Center of Excellence for Learning in Education, Science and Technology A National Science Foundation Science of Learning Center

Karthik Srinivasan

Karthik SrinivasanKarthik Srinivasan did his B.Tech in Information Technology from Anna University, Chennai, India from 2002-2006. As an undergraduate, he got interested in neurobiology and decided to switch from engineering to neuroscience. He did his thesis under Venkateswaran Nagarajan on coding in visual pathways, models of large scale spiking neural networks and high performance computing architectures for the same. This led him to work at the Blue Brain Project, EPFL, Switzerland on a 6 month fellowship as a Visiting Research Student in 2007. He was responsible for writing a software, the Blue Repair, crucial to the project’s toolchain. The system automates morphometric analysis, 3D visualization and repair of single neuron morphologies from Neurolucida drawings. Karthik joined the Department of Cognitive and Neural Systems at Boston University in Fall 2007 convinced that a mathematical and computational theory is essential for understanding neurobiology and behavior. The manifold interactions, interdisciplinary efforts and the intensive coursework mediated under the CELEST umbrella have provided a great opportunity to further his interests.

He is currently advised by Stephen Grossberg and Arash Yazdanbakhsh. They are interested in providing a neural and computational framework for understanding how invariant 3D visual perception is enabled by cortical circuits. They are currently working on how fast binocular fusion, during saccadic eye movements, helps to provide an invariant representation of surfaces and objects despite changes in their retinotopic locations? This effort would help in clarifying the computations performed by the underlying neural circuits which enable such transformations. In the visual attention and spatial perception literature, this is referred to as the predictive remapping problem. In their earlier work, published in Neural Networks and presented at SfN and VSS meeting, they provided a mechanistic basis of how the cortical circuits from visual areas V1 to V2 compute relative disparity from absolute disparity. Combining these two approaches, they are working towards how surfaces hidden in Magic Eye autostereograms can be perceived.

Though Karthik works primarily in vision, his interests extend to experimental techniques and theories to understand sensory processing, neural representation, memory and in particular the development, formation and functions of cortical maps. He has been actively participating in several CELEST related activities. These include inviting and hosting speakers for the SLC seminar series, volunteering for the ICCNS conference, attending the iSLC student and postdoctoral conference and organizing student workshops and tutorials on image processing and computer vision.

Last updated on May 18, 2011