While many people with mild to moderate nonfluent aphasia show aided effects (that is, they produce better speech when using SentenceShaper than they can produce without the prosthesis), these effects do not always carry over into treatment effects (that is, their spontaneous, unaided speech may or may not improve). And people who do show treatment effects – that is, improvements in their spontaneous, unaided speech in laboratory tasks such as retelling narratives – may still experience great difficulty in stressful or highly interactive contexts. Thus treatment effects may be evident in certain social situations but not in others.
Regarding functional use, the literature on Augmentative and Alternative Communication (AAC) has documented that considerable support by family members, clinicians, or volunteers may be required to generalize aided effects from the laboratory to real-life, functional situations.
Another important caveat is that the research studies with the program to date have been relatively small. Larger studies will be required to determine SentenceShaper’s effectiveness for different populations, taking into account not only language impairments but attentional and memory problems as well.
Also unclear is the program’s “mechanism of action”…when it works well, WHY does it work? This, too, is likely to vary across individuals. Some may benefit from the memory support, which may allow them to perform language tasks (such as sentence or narrative construction) that would otherwise be impossible. Others may benefit from the opportunity to self-monitor. Still others may learn effective strategies from the program, such as “bootstrapping” from prepositions or consciously seeking opportunities to evoke a cloze effect.
Another caveat is that SentenceShaper has only been studied with nonfluent aphasia. There are hints (but no data as yet) that it may be useful for people with apraxia of speech and developmental language disorders.
In addition, as noted in the section about research on treatment effects, treatment studies with SentenceShaper have focused upon a narrative-based protocol: people used the program to construct narratives, such as retelling TV shows or life events. But, in fact, the program could be used to implement a variety of treatments: for example, script training (Holland, Milman, Munoz, & Bays, 2002) or Verb Network Strengthening Treatment (VNeST; Emonds, Nadeau, Kiran, 2009). Its efficacy with these and other protocols is unknown.
Please feel free to contact us if you are planning to use SentenceShaper in a research study; will will be happy to provide any support you may need in working with the program. We are also planning to make available to researchers some data logging tools that record every user action while running the program, as well as usage times and so forth.