Neurolinguistic Job Talk - Philip J. Monahan "From Variation to Representation and Back"
Fri, February 24, 2012 • 12:00 PM - 1:30 PM • UTC 1.132
We comprehend spoken language with little effort, despite significant variation inherent in the speech signal. The mechanisms by which the brain accomplishes this task, however, remain poorly understood. In this talk, I argue that the internal knowledge of our phonological system is central in imposing structure on this highly variable signal. I support this position via a combination of cognitive neuroscience and formal linguistic theory. First, I present evidence for a specific auditory scaling mechanism with neurobiological plausibility that helps us arrive at our structured linguistic category representations from the noisy signal. Additionally, it appears that the relative density of the phonetic space to which the signal is being mapped determines the sensitivity of this algorithm. Second, with behavioral responses and real-time brain measures, I demonstrate that some aspects of a speech sound are not represented in their long-term representations. Importantly, these representations provide the basis for which we interpret the signal and construct expectations about it. Finally, I provide an example in which the brain exploits non-contrastive acoustic variation to construct high-level, i.e., dialectal, categorizations based on limited speech input. This work has important implications for understanding how we perceive discrete phonetic categories from continuous physical quantities and the underlying neural computations.
Philip J. Monahan of Basque Center on Cognition, Brain and Language (BCBL), Donostia-San Sebastián, Spain