New research from Princeton University sheds light on a essential difference between human and artificial intelligence: the brain’s remarkable ability to learn and adapt without starting from zero.Unlike AI, which often requires exhaustive retraining for each new skill, the brain appears to repurpose existing cognitive structures-what researchers are calling “cognitive LEGOs”-to master new tasks. The findings, published this week, offer potential breakthroughs in both AI development and the treatment of neurological conditions, building on decades of research into the brain’s estimated 86 billion neurons [[2]] and its complex functions [[1]].
The human brain learns new skills by repurposing pre-existing cognitive building blocks, a new study suggests, offering a potential key to understanding why people can acquire abilities far more quickly and flexibly than even the most advanced artificial intelligence.
Researchers at Princeton University discovered that, unlike AI systems that require extensive training for each new task, the brain doesn’t start from scratch. Instead, it utilizes what researchers describe as “cognitive LEGOs” – pre-built cognitive units that are recombined to produce new behaviors.
This discovery has implications for fields ranging from artificial intelligence development to the treatment of neurological disorders, and could offer insights into how we learn and adapt throughout our lives.
Brain’s Flexibility Outpaces AI
While artificial intelligence excels at tasks like writing articles and assisting doctors with diagnoses, machines often struggle to rapidly learn new skills or immediately adapt to unfamiliar environments.
The human brain, however, demonstrates a remarkable ability to:
Learn the rules of a new game within minutes.
Apply previously learned skills to facilitate the acquisition of new ones.
Seamlessly switch between different tasks.
This capability stems from a key cognitive property called compositionality – the ability to build new skills from smaller, pre-existing skills.
How Researchers Tested the Concept
The research team trained two rhesus macaque monkeys on three different visual tasks. The monkeys were tasked with categorizing colored shapes on a screen, either by shape (rabbit/T-shape) or by color (red/green), using eye movements to indicate their answers.
The tasks were designed to:
Share some underlying rules.
Differ in others.
This allowed researchers to investigate whether the brain reused the same neural patterns when tasks shared common cognitive elements.
The Prefrontal Cortex: A Repository of Cognitive Building Blocks
Neural recordings revealed that the prefrontal cortex – the brain region responsible for complex thinking – maintains recurring patterns of neural activity across different tasks whenever a shared skill is required, such as:
Color discrimination.
Shape recognition.
Eye movement control.
These patterns represent the “cognitive LEGOs” that are repeatedly reused.
How Does It Work?
When the brain transitions from one task to another:
It activates the “cognitive LEGO” specific to the new task.
It reuses the shared “LEGOs.”
It deactivates unnecessary “LEGOs” to maintain the composition.
“It’s that simple, and that ingenious,” researchers said.
Why AI Struggles to Replicate This Approach
Most AI systems suffer from:
Catastrophic interference – when learning a new task causes the AI to forget previously learned ones.
However, the brain, thanks to its “cognitive LEGOs,” can:
Learn new tasks without losing existing skills.
Medical and Research Significance of the Discovery
This new understanding could potentially contribute to:
The development of a new generation of AI capable of cumulative learning.
Improved treatment for disorders such as schizophrenia and obsessive-compulsive disorder.
Enhanced rehabilitation after brain injuries.
A better understanding of adaptation difficulties and task-switching challenges experienced by some patients.
These findings could pave the way for therapies that target how the brain assembles and reuses skills.
The research underscores that the brain isn’t merely a thinking machine, but a flexible and intelligent system that utilizes pre-built cognitive components, rearranging them to invent new behaviors. This unique ability currently sets humans apart from artificial intelligence.
