Principal Investigator: Jack Gandour
Crosslinguistic PET Studies of Speech Prosody
Speech prosody is a critical component of language, conveying the melodic aspects of speech that extend beyond a single segment (e.g., pitch). How the brain processes speech prosody remains largely inconclusive, given that speech prosody is characterized by multiple phonetic parameters (e.g., pitch, duration), multiple functions (e.g., linguistic, affective) and levels of representation (e.g., syllable, word, sentence).
The objective of this research is to develop a functional model of the neuroanatomy underlying prosodic processing by observing the human brain in vivo while having normal adult speakers perform various tasks related to different aspects of prosody. Because of their phonological use of pitch at the syllable level, tone languages afford us a unique opportunity to develop experimental tasks addressing questions of speech prosody. With respect to pitch, we are able to focus primarily on how a single phonetic parameter is processed differentially depending on its role and level of representation in a language.
A working hypothesis is that phonological units of language, whether suprasegmental (prosodic) or segmental, are processed in a distributed neuralnetwork that includes the left perisylvian region near Broca's area. PET (positron emission tomography) imaging techniques are employed to compare patterns of activated brain foci while normal adult speakers of tone (Chinese) and nontone (English) languages perform same-different judgments of selected prosodic aspects of natural speech and filtered speech stimuli. Such crosslinguistic group comparison of brain activation patterns permit us to distinguish language-universal neural mechanisms that underlie analysis of complex auditory stimuli from those language-specific mechanisms attributed directly to language experience.
Specific aims of the proposed research are: (1) to identify the network
of brain areas activated with processing of segmental (consonants) and
suprasegmental (tones) phonological units (Tone vs Consonant in Chinese),
(2) to determine whether the network activated in processing speech prosody
is influenced by the level of linguistic representation (Tone vs Intonation
in Chinese), and (3) to determine whether the network activated with prosodic
processing depends on its use in a linguistic or affective context (Propositional
vs Affective Prosody in Chinese).
