Writing treatment for aphasia using phonological training works best when speaking nonsense. Either nonsense or Italian. Six writing treatment studies demonstrate that research for phonological training treatment is at Phase II. Such treatment would be in a Phase III level of evidence if the studies reviewed had quasi-experimental studies or closer examinations of efficacy. Phonological writing treatment is still in Phase II as illustrated by these three similarities in research efficacy: first, how inconsistent procedures in research obscure a standardized research protocol; second, how irrelevant outcomes confuse data; and third, how the lack of clearly defined participant characteristics mar research results. The purpose of this paper is to demonstrate that though writing treatment using phonological training may have some benefits for some patients with aphasia, the treatment is generally at a Phase II level of evidence with little sign of moving to the quasi-experimental Phase III level.
One barrier for phonological research to move to Phase III is inconsistent procedures. The cognitive model is a central feature of the treatment, but the studies in this review used different versions of the cognitive model. Beeson, Rising, Kim, and Rapcsak (2008) use a simple version of the cognitive model comprised of “dual-route model of single-word processing”(p.708) showing the distinction between written and spoken words and their relationship to the internal language concept. Kiran’s (2005) cognitive model shows similar visual and aural inputs but no clear writing or speaking output. Luzzatti, Columbo, Frustaci, and Vitolo (2000) include the “information processing model for picture naming, word naming, and spelling” (p.252). Their model is similar in some ways to the Beeson and colleagues (2008) model but also includes a sole visual analysis separate from the lexical or sublexical language analysis. Beeson, Rising, Kim, and Rapcsak (2010) and Bowes and Martin (2007) mention the use of a cognitive model of language theory but do not elaborate on its implementation. To add to the inconsistencies, Carlomagno, Columbo, Casadio, Emanuelli and Razzano (1991) contrast “phoneme-grapheme conversion procedures or lexical (whole word form) writing strategies” without referring to a model. Luzzatti and colleagues (2000) assert, “if patients are not able to retrieve a direct lexical representation, they will try to compensate the graphemic lexical damage using sub-word-level abilities” (p.251). Here, the phoneme to grapheme sublexical method is not only one tool for the clinician to use in achieving greater writing ability, but also a tool the individual may use independently of the clinician. The cognitive model is integral in formulating the procedures for teaching phoneme-grapheme writing at the phonological level. The lack of a cohesive cognitive model demonstrates that phonological writing treatment research remains...