Brown researchers are using computational approaches to better understand the workings of the brain and to enable new treatments for conditions from ADHD to paralysis.
Understanding the brain is fundamentally a big data problem. Within each human brain, billions of neurons connect with each other in trillions of ways, giving rise to thoughts, memories, emotions and all the rest of the human experience. Understanding how those connections work — and how they can go wrong — is one of the great frontiers of modern science.
Michael Frank is a professor of cognitive, linguistic and psychological sciences and the founding director of CCBS.
At Brown, researchers are using data science and computational modeling to discover fundamental aspects of brain function, which could lead to new treatment approaches for brain disorders. In 2020, Brown’s Carney Institute for Brain Science launched the Center for Computational Brain Science (CCBS). The center aims to harness expertise from neuroscience, engineering, mathematics, computer science and other areas to create new ways of understanding brain function, and to develop new therapies for disorder and disease.
Michael Frank, a professor of cognitive, linguistic and psychological sciences, is the founding director of CCBS. His research combines computational modeling with experimentation to reveal fundamental aspects of how people and other animals think, make decisions and learn.
In a 2020 study published in the journal Science, Frank and his colleagues showed the mechanism through which ADHD drugs like Ritalin operate in the brain to enhance cognitive output. Ritalin enhances release of the neurotransmitter dopamine in the striatum, a key region in the brain related to motivation, action and cognition. Using a behavioral economic strategy, the study found that the drugs direct the brain to fix its attention on the benefits, rather than the costs, of completing difficult tasks. The findings were captured by Frank’s mathematical model of how dopamine affects decision making and cognition, and could be helpful in developing new therapies for ADHD, as well as other dopamine-related disorders such as anxiety, depression and schizophrenia.
“ My hope is to facilitate a deeper understanding of the interconnections among multiple levels of computational brain science.
”
Michael Frank
Director of the Center for Computational Brain Science
In another study released in 2021 in the journal Cell, Frank collaborated with Christopher Moore’s lab in the Carney Institute to study more precise dynamics of dopamine in the mouse striatum. Contrary to dominant accounts in which dopamine goes up uniformly to rewarding events, the team discovered that dopamine propagates across the striatum in waves of activity, with the direction of the wave depending on whether the animal had “agency” in obtaining its goal. The findings motivated a new computational model in which dopamine dynamics allow animals to infer when they are in control over rewarding events, and to then assign “credit” to the corresponding brain regions that facilitate adaptive behavior.
Investigating Motivation, Movement Disorders and OCD
Another researcher associated with CCBS, Amitai Shenhav, uses computational modeling to investigate the brain’s role in motivating people to achieve goals. In particular, he is researching what makes some tasks more challenging than others, how people evaluate the costs and benefits of putting in the mental effort to overcome those challenges, and how long they persist in those efforts when there are tempting alternatives.
In early 2021, Shenhav published a study combining computational modeling and brain imaging techniques to reveal new details about the neural circuitry that underlies motivation. Those kinds of insights could be useful in helping people who have motivation problems due to chronic depression or other disorders.
Other researchers affiliated with CCBS are working directly on new therapeutic approaches to brain disorders. David Borton, an associate professor of engineering, is working on the development of next-generation deep-brain stimulation (DBS) technologies. DBS uses small implantable electrodes to deliver subtle electrical pulses directly to the brain. The technique is effective in treating movement disorders like Parkinson’s disease, but today’s devices have limitations.
Brain-Computer Interfaces: Researchers from Brown are pioneers in the development of brain-computer interfaces (BCIs) — devices that sense activity within the brain and use those signals to control external devices such as laptop computers and robotic arms. Recently, a Brown team led the demonstration of the first wireless, high-broadband BCI systems. The research is a critical step toward a fully implantable BCI system that could help people who have lost the ability to move or communicate due to injury or disease.
Currently, doctors adjust the stimulation level according to the patient’s symptom response. Once the level is set, the device stimulates at that level continuously until the doctor readjusts it, requiring patients to return to the clinic. But there’s interest in developing devices that can adjust stimulation automatically in response to real-time changes in symptoms or related biomarkers in the brain, which requires a system that can stimulate brain activity and sense it at the same time.
The problem is that stimulation can create electrical artifacts that corrupt the native brain signals the system is trying to sense. Working with Matthew Harrison, an associate professor of applied mathematics, Nicole Provenza, a Ph.D. student in biomedical engineering, and Evan Dastin-van Rijn, an undergraduate student in biomedical engineering, Borton and team developed an algorithm that successfully distinguishes stimulation artifacts and removes them, revealing the native brain signals. The work could be an important enabling step toward adaptive DBS systems.
In related work from the Borton lab, a 2021 study led by Provenza demonstrated, for the first time, the ability to sense candidate biomarkers in the human brain associated with symptoms of obsessive compulsive disorder (OCD). Technologies from commercial partner Medtronic enabled brain sensing over the course of years, detecting changes in the brain’s electrical activity associated with OCD as people went about their daily lives. The findings could be a critical step toward bringing adaptive deep-brain stimulation to bear in the treatment of OCD — a project Borton is working on with Frank and applied mathematician Matthew Harrison.
These are just a few examples of computational brain science work happening at Brown, and Frank is hopeful that CCBS will catalyze new collaborations harnessing the wealth of expertise across the disciplines at Brown. Making progress in understanding the brain in all its complexity will require diverse perspectives, and Brown’s collaborative research ethos makes it a perfect place for such dynamic research.
Brown’s expanding Department of Computer Science is fortifying its research and teaching excellence in data science, artificial intelligence and socially responsible computing.
Researchers in Brown’s Department of Biostatistics are gathering data from MRIs, CT scans and other medical imagery to improve diagnosis and clinical assessment for a range of medical conditions.
Brown researchers are using machine learning, artificial intelligence and other cutting-edge data techniques to piece together human genetic history and fight disease.
