Successful blending in a phonological reading treatment for deep alexia

Abstract

Background: Patients with deep alexia have great difficulty reading pseudowords, indicating an impairment in phonological reading. One method that has been used in attempting to retrain phonologic reading is to teach the patient to sound out words using grapheme-to-phoneme correspondence rules. Most of these studies found that the patients could learn the individual correspondences, but then could not blend them together in order to read words. Previous studies have reported that both children learning to read and patients with phonologic alexia are more successful blending syllables, rather than individual phonemes, into words. This may be due to the fact that phonemes are pronounced differently in isolation than they are in context. Aims: The hypothesis of the current study is that patients will be successful blending letter-to-sound correspondences into words if those correspondences are trained in context rather than in isolation. Methods & Procedures: The correspondences were trained, therefore, as bi graph-phoneme correspondences, rather than as individual grapheme-phoneme correspondences (e.g., ''pa''-/pæ/ and ''at''-/æt/, rather than ''p''- /p, ''a''-/æ/ and ''t''-/t). The study followed a single-subject multiple baseline design in which three sets of bigraphs, and words composed of those trained bigraphs, were trained sequentially. Two subjects with deep alexia participated in the study. Outcomes & Results: Subjects LR and KT successfully blended trained bigraphs in order to read both trained and untrained words that were composed of the trained bigraphs. Conclusions: It is suggested that the patients are learning a new means of decoding words, and that the underlying phonologic processing deficit remains. In terms of clinical application, the bigraph approach offers a viable alternative approach for re-training reading for those patients who are unable to blend individual phonemes into words.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 36 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.