How the Human Brain Outperforms Artificial Intelligence in Learning and Adaptation

In recent years, artificial intelligence has made tremendous progress, but the human brain still remains unmatched in its ability to learn quickly and apply knowledge to new situations. A recent study conducted at Princeton University sheds light on how exactly the brain manages this complex process. Instead of traditional human testing, researchers turned to rhesus macaques (Macaca mulatta), whose brain structure and cognitive abilities closely resemble those of humans.

During the experiments, the monkeys performed tasks on a screen, distinguishing between colors and shapes and choosing the direction of their gaze to respond. Meanwhile, the researchers recorded brain activity to determine which areas were engaged during different tasks. It turned out that the brain uses separate groups of neurons that can be compared to building blocks—a kind of ‘cognitive LEGO.’

Cognitive flexibility

These neural blocks are not tied to a single function. The brain can combine them in different ways, creating new circuits to tackle unfamiliar tasks. This approach gives rise to remarkable cognitive flexibility—something even the most advanced AI systems still lack. For example, if one group of neurons is responsible for distinguishing colors, it can be linked to another group that controls movement, allowing the brain to quickly master new challenges.

Researchers note that it is precisely this ability to ‘repackage’ cognitive elements that allows humans and animals to quickly adapt to change. Unlike neural networks, which often forget old skills when learning new ones, the brain retains and reuses already mastered patterns.

The Role of the Prefrontal Cortex

Scientists paid special attention to the prefrontal cortex—the area of the brain responsible for planning, decision-making, and tackling complex tasks. This is where they found the main ‘building blocks’ of cognitive processes. When certain blocks are not needed, their activity decreases, allowing the brain to focus on the current task and avoid wasting resources.

This mechanism is similar to functions in computer programs: some neurons perform basic operations, and their results are passed along the chain. Thanks to this, the brain can quickly switch between tasks without losing efficiency.

An Advantage Over AI

Modern artificial neural networks can achieve impressive results in specific areas, but they struggle to master multiple different tasks at once. When AI is trained in new skills, it often experiences ‘catastrophic forgetting’—the system loses the ability to perform previous tasks.

In contrast, the brains of humans and animals are able to use prior experience to solve new problems. This enables them to respond quickly to changes in their environment, learn from mistakes, and apply knowledge in unexpected situations.

Opportunities for medicine

The discovery of the ‘cognitive Lego’ mechanism could be a breakthrough not only for the development of artificial intelligence, but also for medicine. Understanding how the brain switches between tasks and uses the same neural blocks will help create new treatments for neurological and mental disorders. This is especially relevant for patients who find it difficult to apply familiar skills in new circumstances.

Scientists believe that in the future, their work will help create more flexible and adaptive AI systems, as well as improve rehabilitation for people with cognitive impairments.

If you didn’t know, Princeton University is one of the oldest and most prestigious universities in the United States, renowned for its research in neuroscience and artificial intelligence. The university employs leading experts from around the world, and its laboratories regularly publish discoveries that are reshaping our understanding of the brain and the possibilities of modern technology.