Researchers at Oregon Health & Science University and the University of Washington have published an article on mobile technology use by persons with primary progressive aphasia (PWPPA) during communicative exchanges with or without support of visual scene displays (VSD).
CoChat, the custom AAC technology used, gathers comments regarding PWPPA-generated photographs from simulated social media, analyzes them with natural language algorithms, and then selects sets of lexical items for textual VSD display to stimulate verbalizations. The goal was to study subjects’ lexical retrieval with and without these technological supports during structured dialogues.
Six subjects, drawn from the three identified subtypes of PPA, were enrolled in single-subject research protocols employing alternating treatment designs, to study lexical retrieval patterns during structured interviews in three conditions, namely, (i) absent all technological supports, (ii) in the presence of the photographs only, and (iii) using CoChat to present the subjects’ photographs surrounded by textual display of the ten algorithmically-generated lexical suggestions. In Phase I, these conditions were presented to all subjects in fixed order; in Phase II, presentations were counterbalanced among subjects. Finally, maintenance probes were administered for condition (iii) to one available subject. Researchers recorded, tallied, and compared numbers of the target words uttered by subjects in each condition, as the dependent variable.
In Phase I, all subjects performed best in condition (iii), that is, they included more target words in their verbal responses when supported by the photographs plus textual prompts displayed on CoChat than they did when supported by photographs alone, or without any technological support whatever. In Phase II, with conditions counterbalanced, four of the six subjects performed best in condition (iii), that is, using the CoChat display of photographs plus textual word prompts. In maintenance probes taken at six and nine months, the available subject’s performance success was found to be unchanged.
PWPPA are shown to benefit most often from VSD displays of personally generated photos with lexical suggestions in text. The researchers demonstrate a method of selecting word suggestions algorithmically by processing comments posted to emulated social media. The approach merits further research.
For further reading: A. Mooney, S. Bedrick, G. Noethe, S. Spaulding, & M. Fried-Oken. 2018. Mobile technology to support lexical retrieval during activity retell in primary progressive aphasia. Aphasiology.