The Curious Case of the Cat Rescue: Can Picture Narrative Description Inform Visuospatial Processing in Aphasia?

Sarah Grace Dalton^1*Javad Anjum¹Davida Fromm²Brian MacWhinney²

• ¹University of Georgia, Athens, United States
• ²Carnegie Mellon University, Pittsburgh, Pennsylvania, United States

Background: Examining discourse production during picture description performance holds great promise for understanding the nature of and the interconnectedness between visuospatial processing and language production in aphasia – a language disorder following acquired brain damage. There is a paucity of studies concurrently investigating the two processes in discourse production tasks, despite their potential clinical utility in aphasia rehabilitation. In the current study, we compared the core lexicon (CoreLex) word production performance of PWA and matched healthy control participants (HCP) along the dimensions of typicality of words (e.g. the words most frequently used by a normative sample of healthy controls), CoreLex word production timing, as well as the indirect visuospatial measures of order and spatial location of CoreLex word productions across the four quadrants of the Cat Rescue picture in a story telling task. Methods: A total of 319 transcripts from HCP and 400 transcripts from PWA, all of whom completed the picture description task of the Cat Rescue were drawn from the AphasiaBank database – the largest repository of aphasic discourse samples in the world. For each transcript, CoreLex scores and timing data of word production – including the elapsed time to produce each core lexicon content item and the first core lexicon content word in each quadrant – were indexed across the four quadrants of the Cat Rescue picture using the CLAN (Computerized Language ANalysis) program. Results: CoreLex analysis revealed that PWA were significantly slower compared to HCP in producing the first CoreLex word for each quadrant of the picture. PWA also demonstrated delayed CoreLex word production times across all the four quadrants, as well as lower rates of certain CoreLex word production compared to HCP. Aphasia severity was inversely related to the latency and accuracy of CoreLex production. Conclusion: Study findings offer preliminary evidence for the clinical utility of integrating concurrent visuospatial processing and language production tasks as part of discourse assessment in PWA. Future implications for further expanding and refining discourse-based visuospatial processing assessment tools in aphasia rehabilitation are discussed.