Holzgrefe, J. 1 , Petrone, C. 2 , Schröder, C. 1 , Höhle, B. 1 , Truckenbrodt, H. 2 & Wartenburger, I. 1
1 Department of Linguistics. University of Potsdam. Potsdam, Germany.
2 Centre for General Linguistics, Berlin, Germany
Listeners’ comprehension of spoken language is guided by prosodic information provided in the uttered speech stream. Prosodic cues such as pauses, pitch changes, and durational properties mark prosodic phrase boundaries, which are used to structure an utterance, for instance to resolve syntactic ambiguities. Due to this immediate interplay with other linguistic domains, prosody got in the focus of attention of online investigations such as the Event Related Potential (ERP) technique. Indicating the perception of intonational phrase boundaries (IPBs), the closure positive shift (CPS) has been established as a neuro-physiological correlate of prosody processing. We report data of 18 monolingual speakers of German, who took part in an ERP experiment with auditory presented stimuli. In contrast to previous studies, we examined the perception of IPBs using short coordinate structures that contained three differently grouped names. In condition A, the IPB appeared after the first name (e.g., [Lola] # [oder Lena und Manu]), whereas in condition B, the boundary occurred later in the stimulus trial, namely after the second name (e.g., [Lola oder Lena] # [und Manu]). Comparing these two types of prosodic grouping, a CPS was elicited, but only coinciding with the late boundary (condition B); although acoustic analyses revealed that the relevant boundary cues were present in both conditions. Our results indicate that a CPS is not necessarily elicited when the relevant acoustic cues are present. Instead, the boundary position seems to play a crucial role, presumably because the relative strength of acoustic cues marking a prosodic boundary increases with the amount of speech material previously processed. This finding has to be taken into account for experimental designs focusing on boundary cue processing, but also strengthens the view of the CPS as a truly prosodic component that is not tied to mere acoustic properties of the stimulus material.