another, much as a computer programmer will irrevocably destroy an old program instruction when a new one is created.
A recent distinction has been made between active and inactive memory ( Lewis, 1979). Active memory (AM) is a subset of inactive memory (IM) and contains either newly formed memories or established retrieved memories or both. A body of evidence suggests that while in AM, memories are particularly open to disruption either by amnesic agents or through other forms of interference. Most of this evidence derives from the animal memory literature, yet it leads to the strong speculation that human memory may have to be evoked for it to be altered or distorted. Active memory is considered a changing subset of all permanent memories possessed by an organism. Over the course of time, numerous memories may become activated and these may be especially subject to change. Yet, at any given time, many of the permanent memories, which have the potential of being active, are in a relatively inactive state and have little effect on current behavior. Some of these memories may never be brought into an active state, and thus may not ever be subject to interfering events that could potentially cause their alteration or distortion.
We have raised the topic of permanence of memory and noted that over the years many researchers have implicitly held the view that information, once stored, remains forever in memory even though it may not be accessible. Such an idea is embodied in many of the traditional statements on memory. And yet, as we have argued, this traditional memory research may have very little to tell us about the workings of memory for natural events. The traditional research left little room for constructive errors and faulty attention to play a role. What a person stores after an experience is almost certainly a departure from what really happened, and yet how much of a departure is something that is enormously difficult to know. Even assuming that an accurate representation of reality is stored as the result of perceiving a complex episode, we know very little about how this representation changes over time. One possibility is that memories that are never again activated may remain relatively the same as they were initially stored; on the other hand, memories that are evoked become fragile and subject to potential transformations.
Adams J. A. Human Memory. New York: McGraw-Hill, 1967.