Effects of keyword on long-term retention: Help or hindrance?

Three experiments assessed the long-term effectiveness of the keyword mnemonic relative to a nonmnemonic (i.e., semantic-context) learning strategy. Following incidental-learning instructions, cued recall was assessed either immediately or after a 2-day delay. The keyword mnemonic produced superior immediate performance relative to the semantic-context strategy. However, after 2 days, there was a marked reversal in performance, with higher levels of delayed recall associated with semantic-context learning. This pattern of findings was obtained when obscure English words (Experiment 1) and second-language vocabulary (Experiment 2) were the learning stimuli. When practice frequencies were manipulated (Experiment 3), increased opportunities for study were more likely to boost the long-term retention of keyword learners compared with semantic-context learners. The implication is that keyword-based memories are especially fragile over time and will benefit from repeated testing and rehearsal.