A groundbreaking study by researchers from Carnegie Mellon University and the University of Pittsburgh has confirmed the presence of “one-way paths” in brain activity, providing empirical evidence for long-hypothesized neural network principles. These stereotyped sequences of neural population activity, crucial for brain functions like motor control, memory, and decision-making, were explored through a clever experiment using a brain-computer interface (BCI). The findings, recently published in Nature Neuroscience, offer significant insights into computational neuroscience and practical applications for brain injury recovery.
The research team, affiliated with the Center for the Neural Basis of Cognition (CNBC), challenged nonhuman subjects to reverse natural neural activity sequences in the motor cortex. Even with visual feedback and rewards, subjects were unable to override these sequences, supporting the theory that neural activity is constrained by the brain’s circuitry. “This outcome validates principles that researchers have brought out in neural network models for decades,” said Byron Yu, professor of biomedical engineering and electrical and computer engineering at CMU.
The study was made possible by an interdisciplinary collaboration. Alan Degenhart, a former postdoctoral researcher at CMU and Pitt, noted, “We hypothesized that if these sequences were truly constrained by neural circuitry, it would be extremely difficult for subjects to modify them—and that’s exactly what we found.” Emily Oby, a former research professor at Pitt and now assistant professor at Queen’s University, highlighted the growing synergy between neural network modeling and understanding the brain: “Our findings are relevant not only for computational neuroscience but also for BCIs, stroke recovery, and learning.”
The project was co-led by Aaron Batista, professor of bioengineering at Pitt, who emphasized the unique collaboration between computational and experimental neuroscientists. “A team like ours, combining state-of-the-art expertise from two disciplines, really makes transformative work possible,” Batista said. The research team included postdoctoral fellow Asma Motiwala, and former Ph.D. students Nicole McClain and Patrick Marino, with key contributions from Erinn Grigsby and Douglas Weber.
Understanding these stereotyped sequences has the potential to reshape rehabilitation for patients recovering from brain injuries or disorders. “If we know how constrained activity sequences are, we can design optimized learning strategies to help patients regain motor control,” said Erinn Grigsby, a former Pitt Ph.D. student.
This work exemplifies the power of collaboration through the CNBC, a partnership between CMU and Pitt. By combining the strengths of both institutions, the team is now pursuing BCI-driven projects to further connect stereotyped activity sequences with physical movements, aiming to improve our understanding of motor planning and execution.
Interested? Read more about this story from the CMU and Pitt news rooms.