Help Privacy Policy Disclaimer
  Advanced SearchBrowse





Event-related brain activation in speech perception: From sensory to cognitive processes


Sivonen,  Päivi
Department Neuropsychology, MPI for Human Cognitive and Brain Sciences, Max Planck Society;

External Resource
No external resources are shared
Fulltext (public)

(Any fulltext), 6MB

Supplementary Material (public)
There is no public supplementary material available

Sivonen, P. (2006). Event-related brain activation in speech perception: From sensory to cognitive processes. PhD Thesis, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig.

Cite as: http://hdl.handle.net/11858/00-001M-0000-0010-B2BE-5
This thesis investigated electrophysiological correlates of underlying brain activation at different levels of speech processing. In Experiment I, different phonation types of spoken vowels activated slightly separate neural assemblies within the auditory cortex, indicating processing of voice quality. Processing speed differed between speech and a simple non-speech tone. In Experiment II the formant structure influenced cortical activation according to the peciodicity of the glottal excitation, its influence being larger to a non-periodic excitation. A periodic non-speech excitation elicited equally large responses than a periodic speech sound. The amplitude of the N1(m) response was suggested to reflect acoustic rather than phonetic properties of speech sounds. Further, the results pointed to the importance of periodicity as a general acoustic feature. Experiments III– V studied influence of contextual semantic information on the processing of words in sentences when the word’s onset was obliterated. The type of the replacement was varied in order to reduce a possible support of coarticulatory and acoustic information in recognising the degraded words. The N1–P2 -response complex reflected an automatic detection of the replacement. A delay of the N400 indicated increased demands in semantic processing when the word’s onset was obliterated. The N400 amplitude was not significantly increased after diminishing the immediate coarticulatory information or removing the acoustic information completely. The contextual information helped the word recognition and integration, possibly having an influence already before the whole word was heard. Experiment VI indicated an involvement of left superior and anterior temporal areas and the left inferior frontal cortex in semantic processing. Furthermore, an active detection task of the semantic violations compared to listening increased the N400 response, as well as delayed the activation in the left inferior frontal cortex. In summary, the results suggested periodicity as an important acoustic feature in the auditory analysis. No hemispheric lateralisation was present at early processing levels of individual speech sounds whereas later in semantic processing the activation was clearly left lateralised. The semantic processing was influenced by increasing task demands and by the phonetic information provided at the word’s onset. Absent onset phonemes delayed the semantic processing of words, however, the sentence context provided supporting information for the word recognition.