York University AI Study: Hidden Flaws in Brain-Like Models | AcademicJobs
A York University study introduces reverse predictivity, exposing flaws in brain-mimicking AI and implications for neuroscience and higher education in Canada.
No reviews yet. Be the first to rate Kohitij!
Kohitij Kar is an Assistant Professor in the Department of Biology within the Faculty of Science at York University. He holds the position of Canada Research Chair in Visual Neuroscience. Kar joined York University in 2022. Prior to this appointment, he served as a Research Scientist and Postdoctoral Associate in the DiCarlo Lab at the McGovern Institute for Brain Research at the Massachusetts Institute of Technology. He completed his Ph.D. in 2015 in the Department of Behavioral and Neural Sciences at Rutgers University under the supervision of Bart Krekelberg.
Kar’s research centers on visual neuroscience, with emphases on object recognition, visual electrophysiology, and the development of brain-like artificial intelligence models. He is a member of York University’s Centre for Vision Research and Centre for Integrative and Applied Neuroscience. In 2025, he received the Ontario Early Researcher Award in recognition of his contributions to early-career research. Kar has authored or co-authored numerous peer-reviewed publications on topics including neural representations in the primate ventral visual stream and comparisons between artificial neural networks and biological vision systems. He has contributed to public lectures and seminars on recurrent computations in visual object perception and the integration of neuroscience with artificial intelligence approaches.
A York University study introduces reverse predictivity, exposing flaws in brain-mimicking AI and implications for neuroscience and higher education in Canada.