The hearer therefore uses a variety of fine-grained clues, as proposed in the beginning of the article, to enhance his or her comprehension of a spoken utterance. Article 4: Electrophysiological evidence for early contextual influences during spoken-word recognition

The study depicted in this document focuses more on the physiologically cognitive basis of spoken-word recognition than the specific workings of the phenomenon itself. The aim is to determine the electrophysiological manifestation of contextual effects on spoken-word recognition. As such, the aim is further to validate early electrophysiological manifestation in terms of the interface between words and their context. As a component of event-related potentials (ERPs) is the N400, a negative-going component that denotes the semantic processing of a word. The N400 component is larger for words recognized to be incongruent in the context of their surrounding utterances. N400 however only denotes semantic integration, which is the final process of word recognition. One of the aims of the study is then to study the N400 process to determine whether there are prior processes related to the entire word-recognition sequence prior to integration. This study has been preceded by several other studies involving N400, with differing results and interpretations. One of these is the phonological expectation in terms of preceding context. This is identified as the early N400 effect. Another study concluded that a semantic expectation was formed. While the results contrast, both findings indicate processes that precede the final integration indicated by the N400 component. While the study in the document did indeed identify an early negativity, this did not substantiate the hypothesis of phonetic expectation. The conclusion is that bottom-up information received by acoustic processing influences word integration before top-down information has an effect.

Article 5: The cascaded nature of lexical selection and integration in auditory sentence processing.

The study addresses the question of whether lexical selection and semantic word integration into its preceding context can be separated or are parallel. The document cites studies...

The finding, after a number of experiments involving the N400 component, indicates a cascading effect: semantic word integration and lexical selection tend to be parallel rather than separate processes. This also substantiates the favor assigned to the bottom-up approach, as seen in the previous article. The idea of integration in terms of previous context is furthermore emphasized by the study. The authors also emphasize that speech is a very rapid phenomenon, and as such there is no time for full recognition before word selection. As such, words that are incongruent in the preceding context are sometimes selected due to time constraints, and these would then have to be repeated to be fully understood.
The distinguishing factor of the study is the fact that it uses natural connected speech in its investigations. In this way, more accurate results are obtained in terms of the way in which the speech process is hypothesized. The conclusion is then that lexical selection and semantic integration are two processes that happen in a cascading rather than separate form. The one cannot be completely separated from the other.

Sources

Norris, D., Cutler, a., McQueen, J.M., and Butterfield, S. (2006). Phonological and conceptual activation in speech comprehension. Cognitive Psychology 53; 146-193.

Shatzman, K.B., and McQuenn, J.M. The activation of offset-embedded words: Evidence from eye-tracking and identity priming.

Salverda, a.P., Dahan, D., McQueen, J.M. (2003). The role of prosodic boundaries in the resolution of lexical embedding in speech comprehension.

Van den Brink, D., Brown, C.M., and Hagoort, P. (2001). Electrophysiological evidence for early contextual influences during spoke-word recognition: N200 versus N400.

Van den Brink, D., Brown, C.M., Hagoort, P. (2006). The cascaded nature of lexical selection and integration in auditory sentence processing. Journal of Experimental Psychology: Learning, Memory, and Cognition, 32 (3): 364-372.