Frontiers in Psychology (Apr 2015)
Modality differences between written and spoken story retelling in healthy older adults
Abstract
Story retelling is frequently used to measure spoken language. It differs from other forms of discourse by requiring the speaker to synthesize information, retain story elements in temporal order, retrieve elements from memory and summarize them linguistically (Doyle et al., 1998). The Discourse Comprehension Test (DCT) was originally created to examine discourse comprehension for brain-damaged adults (Brookshire & Nicholas, 1993). It has also been implemented as a measure of spoken story retelling in people with aphasia and in non-brain damaged adults (Doyle et al., 2000; McNeil et al., 2001). The stimuli from DCT has been used in research to create discourse elicitation probes based on standardized stimuli (Doyle, et al., 1998), create parallel forms of the story retelling task (Doyle, et al., 2000) and to validate the information unit measure (McNeil, et al., 2001). Although DCT has been used to examine the spoken story retelling ability of healthy older adults and people with aphasia, it has not been used as a measure of writing skills in healthy older adults. This information will provide insight into the modality differences between spoken and written discourse in normal adults, which can provide comparative data for persons with aphasia (our long term goal). The purpose of this study was to determine if and to what extent modality differences exist between spoken and written retellings of the story stimuli of the DCT in healthy adults. Methods: Ten native English speaking healthy elderly participants between the ages of 50 and 80 were recruited. Exclusionary criteria included neurological disease/injury, history of learning disability, uncorrected hearing or vision impairment, history of drug/alcohol abuse and presence of cognitive decline (based on Cognitive Linguistic Quick Test). Spoken and written discourse was analyzed for micro linguistic measures including total words, percent correct information units (CIUs; Nicholas & Brookshire, 1993) and percent complete utterances (CUs; Edmonds, et al. 2009). CIUs measure relevant and informative words while CUs focus at the sentence level and measure whether a relevant subject and verb and object (if appropriate) are present. Results: Analysis was completed using Wilcoxon Rank Sum Test due to small sample size. Preliminary results revealed that healthy elderly people produced significantly more words in spoken retellings than written retellings (p=.000); however, this measure contrasted with %CIUs and %CUs with participants producing significantly higher %CIUs (p=.000) and %CUs (p=.000) in written story retellings than in spoken story retellings. Conclusion: These findings indicate that written retellings, while shorter, contained higher accuracy at both a word (CIU) and sentence (CU) level. This observation could be related to the ability to revise written text and therefore make it more concise, whereas the nature of speech results in more embellishment and “thinking out loud,” such as comments about the task, associated observations about the story, etc. We plan to run more participants and conduct a main concepts analysis (before conference time) to gain more insight into modality differences and implications.
Keywords
