Computational modeling of an auditory lexical decision task using jTRACE
Loading...
Date
Author(s)
Citation for Previous Publication
Link to Related Item
Abstract
Description
The TRACE model of spoken word recognition has been widely discussed and used, but was never implemented to simulate the auditory lexical decision task, particularly on a larger number of items. In this study, we attempt to model accuracy and latency estimates and compare the obtained values to actual listener responses collected as part of the Massive Auditory Lexical Decision project. We find that when the lexicon of competitors includes a number of close competitors to the target, model accuracy in selecting the right winner is unsatisfactorily low. Similarly, the correlation between model estimates and mean logged participant response latency was very low under any settings we took into consideration. We discuss plans for future simulations which could potentially increase both model accuracy and fit to participant response latencies.
Item Type
http://purl.org/coar/resource_type/c_6501 http://purl.org/coar/version/c_970fb48d4fbd8a85
Alternative
Other License Text / Link
Language
en