This research investigates the role of prosody (intonation and rhythm of speech) in the processing of syntax in Turkish. Many experiments have shown that the prosody with which a sentence is uttered can provide listeners with cues to its syntactic structure; some have also observed that an inappropriate prosodic contour can mislead the syntactic parsing routines, causing comprehension difficulty. It has been proposed that when prosodic cues are not supportive of the syntactic structure, the language processing mechanism ignores the prosodic input and relies on syntactic parsing strategies such as Late Closure (a locality preference). However, a closer look at the available findings suggests that the lengths of phrases could have caused what appeared to be an effect of default syntactic strategies, since it is known that phrase lengths can influence the perceived informativeness of prosodic cues.

In the present study, four experiments will be conducted to distinguish these alternative explanations. Misleading and facilitatory effects of prosody will be examined in Turkish, a language that is typologically different from more commonly studied languages (such as English) and has properties well-suited to this purpose. Phrase lengths will be systematically manipulated in all experimental sentences, such that optimal-length prosodic phrasing is associated with late closure syntax in one condition, and with early closure syntax in the other.

This study will make an original contribution to current knowledge about the mental mechanisms underlying human sentence processing. Outcomes for Turkish similar to those for previously studied languages would strengthen the general hypothesis that prosody-syntax interactions are guided by fundamental universal properties of the human parsing mechanism. The specific results will disentangle two potential accounts of how that mechanism responds to imperfect input with conflicting cues, as is not uncommon in everyday language use. Other potential contributions include text-to-speech/speech-to-text software development in Turkish with a wide range of applications including programs to aid the visually handicapped.

Agency
National Science Foundation (NSF)
Institute
Division of Behavioral and Cognitive Sciences (BCS)
Type
Standard Grant (Standard)
Application #
1250473
Program Officer
William Badecker
Project Start
Project End
Budget Start
2013-04-15
Budget End
2015-03-31
Support Year
Fiscal Year
2012
Total Cost
$9,984
Indirect Cost
Name
CUNY Graduate School University Center
Department
Type
DUNS #
City
New York
State
NY
Country
United States
Zip Code
10016