Discontinuity between episodes

The extended present and the immediate past are phenomenally discontinuous. In contrast to the present, which is immediately available, to reexperience the past requires an act of episodic recollection (Tulving, 1983). Discontinuity theories in memory, notably the ones exemplified by the short-term memory/long-term memory (STM, LTM) distinction, are supported by experiments that demonstrate discontinuities in the function that relates performance on memory tasks to time elapsed since the memoranda were presented. The present chapter introduces evidence of an analogous discontinuity in the domain of estimation of time past, and of a specific pathology causing a selective loss of one phase in remembering time past. In a paradigm in which information flow to the subject is continuous, the findings isolate a time period within which information is retrievable to an extent inversely proportional to the time elapsed since it was presented. We interpret this in terms of William James's sweeping formulation of the extended present as representing “the lingerings of the past dropping successively away, and the incomings of the future making up the loss” (James, 1890, p. 606). We take this to mean that for James, the extended present is the contents of the present experience, the “episode,” sliding across time.

Short-term memory is usually measured in relation to an input phase, during which stimuli are presented in sufficient amount fully to engage or even overload the cognitive system. The ability to recover this information at varying times afterward is measured (e.g., Peterson & Peterson, 1959). But in the absence of informational overload, can one infer a critical period of time, beyond which experienced events are relegated to the past?