Cognitive blocks are what give the brain its advantage over AI

Cognitive blocks are what give the brain its advantage over AI

Artificial intelligence may write award-winning papers and diagnose diseases with remarkable accuracy, but biological brains still have the upper hand in at least one crucial area: flexibility.

For example, humans can adapt quickly and relatively easily to new information and unfamiliar challenges—by learning new computer software, following a recipe, or learning a new game—while AI systems have difficulty learning on the fly.

In a new study, Princeton neuroscientists discover one reason the brain has an advantage over AI: It reuses the same cognitive "blocks" across many different tasks. By combining and recombining these blocks, the brain can quickly develop new behaviors.

State-of-the-art AI models can achieve human or even superhuman performance in individual tasks. But they find it difficult to learn and perform many different tasks. We found that the brain is flexible because it can reuse components of cognition across many different tasks. By putting these “cognitive Lego blocks” together, the brain is able to develop new tasks.”

Tim Buschman, Ph.D., senior author of the study and associate director of the Princeton Neuroscience Institute

The results were published November 26 in the journal Nature.

Reuse skills for new challenges

If someone knows how to repair a bicycle, perhaps repairing a motorcycle comes more naturally. This ability to learn something new by reusing simpler skills from related tasks is what scientists call compositionality.

“If you already know how to bake bread, you can use that skill to bake a cake without having to learn baking from scratch,” said Sina Tafazoli, Ph.D., a postdoctoral researcher in the Buschman lab at Princeton and lead author of the new study. “You take existing skills—operating an oven, measuring ingredients, kneading dough—and combine them with new skills, like beating dough and making frosting, to create something completely different.”

However, there is limited and sometimes contradictory evidence on how the brain achieves such cognitive flexibility.

To clarify how the brain achieves its ingenuity, Tafazoli trained two male rhesus monkeys to perform three related tasks while their brain activity was monitored.

Instead of baking bread or repairing bicycles, the monkeys completed three categorization tasks. Similar to trying to decipher the often ambiguous spelling of a handwritten doctor's note, the monkeys had to judge whether a colorful, balloon-like blob on a screen in front of them looked more like a rabbit or the letter "T" (shape categorization), or whether it was more red or green (color categorization).

The task was deceptively difficult: the blobs varied in their ambiguity, sometimes obviously resembling a rabbit or a deep red, while other times the differences were subtle.

To indicate what shape or color they thought the blob was, a monkey hummed in response by looking in one of four different directions. In one task, looking to the left meant the animal saw a bunny, while looking to the right suggested it looked more like a “T.”

A key feature of the design was that while each task was unique, they also shared certain elements with the other tasks.

One of the color tasks and the shape task required looking in the same directions, while both color tasks required the animal to categorize color in the same way (either more red or more green) but had to look in different directions to judge its hue.

This experimental design allowed researchers to test whether the brain reuses neural patterns—its cognitive building blocks—during tasks with common components.

Blocks build cognitive flexibility

After analyzing activity patterns across the brain, Tafazoli and Buschman found that the prefrontal cortex — a region at the front of the brain involved in higher cognition — contained multiple common, reusable patterns of activity across neurons that worked toward a common goal, such as color discrimination.

Buschman described these as the brain’s “cognitive Legos” – building blocks that can be flexibly combined to create new behaviors.

“I think of a cognitive block like a function in a computer program,” Buschman said. "A group of neurons could distinguish colors, and their output could be mapped to another function that triggers an action. This organization allows the brain to perform a task by executing each component of that task in turn."

To perform one of the color tasks, the animal assembled a block that calculated the color of the image with another block that moved the eyes in different directions. When switching tasks, such as switching from colors to shapes, the brain simply put the relevant blocks together to calculate the shape and make the same eye movements.

This block release was largely observed in the prefrontal cortex and not in other brain regions, suggesting that this type of compositionality is a special property of this area.

Tafazoli and Buschman also found that the prefrontal cortex clears cognitive blocks when not in use, likely helping the brain better focus on the relevant task.

“The brain has a limited capacity for cognitive control,” Tafazoli said. "You need to condense some of your skills so you can focus on the ones that are important right now. For example, focusing on categorizing shapes temporarily reduces the ability to encode colors because the goal is shape discrimination, not color."

A more efficient way to learn – for AI and for the clinic

These cognitive Lego blocks could explain why people learn new tasks so quickly. By drawing on existing mental components, the brain minimizes redundant learning - a trick that AI systems have yet to master.

“A big problem in machine learning is catastrophic failure,” Tafazoli said. "When a machine or neural network learns something new, it forgets and overwrites previous memories. If an artificial neural network knows how to bake a cake but then learns to bake cookies, it will forget how to bake a cake."

In the future, integrating compositionality into AI could help create systems that continually learn new skills without forgetting old ones.

The same finding could also help improve medicine for people with neurological and psychiatric disorders. Conditions such as schizophrenia, obsessive-compulsive disorder, and certain brain injuries often impair a person's ability to apply known skills in new contexts—possibly due to disruptions in the recombination of the brain's cognitive building blocks.

“Imagine being able to help people regain the ability to change strategies, learn new routines or adapt to change,” Tafazoli said. “In the long term, understanding how the brain reuses and recombines knowledge could help us develop therapies that restore this process.”

Sources: