If you feel fairly sure it was, you have just created a false memory. "Sleep" never appeared. This is the Deese–Roediger–McDermott (DRM) procedure, first used by James Deese in 1959 and revived by Henry Roediger and Kathleen McDermott in 1995. In their studies people recalled the missing theme word about as often as words that were actually there, and many said they vividly "remembered" hearing it.
The forgetting curve
In 1885 the German psychologist Hermann Ebbinghaus published Über das Gedächtnis (On Memory), the result of years spent memorising lists of nonsense syllables like "DAX" and "BOK" and testing himself. He found that forgetting is steep at first and then levels off: much of what you learn can vanish within a day unless revisited. A 2015 replication by Jaap Murre and Joeri Dros reproduced his curve remarkably closely.
The practical lesson is the spacing effect, which Ebbinghaus also noticed. Reviewing material a few times spread out over days beats one long cramming session. It is the principle behind flashcard apps and why revising little and often works.
Memory as reconstruction
In 1932 the Cambridge psychologist Frederic Bartlett asked English participants to read and retell "The War of the Ghosts", a Native American folk tale. With each retelling, unfamiliar details disappeared and the story shifted toward what made sense to the readers' own culture. Bartlett concluded that remembering is "imaginative reconstruction", built from fragments and expectations.
Elizabeth Loftus took this into the courtroom. In a 1974 study with John Palmer, people watched film of a car accident and were asked how fast the cars were going when they "smashed into" or "hit" or "contacted" each other. The verb changed the estimates: about 40.5 mph for "smashed" versus 31.8 mph for "contacted". A week later, people in the "smashed" group were more likely to report seeing broken glass, though there was none. Loftus's work has influenced how police are advised to question witnesses.
Research check: Ulric Neisser and Nicole Harsch asked students the morning after the 1986 Challenger shuttle disaster how they heard the news, then asked again about two and a half years later. Many accounts had changed substantially, yet people felt highly confident. Vividness is not the same as accuracy.
What we fail to see
Perception has its own blind spots. In 1999 Daniel Simons and Christopher Chabris filmed two teams passing basketballs and asked viewers to count passes by the team in white. Midway through, a person in a gorilla suit walked into the scene, thumped their chest and walked off. Around half of viewers never noticed. This is inattentional blindness: attention is a spotlight, and outside it the world goes surprisingly dark.
Its cousin, change blindness, is even more unsettling. In a 1998 study by Simons and Daniel Levin, an experimenter asked a passer-by for directions; two people carrying a door walked between them, and the experimenter was swapped for someone else. About half of the people giving directions didn't notice they were now talking to a different person.
Illusions that persist
The Müller-Lyer illusion, published by Franz Carl Müller-Lyer in 1889, shows two equal lines that look different lengths because of arrow-shaped fins at their ends. The remarkable thing is that knowing the lines are equal doesn't make the illusion disappear. Your perceptual system keeps delivering its best guess, regardless of what you know. The impossible triangle on our home page works similarly: Swedish artist Oscar Reutersvärd drew such figures in 1934, and Lionel and Roger Penrose popularised the shape in 1958.
None of this means memory and perception are broken. They are built for speed and usefulness, filling gaps sensibly most of the time. Being humble about them, and writing things down, is simply good lab practice. Next: how routines take over the work.
