These paradigms have led to discordant results both in terms of performance and brain activity. For example, Nyberg et al. (2000) observed relatively poorer memory performance accuracy for words that had been paired with sounds during encoding vs. words that had not (76% vs. 84%, respectively).