Manipulating task instructions to change narrative discourse performance

Abstract

Background: A frequent method for eliciting narratives involves presenting picture stimuli and instructing the participant to “tell me what is happening in the picture(s) or tell me everything you see going on in the picture(s)”. It has been suggested that such instructions do not make an explicit request for providing temporal–causal information and so narrative samples frequently comprise listed information. Olness (2006) investigated the effect of task instruction on verb forms produced by adults with aphasia when describing pictures and found that using instructions that explicitly requested temporal–causal information—“Make up your own story about what happened, with a beginning, middle, and an end” (p. 179)—improved the ability of participants to convey the temporal–causal interrelationship depicted in single pictures. Further interpretation of the results was limited due to the lack of non‐brain‐injured participants for comparison. Aims: The purpose of the study was (1) to compare the performance of two groups of healthy older adults on the ability to convey main events in pictured stimuli when different task instructions were provided and (2) to quantify linguistic performance across the two groups and determine if groups significantly differed for any of the linguistic features. Method & Procedures: Participants included 24 healthy older adults assigned to one of two groups based on task instruction: picture description (PD) or storytelling (ST). Instructions were as follows—PD group: “Talk about what is going on in the picture(s)”; ST group: “I want you to look at the picture(s) and tell me a story that has a beginning, middle, and end”. Outcomes & Results: The ST group produced a significantly higher proportion of main events and significantly more past‐tense verbs than the PD group. Two significant relationships among the linguistic measures and main events measure were found for the PD group: a positive relationship between percent of information units (IUs) produced and proportion of main events for the sequential pictures and a negative relationship between D (Malvern & Richards, 1997) and proportion of main events for the single picture stimuli. Conclusions: Findings suggest that task instructions do affect the quality of the narrative produced by healthy older adults. The instructions to provide a story with a beginning, middle, and end may have acted as a scaffold for participants; cueing participants to provide the temporal–causal relationships depicted. Therefore, to better evaluate discourse production abilities in clinical populations, care should be taken in the task instructions provided.