Project Report

The processes employed to read (convert letters to sounds) are proposed to be different for different native languages. There is convincing evidence that native speakers of a character-based language such as Chinese process words as "chunks", due to Chinese having a poor grapheme-to-phoneme correspondence (parts of Chinese characters do not provide reliable clues to the word’s pronunciation). As a result, Chinese readers do not appear to process individual sounds of Chinese words before making a decision about a whole character’s pronunciation (Feng, Miller, Shu, & Zhang, 2001; Liu, Wu, Sue, & Chen, 2006). A Chinese reader must match the appearance of the complete character to a corresponding set of sounds found in the mental lexicon. This process is analogous to looking up a word in a dictionary, but without the benefit of individual letters to aid in the identification of the correct correspondence. Unlike Chinese, Korean is an alphabetic language where each letter (a line or two oriented in a particular manner, or a circle, or some combination thereof) corresponds to a single unique sound (Sohn, 1999). Co-occurring with this transparent grapheme-to-sound relationship is a tendency for Korean readers to process individual sounds as they read a word, thus allowing them to narrow the list of possible word-to-sound correspondences during the reading process before making a final decision about a complete word’s pronunciation. This study helps establish and disambiguate the effects that native language and English proficiency (operationalized as amount and type of English exposure and production) have on English letter-to sound processing during reading, also known as decoding. The study asked native speakers of Korean (n=23) and Chinese (n=89) to perform an English word reading task. Decoding (the conversion of print to sound) was isolated by having participants read rhyming and non-rhyming English word pairs. The word pairs were orthographically dissimilar and controlled for frequency and length (die / sky). After reading a word pair, participants were asked to decide whether or not the words rhymed. The time needed to make a correct affirmative or negative judgment was measured and recorded in milliseconds. Results indicated that native language does not significantly impact decoding time, t (110) = -.466, p = .321. There was concern that the result of this test was unduly influenced by the unbalanced sample sizes (Chinese = 89, Korean = 23), so five subsamples of the Chinese data, all random samples of 23 participants, were compared to the Korean data, and all exhibited similar results: t (44) = -.127, p = .450; t (44) = .448, p= .328; t (44) = -.343, p = .367; t (44) = -.471, p = .320; t (44) = .106, p = .458, indicating that there were not statistically significant differences between Chinese and Korean group decoding speeds, meaning that native speakers of Korean do not decode English words more quickly or slowly than their Chinese counterparts. To investigate proficiency effects, decoding times were regressed on overall English proficiency ratings in a bivariate linear regression, with the results indicating that proficiency explained a statistically non-significant portion (less than 1%) of the variance (R2 = .002, F (1, 87) = .178, p = .674) in decoding times for the Chinese group. Nor did English proficiency significantly predict Korean decoding speeds, with regression results indicating that proficiency explained only 3.4% of the variance (R2 = .034, F (1, 21) =.736, p = .401) in decoding times for the Korean group. The study implies that native language reading processes do not significantly affect the speed of second language decoding. A further implication is that designing English reading curricula that target specific native language reading processes for alteration is not advisable in light of current research and may be a poor use of the limited and dwindling funds available for language education. Conversely, populations with different native languages may pool their resources to create a common English reading curriculum, confident that any particular linguistic group will not be put at a disadvantage when it comes to learning how to read. Results of the above study were presented at the Second Language Research Forum in October of 2011. A manuscript is currently being prepared for publication in a peer-reviewed journal of Second Language Acquisition or Applied Linguistics. References Feng, G., Miller, K., Shu, H., & Zhang, H. (2001). Rowed to recovery: The use of phonological and orthographic information in reading Chinese and English. Journal of Experimental Psychology: Learning, Memory, and Cognition, 27 (4), 1079-1100. Liu, I.M., Wu, J.T., Sue, I.R., & Chen, S.C. (2006). Phonological mediation in visual word recognition in English and Chinese. In Li, P., Tan, H., Bates, E., & Tzeng, O. J. L. (Eds.), The handbook of East Asian psycholinguistics: Vol. 1. Chinese (pp. 218-224). Cambridge, UK: Cambridge University Press. Sohn, H.-M. (1999). The Korean Language. New York: Cambridge University Press.

Agency
National Science Foundation (NSF)
Institute
Office of International and Integrative Activities (IIA)
Application #
1107341
Program Officer
Carter Kimsey
Project Start
Project End
Budget Start
2011-06-01
Budget End
2012-05-31
Support Year
Fiscal Year
2011
Total Cost
$5,000
Indirect Cost
Name
Amoroso Luke W
Department
Type
DUNS #
City
Arlington
State
VA
Country
United States
Zip Code
22201