Speech in contextProject reference: 623072
Funded under :
The neural implementation of contextual influences in speech perception
Total cost:EUR 259 293,3
EU contribution:EUR 259 293,3
Topic(s):FP7-PEOPLE-2013-IOF - Marie Curie Action: "International Outgoing Fellowships for Career Development"
Call for proposal:FP7-PEOPLE-2013-IOFSee other projects for this call
Funding scheme:MC-IOF - International Outgoing Fellowships (IOF)
Human listeners can perceive speech quite effortlessly, although it is, computationally, a very challenging task. An important reason why humans perform so well is because they make use of acoustic information that is available in the context. Such contextual information is useful because it provides a listener with information about a speaker's voice properties and his or her speaking rate. Although the influence of context on speech perception was noted decades ago and has been extensively investigated behaviourally, we know little about the underlying neural mechanisms, and the functional levels at which such influences operate remain hotly debated. Recently, the development of a technique that is largely novel to the language sciences, intracranial recordings (electrocorticography; ECoG), has started to provide important insights into the mechanisms underlying human speech perception. In the proposed project I will use this method to unravel the neural mechanisms by which human listeners manage to make use of contextual information in speech perception, focussing on three known perceptual normalization mechanisms. For each of these, I will determine whether they operate at a pre-categorical or at a categorical level of processing. In addition, the project will directly address the role of the posterior Superior Temporal Gyrus as a possible mapper of spectral representations onto speech sound categories. For the implementation of the project, I will spend 2 years at the University of California at Berkeley, USA, to gain experience with the highly specialized analysis techniques necessary for the analysis of ECoG measures, and work at the Donders Institute at the Radboud University Nijmegen, The Netherlands, during the third year. The proposed research programme bridges between the fields of linguistics and auditory neuroscience.
EU contribution: EUR 259 293,3
GEERT GROOTEPLEIN NOORD 9
6525 EZ NIJMEGEN
Tel.: +31 24 3610751
Fax: +31 24 3610652