Ramanujan Srinath is interested in how we see and interact with objects in the world. In his Ph.D., he studied how the brain quickly transforms image information from the retina to decodable information about the geometry of 3D objects. He is an experienced electrophysiologist, optophysiologist and programmer, and is interested in multifaceted projects that can answer fundamental questions about how we make inferences about and physically interact with our environment. He is currently training monkeys to do a version of the task described above and gearing up to start multi-channel, multi-area recordings in the brain. He is also building deep network models that can do the task as well as (or better than) humans.
Principal Investigator: Marlene Cohen
Fellow: Neha Murthy
“Supervised and unsupervised shifts in learned correlations between stimulus parameters in humans and monkeys”
Our brain is tasked with making sense of the ever-changing visual information about objects in our environments. In this project, we will study how two visual properties of objects can be learned together and how behavior would change if the learned associations between these properties changes. We are currently training monkeys to do a task where the monkey is reporting the curvature of an object on a continuous scale. We are also doing online human psychophysics experiments using the same task to assess how humans would solve the problem with and without explicit instructions about the shifts in associations. The undergraduate researcher will be involved in both aspects of this exciting project which would include hands-on training with custom software, human experiments and behavioral data analysis.