Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form


Chang, Li-Yun (2015) VISUAL ORTHOGRAPHIC VARIATION AND LEARNING TO READ ACROSS WRITING SYSTEMS. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

Primary Text

Download (1MB)


This research examined the extent to which visual characteristics of orthographies affect learning to read within and across writing systems, with an eye toward the role of mapping principles – the manner in which graphemes map to linguistic units (e.g., phonemes, syllables, and morphemes) in this process. Study 1 explained visual orthographic variation by developing a measurement system to quantify complexity of graphemes in 131 orthographies. The results show that grapheme complexity varies across writing systems and that this variation is driven by grapheme inventory, a consequence of mapping principles. Next, we questioned how visual orthographic variation impacts individuals’ perceptual learning of graphemes – one of the initial stages of learning to read. Study 2 tested the degree to which mastering first-language (L1) graphemes with different complexities affects visual perceptual discrimination for individuals using different mapping principles (Online cross-writing-system experiment; eight participant groups: Hebrew, English, Russian, Arabic, Hindi, Telugu, Japanese, and Chinese, n = 60, respectively) and individuals using the same mapping principle (Lab within-writing-system experiment: simplified vs. traditional Chinese, n = 60, respectively). Consistent results from both experiments show that discrimination difficulty is a function of grapheme stimulus complexity itself as well as its relationship to the complexity of participants’ L1, regardless of mapping principles. These results were confirmed in Study 3, in which we developed a universal orthographic neural network encoder focus on statistical properties of visual patterns to simulate human behaviors. We trained each of 131 identical encoders to learn the structure of a different orthography; a strong, positive association was found between grapheme complexity and network learning difficulty. Taken together, our results suggest that visual orthographic variation, encompassing both grapheme complexity and grapheme inventory required for orthographic mastery, affects visual discrimination processing of graphemes; these complexity effects are driven significantly, but not absolutely, by mapping principles across writing systems.


Social Networking:
Share |


Item Type: University of Pittsburgh ETD
Status: Unpublished
CreatorsEmailPitt UsernameORCID
Chang, Li-Yunlic72@pitt.eduLIC72
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairPerfetti, Charles A.perfetti@pitt.eduPERFETTI
Committee MemberPlaut, David
Committee MemberFiez, Julie A.fiez@pitt.eduFIEZ
Committee MemberTokowicz, Natashatokowicz@pitt.eduTOKOWICZ
Date: 17 June 2015
Date Type: Publication
Defense Date: 8 January 2015
Approval Date: 17 June 2015
Submission Date: 22 January 2015
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Number of Pages: 136
Institution: University of Pittsburgh
Schools and Programs: Dietrich School of Arts and Sciences > Psychology
Degree: PhD - Doctor of Philosophy
Thesis Type: Doctoral Dissertation
Refereed: Yes
Uncontrolled Keywords: learning to read, visual orthography, writing systems
Date Deposited: 17 Jun 2015 16:56
Last Modified: 15 Nov 2016 14:26


Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item