First evidence of mirror-like activity found in individual human neurons

In a groundbreaking study, Caltech researchers have uncovered the first evidence of mirror-like activity in individual human neurons. This discovery challenges previous understanding of how we comprehend the actions of others and our ability to imagine actions we have never performed ourselves.

Mirror neurons, initially discovered in nonhuman primates in the 1990s, are specialized brain cells that fire both when an individual performs an action and when they observe someone else performing the same action. These neurons were thought to facilitate understanding of others' actions by internally simulating them. However, the new study by Caltech researchers suggests that mirroring is more nuanced and context-dependent than previously believed.

The findings, published in the journal Cell, indicate that mirror activity occurs only in a higher-level brain region called the posterior parietal cortex (PPC), which is involved in planning and intention. Crucially, mirror activity does not happen automatically and can be suppressed depending on the task at hand.

The research, led by postdoctoral scholar Vasiliki Bougou, involved working with two tetraplegic individuals. By implanting electrode arrays in their PPCs and motor cortex (MC), the researchers measured neural activity during tasks involving observing and performing actions such as sliding, lifting, and rotating a cube. The PPC neurons showed mirror-like properties, with similar patterns of neural activity when participants observed or attempted the actions themselves. However, the MC neurons only encoded the attempted actions and did not show mirror activity.

The study reveals that the brain can selectively encode and ignore visual information based on its relevance to the task at hand. When participants performed tasks and were asked to report what they had observed, neurons in the PPC encoded both the attempted and observed actions. In contrast, the MC only encoded the attempted action. This dissociation between action and observation suggests that our brains build internal models of others' actions in the PPC, but this process is not automatic.

The findings have significant implications for the development of brain-machine interfaces (BMIs). By understanding how the brain constructs mental models of observed actions, researchers can improve the flexibility and adaptability of BMIs. For example, a BMI could potentially allow a tetraplegic individual to drive a car using their thoughts alone, without interference from observing actions in their visual field.

The researchers are exploring how the PPC builds internal models to enhance BMIs and improve the overall user experience.