On the permanence of stored information in the human brain.

Abstract

Many people believe that information that is stored in long-term memory is permanent, citing examples of techniques that are alleged to uncover previously forgotten information. Such tech- niques include hypnosis, psychoanalytic procedures, methods for eliciting spontaneous and other conscious recoveries, and—perhaps most important—the electri- cal stimulation of the brain reported by Wilder Pen- field and his associates. In this article we first evaluate • the evidence and conclude that, contrary to apparent popular belief, the evidence in no way confirms the view that all memories are permanent and thus poten- tially recoverable. We then describe some failures that resulted from attempts to elicit retrieval of pre- viously stored information and conjecture what cir- cumstances might cause information stored in memory to be irrevocably destroyed. Few would deny the existence of a phenomenon called forgetting, which is evident in the com- mon observation that information becomes less available as the interval increases between the time of the information's initial acquisition and the time of its attempted retrieval. Despite the prevalence of the phenomenon, the factors that underlie forgetting have proved to be rather elusive, and the literature abounds with hypothesized mechanisms to account for the ob- served data. In this article we shall focus our attention on what is perhaps the fundamental issue concerning forgetting; Does forgetting consist of an actual loss of stored information, or does it result from a loss of access to information, which, once stored, remains forever? It should be noted at the outset that this ques- tion may be impossible to resolve in an absolute sense. Consider the following thought experiment. A person (call him Geoffrey) observes some event, say a traffic accident. During the period of ob- servation, a movie camera strapped to Geoffrey's head records the event as Geoffrey experiences it. Some time later, Geoffrey attempts to recall and