Effects of imagery instructions on false memories produced on implicit and explicit memory tests

In two experiments, we investigated whether the Deese-Roediger-McDermott paradigm using implicit and explicit memory tests would produce critical lure (CL) words. Participants studied lists of semantic associates (e.g., newspaper, book, write) to induce memories for CL words (e.g., read). Afterwards, participants participated in implicit and explicit memory tests in each experiment. The level of priming of the CL words was quite high and similar to that of the presented words when participants were instructed to image the interword relation on lists during study. Participants explicitly recognized many CL words as having been presented on the lists in Experiment 1 and recalled many CL words in Experiment 2. The results are discussed in light of Underwood's (1965) implicit associative response hypothesis.