How Memory Is Built, Lost and Rewritten

Hermann Ebbinghaus first documented that material is remembered better when study sessions are spread out than when they are crammed into one sitting. It is the natural starting point because it is the oldest finding here, dated to 1885, and it still underlies spaced repetition study tools.

The first documented studies of this effect were published in 1909 by Edwina Abbott. Actively retrieving information, as with a practice test, strengthens long-term retention more than the same time spent re-reading, which pairs naturally with spacing.

Ebbinghaus also named this pattern from experiments on himself. In recall of a list, items at the start and end are remembered best and items in the middle worst, which shows that where something sits in a sequence shapes whether it is kept.

Hedwig von Restorff found in her 1933 study that when a list of similar items contains one distinctive item, that isolated item is remembered better. Position is not the only thing that decides what sticks, since standing out from the rest also helps.

Bluma Zeigarnik's research report appeared in 1927 after Kurt Lewin noticed that a waiter recalled unpaid orders better than paid ones. People tend to remember unfinished or interrupted tasks better than completed ones, so even the state of a task changes what is retained.

Betsy Sparrow, Jenny Liu and Daniel Wegner described this tendency in 2011. People who expect to find information again online are less likely to remember the information itself, a modern example of leaning on an outside source as part of one's memory.

Most adults cannot retrieve episodic memories from before about the age of three to four years. This is the earliest gap in a life story, and it shows that memory is not a continuous record from birth.

A forgotten memory returns without being recognized as one, and the person believes it is new and original. It was first documented in 1874 in the case of the medium Stainton Moses, and it shows that a memory can survive after its source is lost.

Elizabeth Loftus showed that people can recall events that never happened, or recall them differently from how they occurred, through suggestive questioning or misleading information. The memory here is not merely lost or detached from its source but actively wrong, which matters for eyewitness testimony.

This is a proposed phenomenon of recalling the past more positively than it was experienced. Studies of vacations in the 1990s compared real-time ratings with later recollections, which were more favorable, so even accurate-seeming memories can drift in tone.

A 1977 study at Villanova University and Temple University showed that repeated statements are rated as more likely to be true than new ones. It closes the path by showing that the familiarity memory produces can be mistaken for accuracy.