Fedorenko has developed robust and novel techniques that allow her to identify brain regions involved in linguistic processing at the individual-brain level. She has shown that the language network is uniquely engaged in linguistic, and not non-linguistic, tasks. She has shown that word meanings and syntax (the rules for how individual words can combine to create phrases and sentences), engage similar brain regions and are not readily dissociable. This contrasts with previous ideas in the field, and holds true even when using high-resolution methods such as intracranial recording. Fedorenko has also shown that that semantic composition may be the core driver of the language-selective brain regions.
Currently she is focused on a number of avenues:
(a) probing the time-course, effective connectivity, and causal mechanisms of language processing using intracranial recordings and stimulation.
(b) using state-of-the art decoding and deep neural nets to develop a precise model describing language regions.
(c) examining non-literal processing given that much of language comprehension goes beyond the literal string of words.
(d) relating variability in neural language markers between people to behavior and genetics.
(e) probing the cognitive and neural architecture of individuals with exceptional linguistic talent (e.g., “hyper-polyglots”).
Fedorenko joined the McGovern Institute and MIT in July 2019, having established her lab at Massachusetts General Hospital/ Harvard Medical School in 2014. She earned her PhD in Cognitive Sciences at MIT in 2007, where she also conducted her postdoctoral research.