Selecting a phoneme-to-grapheme mapping: Random or weighted selection?
-
1
New York University, Communicative Sciences and Disorders, United States
Introduction
It is widely held that there are two routes for written word production: lexical and sublexical. The dual-route account explains both how people spell words that are familiar (via the lexical route) and novel words (via the sublexical route). Impairment affecting the lexical route often leads individuals to rely on the sublexical route, resulting in frequent production of phonologically plausible errors (PPEs; e.g., “brownie” → BRAUNEE) for real words, and intact spelling of nonwords. The selection of specific graphemes in the sublexical system in the case of PPEs and nonwords may reflect common sound-spelling correspondences (Sanders & Caramazza, 1990), such that graphemes with high probability phoneme-to-grapheme (PG) mappings are preferred to low probability mappings (e.g., /f/ → {F} more than {PH}; /r/ → {R} more than {RH}). However, other studies have reported some PPEs consisting of low probability mappings (Baxter & Warington, 1987).
We investigated how PG mappings are selected in sound-level PPEs in an individual with acquired dysgraphia. In particular, we examined each instance of a sound being produced with a non-target spelling, and compared the errors we obtained to predictions of accounts in which graphemes are selected with equal probabilities (random selection) and with weighted probabilities based on the frequency of occurrences in a language.
Method
MOA (b. 1980), LH, suffered an AVM in 2010, and presents with impaired reading and spelling skills and relatively intact spoken production. A total of 1182 words were administered (17.2% accuracy on word-level, 69.8% accuracy on letter-level spelling). MOA’s spelling exhibits both a length effect (4/5 letter: 43% vs. 7/8 letter: 3.1%, X2=43.35, p < .0001) and frequency effect (LF: 14.8% vs. HF: 32.7%, X2=14.44, p < .001). A test of nonword spelling indicated that he has access to the sublexical spelling route, but his PG conversion system is not entirely intact (319/378 phonemes in nonwords correct, 84.4%).
Procedure
We identified all sound-level PPEs in MOA’s spelling (N=654) and noted the PG mapping of the target and the response (based on syllable-position from Hanna et al., 1966). We then used Monte Carlo simulations to generate chance distributions of what would be expected if graphemes were: (1) selected randomly from all the possible options for a given phoneme; or (2) selected based on their frequency of occurrences (weighted). We then ran the same analyses on nonwords, including each accurately spelled letters based on produced sounds (N=319).
Results
For both PPEs and nonwords, MOA shows an average PG correspondence that is significantly higher than an expected mean by random selection, but significantly lower than an expected mean by weighted selection, in both words and nonwords (see Table 1).
Discussion
Our findings demonstrate that random selection underestimates MOA’s PG correspondences whereas weighted selection predicts higher PG correspondences than he produces. To explain his intermediate spelling performance on PPEs, we will test additional approaches to weighing the relative probability of PG mappings, including using log frequencies, separating consonant and vowel status, and considering the number of grapheme options in each phoneme.
References
Baxter, D. M., & Warrington, E. K. (1987). Transcoding sound to spelling: single or multiple sound unit correspondence? Cortex, 23(1), 11-28.
Hanna, P. R. (1966). Phoneme-Grapheme Correspondences as Cues to Spelling Improvement.
Sanders, R. J., & Caramazza, A. (1990). Operation of the phoneme-to-grapheme conversion mechanism in a brain injured patient. Reading and Writing, 2(1), 61-82.
Keywords:
dysgraphia,
sublexical orthographic processing,
Phoneme-to-grapheme conversion,
phonologically plausible errors,
spelling
Conference:
Academy of Aphasia 53rd Annual Meeting, Tucson, United States, 18 Oct - 20 Oct, 2015.
Presentation Type:
Poster
Topic:
Student first author
Citation:
Lee
B and
Buchwald
A
(2015). Selecting a phoneme-to-grapheme mapping: Random or weighted selection?.
Front. Psychol.
Conference Abstract:
Academy of Aphasia 53rd Annual Meeting.
doi: 10.3389/conf.fpsyg.2015.65.00012
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
01 May 2015;
Published Online:
24 Sep 2015.
*
Correspondence:
Ms. Binna Lee, New York University, Communicative Sciences and Disorders, NY, NY, 10012, United States, bnlee@nyu.edu