The present article is the first of three using response time and accuracy to explore recognition and cued recall of recently presented words. This article examines recognition with the methods of free response and signal-to-respond. Variations of list length and study time per list had a large effect on accuracy, but a small or negligible effect on: 1) free response response time distributions, and 2) retrieval dynamics in signal-to-respond. We fit these data with a model in which the timing of the decision to respond is based on an Assessment of Retrieval Completion (termed ARC), rather than on a sufficiency of evidence in favor of one of the response options. This ARC model is built within the framework of the REM model of Shiffrin and Steyvers (1997). By including an assumption that forgetting occurs during each trial's few seconds of retrieval, the model predicted both the dissociation of accuracy and response time and the finding that errors are slower than correct responses.
Download this paper in PDF format:
To view PDF files, download Adobe Acrobat Reader (it's free) from http://www.adobe.com/prodindex/acrobat/readstep.html.
