Dual vs. single encoding in recognition memory as a function of lag, delay, and availability of semantic information

Abstract

One objective of the study was to assess the view of Winograd and Raines (1972) that the absence of a difference between dual as opposed to single encodings of a word at long lags may be due to the fact that subjects can establish two independent temporal tags when a to-be-remembered (TBRI word is presented at long lags. Winograd and Raines' argument that the effectiveness of any such temporal tags should diminish over time was examined. A second objective was to determine whether providing information about the semantic senses used during encoding operations would facilitate recognition, as "sense tagging" theories require. Words were presented twice in either the same low-frequency context or the same high-frequency context, or once in both a high- and low-frequency context at Lags 0 and 15. A recognition test was given immediately, or after a 1-week delay. The test was either context free or each word was accompanied by phrases which, in the case of TBR words, cued subjects with respect to the semantic senses of the word used in initial encoding. All manipulations influenced performance, but not in accordance with theoretical expectations.