Expectancies generated by recent exposure to melodic sequences

In four experiments, we examined the effects of exposure to unfamiliar tone sequences on melodic expectancy and memory. In Experiment 1, 30 unfamiliar tone sequences (target sequences) were presented to listeners three times each in random order (exposure phase), and listeners recorded the number of notes in each sequence. Listeners were then presented target and novel seq uences and rated how well the final note continued the pattern of notes tha t preceded it. Novel sequences were identical to target sequences, except f or the final note. Ratings were significantly higher for target sequences t han for novel sequences, illustrating the influence of exposure on melodic expectancy. Experiment 2 confirmed that without exposure to target sequence s, ratings were equivalent for target and novel sequences. In Experiment 3, new listeners were assessed for explicit memory for target sequences follo wing the exposure phase. Recognition of target sequences was above chance, but unrelated to expectancy judgments in Experiment 1. Experiment 4 replica ted the exposure effect, using a modified experiment design, and confirmed that the effect is not dependent on explicit memory for sequences. We discu ss the idea that melodic expectancies are influenced by implicit memory for recently heard melodic patterns.