Abstract
Forty undergraduates intentionally learned 36 words presented on a computer screen one-by-one. Each word was superimposed on one of 36 (Load 1) or 2 (Load 18) different 4-s video clips with background sounds (video contexts). In the Load 18, the same 18 contexts were successively presented. After 5-min filled retention interval, undergraduates received 36 old and 36 new words at random order either on one of 18 old and 18 new contexts (Load 1) or on one of 2 old and 2 new contexts (Load 18), and were required to judge whether or not the currently presented word had been presented at learning. When each old word was tested on the old context, the same context as at learning was used. Hit and A' showed the superiority of the old context condition only in Load 1, whereas FA showed the superiority both in Loads 1 and 18.
