Delphine Dahan
University of Pennsylvania
The research that I will be presenting examines the dynamics of spoken word recognition through the use of a relatively new paradigm. Here, listeners' on-going interpretation of a spoken word is assessed through the recording of their eye movements to pictured objects displayed on a computer screen as listeners hear the name of one of the objects and select the correct referent using the computer mouse. Analyses of the location and timing of saccadic eye movements reveal how interpretation evolves as the spoken input accumulates. I will present evidence that listeners make immediate use of fine-grained information in the speech signal. Furthermore, far from ignoring or abstracting away from the variability that characterizes word forms across talkers and contexts, listeners appear to dynamically adjust their interpretations to talkers and contexts. Such findings suggest that listeners' mental representations of word forms may be less abstract and static than traditionally conceived.
A reception follows in 1413 Marie Mount Hall.