Abstract
The purpose of the present study was to examine factors that could influence whether recall performance in the reading span task (Daneman & Carpenter, Citation1980) would benefit from the contextual information from the sentences in the processing component of the task. More specifically, we investigated whether people would benefit from sentence sets that formed short stories or when the entire span task was one continuous story. Overall, there was a clear benefit for contextually related sentence sets (i.e., the story span tasks) compared to the traditional reading span task. However, the benefit was eliminated when the entire set formed one continuous story. These results support the recall reconstruction hypothesis for working memory (Towse, Cowan, Hitch, & Horton, Citation2008), which suggests that people may strategically use the content of the sentences from the processing component of the reading span task as memorial cues to reconstruct the target words of the storage component. However, this benefit is constrained to scenarios when the contextual cues are unique to a specific set.
Acknowledgments
We would like to thank Pamela Cornejo and Steven Lehr for their assistance with stimuli creation for this study. We would also like to thank Hilary Billings, Jennifer Bivens, Pamela Cornejo, Galit Melnick, Ashley Moreno, Adam Osman, Noemi Santana, Travis Shepard, Amber Throop, Jennifer Underhill, and Kathy Van Guilder for their assistance in collecting and analysing the data for this study. This project was not supported by any external funding sources.
Notes
1 We use the phrase “smaller benefit” because, in contrast to the reported analysis, this comparison was significant when using a more lenient scoring procedure that scored words as correct even if they were not recalled in the appropriate serial position (e.g., Friedman & Miyake, Citation2005), F(1, 59) = 5.04, p < .05, ηp 2 = .08.