Principal Investigator: Jonathan King
The Temporal Dynamics of Language Processing
Ever since my earliest work on the effect of working memory capacity on sentence processing, (King & Just, 1991) 1 have been interested in individual differences in linguistic performance, and how these can depend on differences in effective working memory capacity. My interests also extend beyond the usual boundary of psycholinguistics in that they have been focused on the neural substrate of language processing. However, the neural substrate is much more than just a catalog of brain areas whose activity is modulated by the comprehension or production of language, and it is more than just that plus the information about connections between those areas. In an important sense, it is the temporal dynamics of information flow through the neural network that is the true substrate of language. Language processing can be tracked at time scales that range from mere milliseconds (e.g., King & Kutas, in press) to the integration of information across several seconds (e.g., King & Kutas, 1995; Kutas & King, 1996).
This emphasis on the temporal dimension of language processing, and the necessary connection of temporal dynamics to working memory, has forced me to recognize that any disruption in timing induced by our experimental designs must have an effect on the processing being studied. My response has been to work towards the development of more ecologically valid experimental paradigms, including those that utilize naturally presented speech (e.g., Mueller, King & Kutas, 1997) and new techniques for combining eye movement protocols with event-related potential data (e.g., King, Coulson, Federmeier & Kutas, 1997, in preparation). One difficulty with this approach is that as our paradigms become more realistic, the signals themselves (whether ERP or fMRI or something else) become more difficult to analyze. So a third part of my research program focuses on issues of signal processing, most recently with signal processing experts such as Irina Gorodnitsky, and in the next several weeks, local fMRI experts such as Paul Reber.
In the following paragraphs, I will detail my current and intended future
work on the time course of language processing, the implementation of
more ecologically valid language processing paradigms, and improvements
in the analysis of the signals most important to cognitive neuroscience.
