Tid: Torsdag 19 september 2013, kl. 15:00 - 17:00
Plats: C307, Södra huset, Frescati

Postseminarium äger rum direkt efter seminariet i institutionens pentry.


One of the questions that have intrigued language researchers since the foundation of the cognitive sciences is how we are able to understand each other despite the striking 'lack of invariance' in the linguistics signal: multiple instances of the same linguistic category (e.g., a /b/) in the same context, spoken by the same speaker will differ in their physical properties. Furthermore, one speaker's /b/, might physically indistinguishable from another speaker's /p/. Similar variability exists at all levels of linguistic processing, including speaker’s lexical, syntactic, and pragmatic preferences. How then, do we manage to understand each?

I present an overview of efforts in my lab to understand how the human brain/mind achieves this goal. I’ll focus on behaviorial evidence from speech perception and sentence processing experiments and computational modeling of adaptation data. Our studies suggest that comprehenders are able to rapidly adapt their expectations about linguistic distributions (incl. phonetic categories, but also more abstract categories, such as syntax) based on recent language input. In doing so, they are able to converge towards the statistics of the current linguistics environment. This ability is crucial, as it allows efficient and robust language understanding despite the noise and variability of linguistic signal.
Time permitting, I will also present an ongoing study that links phonetic adaptation in adults to implicit learning in infants during the acquisition of phonology.

Some key references:
- Fine, A. B. & Jaeger, T. F. 2013. Evidence for implicit learning in syntactic comprehension. Cognitive Science 37(3), 578–591
- Fine, A. B., Jaeger, T. F., Qian, T., & Farmer, T. submitted. Rapid Expectation Adaptation During Syntactic Comprehension. Cognition.
- Jaeger, T. F. & Snider, N. 2013. Alignment as a consequence of expectation adaptation: syntactic priming is affected by the prime's prediction error given both prior and recent experience. Cognition 127(1), 57–83.
- Jaeger, T. F. & Ferreira, V. in press. Seeking predictions from a predictive framework. Brain and Behavioral Sciences.
- Kleinschmidt, D., & Jaeger, T. F. 2012. A continuum of phonetic adaptation: Evaluating an incremental belief-updating model of recalibration and selective adaptation. CogSci12, 605-610. Austin, TX: CSS.
- Qian, T., Jaeger, T. F., & Aslin, R. 2012. Learning to Represent a Multi-Context Environment: More than Detecting Changes. Frontiers in Psychology 3, 228.

T. Florian Jaeger
Brain and Cognitive Sciences, Computer Science, University of Rochester

[This work is based on collaborations with Richard Aslin, Thomas Farmer, Alex Fine, Robbie Jacobs, Dave Kleinschmidt, Cheyenne Munson, Ting Qian, and Joe Toscano, and funded by an Alfred P. Sloan Fellowship, NSF BCS-0845059, NSF CAREER IIS-1150028, and NIH R01 HD075797-01]


Thomas Hörberg & Ljuba Veselinova