Development of a Novel Writing System

PI: Julie Fiez

Abstract

The goal of this project is to determine whether the processing of visual words can be exclusively mediated by right-hemisphere visual areas. To address this question, we explore a novel orthography in which spoken English phonemes are graphically represented by faces, which evoke a strong activation in right hemisphere visual areas. We assess whether native English subjects can become fluent at reading this "face-based" writing system, and probe whether predominantly right-hemisphere visual areas support the processing of items displayed using this system.

Glossary

PSLC General Glossary

Full Glossary [edit] Research question The basic research question are: 1) can participants learn to read words and text presented using face-based graphemes in an alphabetic writing system (FaceFont), 2) how does the learning compare to a novel alphabetic orthography based upon more traditional letter-like graphemes (KoreanFont)?, and 3) is fluent FaceFont reading supported by right-lateralized activation within the fusiform, while fluent KoreanFont activation is supported by the typical pattern of left-lateralized fusiform activation?

What instructional activities or methods cause students’ learning to be robust? Participants receive direct instruction on the grapheme-phoneme mapping principles of the writing system to which they are assigned, and then practice decoding at the single word level, using software that provides assistance as requested by the subject and information about the correct response after each naming attempt of the participant. In later stages of training, participants read connected text, using the contextual constraints to ascertain the accuracy of their decoding.

[edit] Dependent variables 1) reaction time to read single words, and rate of reading (words per minute) of connected text 2) changes in the BOLD response to visually presented English words, FaceFont words, and KoreanFont words in the left and right fusiform

Measures of basic learning (an immediate, near-transfer post-test) and measures of robust learning (long-term retention, far transfer and accelerated future learning) Our single-word reading tests include words that are part of the training set, words that are not part of the training set, and nonwords; this allows us to examine the degree of transfer to never-before seen grapheme combinations

[edit] Independent variables orthographic system (FaceFont, KoreanFont)

[edit] Hypotheses We hypothesize that participants will be able to learn both KoreanFont and FaceFont, though the learning of KoreanFont may be faster and particpants may be able to achieve much higher rates of fluency with extended training. The visual-perceptual account of fusiform organization predicts that skilled FaceFont reading will be supported by the right fusiform, while the phonological-bridge account predicts that skilled FaceFont reading will be supported by left-lateralized or bilateral fusiform activation. [edit] Explanation

The visual-perceptual account posits that the degree to which the left or right fusiform is engaged during orthographic processing is dependent upon the visual characteristics of the stimuli. Since faces are known to elicit predominantly right-lateralized fusiform activation (potentially because of the greater need for configural processing, or low-spatial frequency analysis) this account predicts that if faces are used a graphemes the fusiform activation will be largely right lateralized.

The phonological-bridge account posits that the left fusiform, in the territory of the "visual word form area" must always be used to support skilled reading, because this region is uniquely situated to connect to left hemisphere language areas (e.g., a region in the left temoroparietal cortex associated with phonological representation). This account thus predicts that regardless of the visual form of the writing system, the visual word form area will always be engaged in a skilled reader.

Pilot results obtained by a PSLC intern in 2008 indicate that the FaceFont system can be learned, although at a slower rate than KoreanFont. Initial imaging results further suggest that FaceFont reading is associated with right lateralized activation of the fusiform. Thus, the initial evidence supports the visual-perceptual account of the fusiform and reading. An NIH grant to study the FaceFont writing system in greater depths is in the final stages of review, and we are hopeful that we will be able to build upon this pilot work beginning in Fall 2009.

[edit] Descendents

   * Interactive Communication.
   * Coordinative Learning.
   * Refinement and Fluency.