Memory: How It Actually Works (and Fails)

Psychology Fundamentals

Chapter 2 · Memory: How It Actually Works (and Fails)

Memory feels like a recording — press play, and the past replays faithfully. Real psychological research shows something closer to the opposite: memory decays predictably, and it can be actively reshaped, sometimes without the person remembering ever realizing it happened.

Ebbinghaus's Real Forgetting Curve

Hermann Ebbinghaus began real memory experiments in 1879, publishing his findings in 1885 — using himself as his own sole subject, memorizing invented "nonsense syllables" (like DAX or YAT) specifically to remove any advantage from prior real-world knowledge. One single investigation alone required a real 15,000 recitations.

A Real, Exponential Pattern

Ebbinghaus found forgetting follows a real exponential curve: the sharpest decline happens within the first 20 minutes, with substantial further decay continuing through the first hour — before the curve genuinely levels off after roughly one day.

Loftus & Palmer's Real 1974 Study — Words Change What You "Remember"

Elizabeth Loftus and John Palmer's real 1974 study showed participants videos of car crashes, then asked them to estimate the vehicles' speed — varying only the single verb used in the question.

Real, Specific Numbers

Asked how fast the cars were going when they "collided," participants estimated a mean of 32 mph. Asked using "hit each other," the estimate rose to 34 mph. Asked using "smashed each other," it jumped to 41 mph — a real, roughly 9 mph difference from a single word choice, on identical footage.

This is real, direct evidence that memory is reconstructive — later information (even something as small as a verb in a follow-up question) can genuinely reshape what someone reports remembering, not merely how they phrase it.

A Real, More Extreme Case: The "Lost in the Mall" Study

Loftus and Jacqueline Pickrell's real 1995 study went further, testing whether an entire false childhood memory could be planted. 24 real participants each received a booklet with four narratives about their own childhood — three genuinely true, one fabricated: being lost in a shopping mall around age 5 and rescued by an elderly stranger.

A Real, Honestly Nuanced Result
Roughly 25% of the 24 participants came to report remembering the fabricated event as real. When later told one of the four stories was false, 5 of the 24 participants misidentified a genuinely true story as the fake one instead. But the result wasn't uniformly convincing: the false memories were typically described with fewer words and less vivid detail than the real ones — a real, honest limitation the original researchers reported rather than hid.

Loftus herself described this as an "existence proof" — not that false memories are common or easy to create, but that they can genuinely be created through suggestion at all, with real, direct implications for how eyewitness testimony and recovered-memory claims should be treated in practice.

Hands-On Exercises

Exercise 1

Explain, in your own words, why Ebbinghaus specifically used nonsense syllables rather than real words for his forgetting-curve research.

📄 View solution
Exercise 2

Explain, in your own words, why the real 9 mph gap between "collided" and "smashed" specifically demonstrates memory being reconstructive, rather than simply showing people are bad at estimating speed.

📄 View solution
Exercise 3

Explain, in your own words, why the real detail that false memories were described less vividly than true ones matters for how confidently this study's own findings should be applied to real-world situations, like eyewitness testimony.

📄 View solution

Chapter 2 Quick Reference

  • Ebbinghaus's real 1879-1885 self-experimentation (nonsense syllables, 15,000 recitations in one study) found forgetting follows an exponential curve — sharpest in the first 20 minutes, leveling off after roughly a day
  • Loftus & Palmer's real 1974 study found a single verb ("collided" vs. "smashed") shifted mean speed estimates from 32 mph to 41 mph — real evidence memory is reconstructive
  • Loftus & Pickrell's real 1995 "lost in the mall" study found 25% of 24 participants came to falsely remember a fabricated childhood event — with the honest nuance that false memories were typically less vivid and detailed than real ones