.

Jonathan Shor

First author, from the NYU Langone Institute for Translational Neuroscience.

Xujin Chris Liu

First author, from the NYU Tandon School of Engineering.

Got photos? Submit your photos here. ›

What they’re saying

“Our work revealed, not just where priors are stored, but also the brain computations involved.”

— Biyu He, Associate Professor (Mirage News)

“This study yielded a directly testable theory on how priors act up during hallucinations, and we are now investigating the related brain mechanisms in patients with neurological disorders to reveal what goes wrong.”

— Biyu He, Associate Professor (Mirage News)

“Although AI has made great progress in object recognition over the past decade, no tool has yet been capable of one-shot learning like humans. We now anticipate the development of AI models with human-like perceptual mechanisms that classify new objects or learn new tasks with few or no training examples. This is more evidence of a growing convergence between computational neuroscience and advances in AI.”

— Eric Oermann, Assistant Professor (Mirage News)

What’s next

The research team is also looking into likely connections between the brain mechanisms behind visual perception and the better-known type of 'aha moment' when we comprehend a new idea.

The takeaway

This study provides important insights into the brain mechanisms underlying the rapid 'one-shot learning' that allows us to recognize objects after seeing them just once. Understanding these processes could lead to new treatments for conditions involving abnormal perceptual experiences, as well as advances in artificial intelligence that mimic human-like learning.