Have a UROP opening you would like to submit?
Please fill out the form.
Audio & Mechanical Turk Experiments to Understand Language Processing in the Brain
6: Electrical Engineering and Computer Science
Andrei Barbu: email@example.com
We want to understand how human brains process language. So far, we've had little data to do this and consequently much less is known about language in the brain than vision in the brain. This project has two phases: annotating a large audio corpus so that we know when a subject heard a word and using that corpus to probe different questions about how words and sentences were processed by their brains. We have a unique dataset recorded from electrodes placed in human brains while subjects were watching movies. Part of the UROP would involve annotating this dataset and helping to improve the tool, then learning to run Mechanical Turk experiments to extend the annotations, while you learn more about machine learning for neuroscience.
It depends on what you want to focus on. If the focus is on the tool, Mechanical Turk, and annotations no experience is strictly necessary but some knowledge of linguistics or web development would be a plus. If the focus is on the neural data some experience in machine learning is required with knowledge of linguistics or neuroscience being a plus.