The MIT Linguistics Colloquium Series presents:
Asaf Bachrach
Massachusetts Institute of Technology
Friday, April 11, 2008
3:30 p.m., Room 32-141
Language comprehension is a complex higher level cognitive task. The subtractive logic and the meta-cognitive tasks used in the conventional fMRI paradigm are ill suited for this domain. We recorded subjects' BOLD response while they passively listen to short narratives. We used linguistic theory and an automated parser to produce word level and sentence level (quasi) continuous predictors along multiple dimensions. These predictors were combined to model the observed hemodynamic time course. Using this novel combination of a naturalistic task (cf. Brennan et al 2007) and a multi-parametric design (cf. Bachrach et al 2007) we can now distinguish among the neuro-cognitive correlates of different aspects of parsing complexity.
