There is a general consensus in theories of human speech recognition that humans engage in predictive processing during online speech processing. There are also claims that predictive processing indicates the operation of a predictive coding (PC) mechanism (Rao & Ballard, 1999). Formally, PC is a generative model where top-down signals consist of predictions while bottom-up signals consist of prediction error (deviations from predictions; Rao & Ballard, 1999). Some researchers have taken decreased neural signals when inputs conform to expectations as evidence for PC (e.g., Blank & Davis, 2016), and that other computational frameworks (e.g., interactive activation, as in the TRACE model of McClelland & Elman, 1986) are incompatible with apparent reductions in prediction error when inputs match expectations. However, these claims were made without linkage to computational models faithful to formal PC because no such implementations existed. This dissertation fills this gap by implementing and testing the first faithful implementation of PC as a model of human spoken word recognition (SWR). After validating the new model on fundamental aspects of SWR (the timecourse of lexical activation and competition), I compared model simulations to two behavioral and neural targets. While the new model readily captures basic activity reduction when inputs match expectations (Gagnepain, Henson, & Davis, 2012), more subtle and detailed predictions regarding patterns of activity under conditions of priming and noise (Blank & Davis, 2016) were not present. This suggests that the more subtle patterns may not be hallmarks of PC, and motivates discussion about pitfalls of inferring complex computational mechanisms from complex neural data. The new PC model fills a fundamental gap in current computational understanding of SWR. Thus far, the results with this model challenge current claims about the nature of neural activity predicted by PC. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com.bibliotheek.ehb.be/en-US/products/dissertations/individuals.shtml.]