Inner voice experiences during processing of direct and indirect speech

Yao, B. and Scheepers, C. (2015) Inner voice experiences during processing of direct and indirect speech. In: Frazier, L. and Gibson, E. (eds.) Explicit and Implicit Prosody in Sentence Processing: Studies in Honor of Janet Dean Fodor. Series: Studies in theoretical psycholinguistics, 46. Springer International Publishing. ISBN 9783319129600

106480.pdf - Accepted Version


Publisher's URL:


In this chapter, we review a new body of research on language processing, focusing particularly on the distinction between direct speech (e.g., Mary said, “This dress is absolutely beautiful!”) and indirect speech (e.g., Mary said that the dress was absolutely beautiful). First, we will discuss an important pragmatic distinction between the two reporting styles and highlight the consequences of this distinction for prosodic processing. While direct speech provides vivid demonstrations of the reported speech act (informing recipients about how something was said by another speaker), indirect speech is more descriptive of what was said by the reported speaker. This is clearly reflected in differential prosodic contours for the two reporting styles during speaking: Direct speech is typically delivered with a more variable and expressive prosody, whereas indirect speech tends to be used in combination with a more neutral and less expressive prosody. Next, we will introduce recent evidence in support of an “inner voice” during language comprehension, especially during silent reading of direct speech quotations. We present and discuss a coherent stream of research using a wide range of methods, including speech analysis, functional magnetic resonance imaging (fMRI), and eye-tracking. The findings are discussed in relation to overt (or ‘explicit’) prosodic characteristics that are likely to be observed when direct and indirect speech are used in spoken utterances (such as during oral reading). Indeed, the research we review here makes a convincing case for the hypothesis that recipients spontaneously activate voice-related mental representations during silent reading, and that such an “inner voice” is particularly pronounced when reading direct speech quotations (and much less so for indirect speech). The corresponding brain activation patterns, as well as correlations between silent and oral reading data, furthermore suggest that this “inner voice” during silent reading is related to the supra-segmental and temporal characteristics of actual speech. For ease of comparison, we shall dub this phenomenon of an “inner voice” (particularly during silent reading of direct speech) simulated implicit prosody to distinguish it from default implicit prosody that is commonly discussed in relation to syntactic ambiguity resolution. In the final part of this chapter, we will attempt to specify the relation between simulated and default implicit prosody. Based on the existing empirical data and our own theoretical conclusions, we will discuss the similarities and discrepancies between the two not necessarily mutually exclusive terms. We hope that our discussion will motivate a new surge of interdisciplinary research that will not only extend our knowledge of prosodic processes during reading, but could potentially unify the two phenomena in a single theoretical framework.

Item Type:Book Sections
Additional Information:The final publication is available at
Glasgow Author(s) Enlighten ID:Scheepers, Dr Christoph
Authors: Yao, B., and Scheepers, C.
Subjects:B Philosophy. Psychology. Religion > BF Psychology
P Language and Literature > P Philology. Linguistics
Q Science > Q Science (General)
College/School:College of Medical Veterinary and Life Sciences > School of Psychology & Neuroscience
Publisher:Springer International Publishing
Copyright Holders:Copyright © 2015 Springer International Publishing Switzerland

University Staff: Request a correction | Enlighten Editors: Update this record