The results of 6 studies (involving 304 children) suggested that 4- and 5-year-olds, but not 3-year-olds, understand that very recent past events determine the present. In Studies 1-3, 3- and 4-year-old children were introduced to 2 empty hiding locations. With children's backs to these locations, a camera recorded an experimenter secretly hiding a puppet in one of them. Children then viewed the videotape of what had just happened, along with another tape that depicted identical events except with a different child and with the puppet hidden in the other location. Only 4-year-olds were subsequently able to locate the puppet, even though 3-year-olds remembered the contents of the tapes and understood the equivalence between the video events and the real world. In Study 4, similar effects were obtained when a verbal analog of the test was presented to 3-5-year-olds. Studies 5 and 6 showed that when children observed 2 events in which they had just participated, only 5-year-olds understood that the most recent events were relevant.