Encoding & Decoding of Natural Language in fMRI
By Alexander Huth



The meaning of language is represented in highly specific patterns of brain activity across a large portion of the human cortex. Using recently developed machine learning methods and very large fMRI datasets collected from single subjects, we construct models that predict brain responses with high accuracy. Interrogating these models enables us to map language selectivity and potentially uncover organizing principles. The same techniques also enable us to construct surprisingly effective decoding models, which predict language stimuli from brain activations recorded using fMRI. These models can decode language while subjects imagine telling a story and while subjects watch silent films with no explicit language content.



Alexander Huth Biography



Hailing originally from southern California, Alex attended Caltech, where he earned a BS in computational neuroscience in 2007. There he began doing neuroscience research under Professor Christof Koch, studying how the brain processes sound in the congenitally blind. Alex stayed in Professor Koch’s lab for a year after graduation, and then moved north to UC Berkeley in 2008. Alex's PhD work at Berkeley with Professor Jack Gallant focused on using modern computational methods to understand and model how our brains extract meaning from both vision and language. After receiving his PhD in 2013, Alex stayed in Professor Gallant’s lab for 3 more years as a postdoc, during which time he published a landmark paper on how the brain processes language. This work received substantial attention from the popular press, and also led to Alex receiving prestigious research awards from the Burroughs Wellcome Fund and the Alfred P. Sloan Foundation. In 2017, Alex moved to UT Austin to begin a position as an assistant professor jointly in the departments of Computer Science and Neuroscience. His growing lab at UT aims to use incredibly large brain response datasets collected from single individuals in order to understand how the brain processes language at an unprecedented level of detail.

If you would like to attend, register in advance for this event.

After the webinar, you would be invited to talk to the speaker and mingle with other alumni in a post-webinar social via a Zoom meeting.

If you would like one of our alumni or professors to share with us his/her work/interests during a future seminar, please let us know.

We look forward to seeing you.