Facebook pixel
Go to home page

New BrainHealth Faculty Member Aims to Tackle Neural Mechanisms of Addiction and Mental Health

Xiaosi Gu, PhD, researches cognitive control and decision-making, with a particular focus on abnormal cognitive processes in addiction.

Center for BrainHealth

Nicotine, the primary addictive substance in tobacco, stimulates neural pathways in the reward circuitry of the brain. However, pure biochemical explanations are not sufficient to account for difficulty in quitting and remaining smoke-free. Xiaosi Gu, Ph.D. recently joined the Center for BrainHealth at The University of Texas at Dallas as an assistant professor to further research into cognitive control and decision-making, with a particular focus on abnormal cognitive processes in addiction. Dr. Gu’s most recent work, published in the Proceedings of the National Academy of Sciences (PNAS), suggested that belief is as important as biochemistry in addiction. “In essence, what the study showed was the power of the cognitive system to override the effects of neuroactive drugs,” said Gu. “This evidence implies that what an individual thinks about the act of engaging with a drug and its subsequent effect on the brain and body has major implications in how the brain responds to the drug.” Gu’s research will focus on poor decision-making and the loss of control, often considered hallmarks of addiction as well as many other psychiatric conditions, by using functional MRI in combination with neuroeconomic tasks to measure neural and behavioral responses. “My ultimate dream is to use neurobiological information to inform individualized therapy,” Gu explained. “Compared to physical diseases, mental disorders have much more heterogeneity and complexity. If you have heart disease or are diagnosed with lung cancer, physicians will follow a treatment plan based on biology. For psychiatric disorders, we don’t have anything like that yet. Diagnoses are made based on the Diagnostic and Statistical Manual of Mental Disorders (DSM), which is a subjective analysis, not an objective one. That gap in knowledge is what I am hoping to fill through my research – to use much more formal objective assessments based on deep cognitive and neural phenotypes to help with accurate diagnosis.” With a keen interest in computational approaches to psychiatry, Gu received her Ph.D. in neuroscience from the Mount Sinai School of Medicine in New York City and spent the last four years as a research fellow at the Virginia Tech Research Institute in Roanoke, Virginia and the Wellcome Trust Centre for Neuroimaging at University College London in London, UK. During her time in London, Gu and colleagues set up the world’s first course on computational psychiatry, which aims to bring together experts in neuroscience, psychiatry, decision sciences and computational modeling to define problems quantitatively in psychiatric disorders, and to train the next generation of scientists and clinicians that wish to apply these models to modern diagnosis and treatment strategies. “We are very pleased to have Dr. Gu join our faculty,” said Dr. Bert Moore, dean of UT Dallas’ School of Behavioral and Brain Sciences and Aage and Margareta Moller Distinguished Professor. “She brings a varied array of interests and research efforts investigating such diverse domains as the underpinnings of empathy and brain mechanisms involved in addiction. Utilizing assorted methodologies, coupled with sophisticated computational analyses, Xiaosi adds important strengths to the School and Center and also opportunities for collaborations and student training.” What is computational psychiatry? Computational psychiatry is a new interdisciplinary field which seeks to characterize mental disorders in terms of aberrant computations at multiple scales. In recent years the field of human neuroscience, particularly functional neuroimaging, has begun to address the underlying neurobiology of changes in brain function related to psychiatric disease. This effort has produced some exciting early discoveries, but it has also highlighted the need for computational models that can bridge the explanatory gap between pathophysiology and psychopathology. The expertise and quantitative tools required to address this gap exist only across disciplines, combining skills and knowledge from investigators and clinicians that are jointly interested in solving problems of mental health.

Share this article