中文摘要：

对47名大学生被试依次进行自然情境中的人为任务、自然情境中的自然任务、实验室情境中的人为任务和实验室情境中的自然任务的测试。结果发现，相同任务情境中的不同性质的前瞻记忆具有一致性，特别是在自然情境中的一致性更为明显；相同任务性质但不同任务情境下的前瞻记忆没有一致性，实验室中的前瞻记忆与自然情境中的前瞻记忆不能相互预测。

英文摘要：

Prospective memory has become one of the hot spots in memory researches since 1970＇ s. There are two paradigms of prospective memory research -- the daily paradigm and the laboratory paradigm. In the daily paradigm research, prospective memory tasks are assigned to the subjects in everyday life （ natural situations）, such as sending postcards to the experimenter. In the laboratory paradigm research, prospective memory tasks are assigned to the subjects in the laboratory （laboratory situations）, such as pressing the prescriptive keys when previously designated words appear on the computer screen. On this basis, Kvavilashvili （1992） divided the tasks which were assigned to the subjects in the course of the experiment into two kinds, i.e. natural task and artificial task. The former refers to the prospective memory task which links the experiment situation logically and is not realized by the subject, which belongs to the experiment task ; the latter refers to the prospective memory that doesn＇ t link the experiment logically and could be realized by the subject, which is the experiment task. So the tasks of prospective memory experiments could be divided into four types according to the experiment situation and task property： artificial task in natural situation （ task type t ）, natural task in natural situation （ task type 2）, artificial task in laboratory situation （ task type 3 ） and natural task in laboratory task （ task type 4）. Kvavilashvili thought the natural task in laboratory situation （ task type 4） was the most ideal for prospective memory research, because it could avoid the ceiling effect and it took account of the ecological effect. This research started from the classification of prospective memory tasks and discussed further the consistence of the performance that subjects finished on the four types of prospective memory tasks. Specifically, it＇ s to make certain whether the performances on the artificial and natural tasks in laboratory situation and natura